Authors: Ricardo Loor-Torres, Yuqi Wu, Esteban Cabezas, Mariana Borras, David Toro-Tobon, Mayra Duran, Misk Al Zahidy, Maria Mateo Chavez, Cristian Soto Jacome, Jungwei W. Fan, Naykky M. Singh Ospina, Yonghui Wu, Juan P. Brito
Abstract: Background We aim to use Natural Language Processing (NLP) to automate the extraction and classification of thyroid cancer risk factors from pathology reports. Methods We analyzed 1,410 surgical pathology reports from adult papillary thyroid cancer patients at Mayo Clinic, Rochester, MN, from 2010 to 2019. Structured and non-structured reports were used to create a consensus-based ground truth dictionary and categorized them into modified recurrence risk levels. Non-structured reports were narrative, while structured reports followed standardized formats. We then developed ThyroPath, a rule-based NLP pipeline, to extract and classify thyroid cancer features into risk categories. Training involved 225 reports (150 structured, 75 unstructured), with testing on 170 reports (120 structured, 50 unstructured) for evaluation. The pipeline's performance was assessed using both strict and lenient criteria for accuracy, precision, recall, and F1-score. Results In extraction tasks, ThyroPath achieved overall strict F-1 scores of 93% for structured reports and 90 for unstructured reports, covering 18 thyroid cancer pathology features. In classification tasks, ThyroPath-extracted information demonstrated an overall accuracy of 93% in categorizing reports based on their corresponding guideline-based risk of recurrence: 76.9% for high-risk, 86.8% for intermediate risk, and 100% for both low and very low-risk cases. However, ThyroPath achieved 100% accuracy across all thyroid cancer risk categories with human-extracted pathology information. Conclusions ThyroPath shows promise in automating the extraction and risk recurrence classification of thyroid pathology reports at large scale. It offers a solution to laborious manual reviews and advancing virtual registries. However, it requires further validation before implementation.
Authors: Lingxi Xiao, Muqing Li, Yinqiu Feng, Meiqi Wang, Ziyi Zhu, Zexi Chen
Abstract: The research explores the utilization of a deep learning model employing an attention mechanism in medical text mining. It targets the challenge of analyzing unstructured text information within medical data. This research seeks to enhance the model's capability to identify essential medical information by incorporating deep learning and attention mechanisms. This paper reviews the basic principles and typical model architecture of attention mechanisms and shows the effectiveness of their application in the tasks of disease prediction, drug side effect monitoring, and entity relationship extraction. Aiming at the particularity of medical texts, an adaptive attention model integrating domain knowledge is proposed, and its ability to understand medical terms and process complex contexts is optimized. The experiment verifies the model's effectiveness in improving task accuracy and robustness, especially when dealing with long text. The future research path of enhancing model interpretation, realizing cross-domain knowledge transfer, and adapting to low-resource scenarios is discussed in the research outlook, which provides a new perspective and method support for intelligent medical information processing and clinical decision assistance. Finally, cross-domain knowledge transfer and adaptation strategies for low-resource scenarios, providing theoretical basis and technical reference for promoting the development of intelligent medical information processing and clinical decision support systems.
Authors: Shezheng Song, Shasha Li, Shan Zhao, Chengyu Wang, Xiaopeng Li, Jie Yu, Qian Wan, Jun Ma, Tianwei Yan, Wentao Ma, Xiaoguang Mao
Abstract: Multimodal aspect-based sentiment analysis (MABSA) aims to understand opinions in a granular manner, advancing human-computer interaction and other fields. Traditionally, MABSA methods use a joint prediction approach to identify aspects and sentiments simultaneously. However, we argue that joint models are not always superior. Our analysis shows that joint models struggle to align relevant text tokens with image patches, leading to misalignment and ineffective image utilization. In contrast, a pipeline framework first identifies aspects through MATE (Multimodal Aspect Term Extraction) and then aligns these aspects with image patches for sentiment classification (MASC: Multimodal Aspect-Oriented Sentiment Classification). This method is better suited for multimodal scenarios where effective image use is crucial. We present three key observations: (a) MATE and MASC have different feature requirements, with MATE focusing on token-level features and MASC on sequence-level features; (b) the aspect identified by MATE is crucial for effective image utilization; and (c) images play a trivial role in previous MABSA methods due to high noise. Based on these observations, we propose a pipeline framework that first predicts the aspect and then uses translation-based alignment (TBA) to enhance multimodal semantic consistency for better image utilization. Our method achieves state-of-the-art (SOTA) performance on widely used MABSA datasets Twitter-15 and Twitter-17. This demonstrates the effectiveness of the pipeline approach and its potential to provide valuable insights for future MABSA research. For reproducibility, the code and checkpoint will be released.
Authors: Alessio Buscemi, Daniele Proverbio
Abstract: Democratic opinion-forming may be manipulated if newspapers' alignment to political or economical orientation is ambiguous. Various methods have been developed to better understand newspapers' positioning. Recently, the advent of Large Language Models (LLM), and particularly the pre-trained LLM chatbots like ChatGPT or Gemini, hold disruptive potential to assist researchers and citizens alike. However, little is know on whether LLM assessment is trustworthy: do single LLM agrees with experts' assessment, and do different LLMs answer consistently with one another? In this paper, we address specifically the second challenge. We compare how four widely employed LLMs rate the positioning of newspapers, and compare if their answers align with one another. We observe that this is not the case. Over a woldwide dataset, articles in newspapers are positioned strikingly differently by single LLMs, hinting to inconsistent training or excessive randomness in the algorithms. We thus raise a warning when deciding which tools to use, and we call for better training and algorithm development, to cover such significant gap in a highly sensitive matter for democracy and societies worldwide. We also call for community engagement in benchmark evaluation, through our open initiative navai.pro.
Authors: Jaehee Ryu, Seonhee Cho, Gyubok Lee, Edward Choi
Abstract: In this paper, we introduce EHR-SeqSQL, a novel sequential text-to-SQL dataset for Electronic Health Record (EHR) databases. EHR-SeqSQL is designed to address critical yet underexplored aspects in text-to-SQL parsing: interactivity, compositionality, and efficiency. To the best of our knowledge, EHR-SeqSQL is not only the largest but also the first medical text-to-SQL dataset benchmark to include sequential and contextual questions. We provide a data split and the new test set designed to assess compositional generalization ability. Our experiments demonstrate the superiority of a multi-turn approach over a single-turn approach in learning compositionality. Additionally, our dataset integrates specially crafted tokens into SQL queries to improve execution efficiency. With EHR-SeqSQL, we aim to bridge the gap between practical needs and academic research in the text-to-SQL domain.
Authors: Rebecca Dorn, Lee Kezar, Fred Morstatter, Kristina Lerman
Abstract: Content moderation on social media platforms shapes the dynamics of online discourse, influencing whose voices are amplified and whose are suppressed. Recent studies have raised concerns about the fairness of content moderation practices, particularly for aggressively flagging posts from transgender and non-binary individuals as toxic. In this study, we investigate the presence of bias in harmful speech classification of gender-queer dialect online, focusing specifically on the treatment of reclaimed slurs. We introduce a novel dataset, QueerReclaimLex, based on 109 curated templates exemplifying non-derogatory uses of LGBTQ+ slurs. Dataset instances are scored by gender-queer annotators for potential harm depending on additional context about speaker identity. We systematically evaluate the performance of five off-the-shelf language models in assessing the harm of these texts and explore the effectiveness of chain-of-thought prompting to teach large language models (LLMs) to leverage author identity context. We reveal a tendency for these models to inaccurately flag texts authored by gender-queer individuals as harmful. Strikingly, across all LLMs the performance is poorest for texts that show signs of being written by individuals targeted by the featured slur (F1 <= 0.24). We highlight an urgent need for fairness and inclusivity in content moderation systems. By uncovering these biases, this work aims to inform the development of more equitable content moderation practices and contribute to the creation of inclusive online spaces for all users.
Authors: Medha Hira, Arnav Goel, Anubha Gupta
Abstract: This paper presents CrossVoice, a novel cascade-based Speech-to-Speech Translation (S2ST) system employing advanced ASR, MT, and TTS technologies with cross-lingual prosody preservation through transfer learning. We conducted comprehensive experiments comparing CrossVoice with direct-S2ST systems, showing improved BLEU scores on tasks such as Fisher Es-En, VoxPopuli Fr-En and prosody preservation on benchmark datasets CVSS-T and IndicTTS. With an average mean opinion score of 3.75 out of 4, speech synthesized by CrossVoice closely rivals human speech on the benchmark, highlighting the efficacy of cascade-based systems and transfer learning in multilingual S2ST with prosody transfer.
Authors: Arnav Goel, Medha Hira, Anubha Gupta
Abstract: The field of prosody transfer in speech synthesis systems is rapidly advancing. This research is focused on evaluating learning methods for adapting pre-trained monolingual text-to-speech (TTS) models to multilingual conditions, i.e., Supervised Fine-Tuning (SFT) and Transfer Learning (TL). This comparison utilizes three distinct metrics: Mean Opinion Score (MOS), Recognition Accuracy (RA), and Mel Cepstral Distortion (MCD). Results demonstrate that, in comparison to SFT, TL leads to significantly enhanced performance, with an average MOS higher by 1.53 points, a 37.5% increase in RA, and approximately a 7.8-point improvement in MCD. These findings are instrumental in helping build TTS models for low-resource languages.
Authors: Jing Li, Zhijie Sun, Dachao Lin, Xuan He, Yi Lin, Binfan Zheng, Li Zeng, Rongqian Zhao, Xin Chen
Abstract: Mixture-of-Experts (MoE) architectures have recently gained increasing popularity within the domain of large language models (LLMs) due to their ability to significantly reduce training and inference overhead. However, MoE architectures face challenges, such as significant disparities in the number of tokens assigned to each expert and a tendency toward homogenization among experts, which adversely affects the model's semantic generation capabilities. In this paper, we introduce LocMoE+, a refined version of the low-overhead LocMoE, incorporating the following enhancements: (1) Quantification and definition of the affinity between experts and tokens. (2) Implementation of a global-level adaptive routing strategy to rearrange tokens based on their affinity scores. (3) Reestimation of the lower bound for expert capacity, which has been shown to progressively decrease as the token feature distribution evolves. Experimental results demonstrate that, without compromising model convergence or efficacy, the number of tokens each expert processes can be reduced by over 60%. Combined with communication optimizations, this leads to an average improvement in training efficiency ranging from 5.4% to 46.6%. After fine-tuning, LocMoE+ exhibits a performance improvement of 9.7% to 14.1% across the GDAD, C-Eval, and TeleQnA datasets.
Authors: Guy Tennenholtz, Yinlam Chow, Chih-Wei Hsu, Lior Shani, Ethan Liang, Craig Boutilier
Abstract: We propose a novel approach for training large language models (LLMs) to adhere to objectives defined within a latent embedding space. Our method leverages reinforcement learning (RL), treating a pre-trained LLM as an environment. Our embedding-aligned guided language (EAGLE) agent is trained to iteratively steer the LLM's generation towards optimal regions of the latent embedding space, w.r.t. some predefined criterion. We demonstrate the effectiveness of the EAGLE agent using the MovieLens 25M dataset to surface content gaps that satisfy latent user demand. We also demonstrate the benefit of using an optimal design of a state-dependent action set to improve EAGLE's efficiency. Our work paves the way for controlled and grounded text generation using LLMs, ensuring consistency with domain-specific knowledge and data representations.
Authors: Jiaxing Li, Chi Xu, Feng Wang, Isaac M von Riedemann, Cong Zhang, Jiangchuan Liu
Abstract: Large Language Models (LLMs) have become increasingly popular, transforming a wide range of applications across various domains. However, the real-world effectiveness of their query cache systems has not been thoroughly investigated. In this work, we for the first time conducted an analysis on real-world human-to-LLM interaction data, identifying key challenges in existing caching solutions for LLM-based chat services. Our findings reveal that current caching methods fail to leverage semantic connections, leading to inefficient cache performance and extra token costs. To address these issues, we propose SCALM, a new cache architecture that emphasizes semantic analysis and identifies significant cache entries and patterns. We also detail the implementations of the corresponding cache storage and eviction strategies. Our evaluations show that SCALM increases cache hit ratios and reduces operational costs for LLMChat services. Compared with other state-of-the-art solutions in GPTCache, SCALM shows, on average, a relative increase of 63% in cache hit ratio and a relative improvement of 77% in tokens savings.
Authors: H\`ector Loopez Hidalgo, Michel Boeglin, David Kahn, Josiane Mothe, Diego Ortiz, David Panzoli
Abstract: Semantic relations among entities are a widely accepted method for relation extraction. PromptORE (Prompt-based Open Relation Extraction) was designed to improve relation extraction with Large Language Models on generalistic documents. However, it is less effective when applied to historical documents, in languages other than English. In this study, we introduce an adaptation of PromptORE to extract relations from specialized documents, namely digital transcripts of trials from the Spanish Inquisition. Our approach involves fine-tuning transformer models with their pretraining objective on the data they will perform inference. We refer to this process as "biasing". Our Biased PromptORE addresses complex entity placements and genderism that occur in Spanish texts. We solve these issues by prompt engineering. We evaluate our method using Encoder-like models, corroborating our findings with experts' assessments. Additionally, we evaluate the performance using a binomial classification benchmark. Our results show a substantial improvement in accuracy -up to a 50% improvement with our Biased PromptORE models in comparison to the baseline models using standard PromptORE.
Authors: Seyed Moein Ayyoubzadeh
Abstract: Homograph disambiguation, the task of distinguishing words with identical spellings but different meanings, poses a substantial challenge in natural language processing. In this study, we introduce a novel dataset tailored for Persian homograph disambiguation. Our work encompasses a thorough exploration of various embeddings, evaluated through the cosine similarity method and their efficacy in downstream tasks like classification. Our investigation entails training a diverse array of lightweight machine learning and deep learning models for phonograph disambiguation. We scrutinize the models' performance in terms of Accuracy, Recall, and F1 Score, thereby gaining insights into their respective strengths and limitations. The outcomes of our research underscore three key contributions. First, we present a newly curated Persian dataset, providing a solid foundation for future research in homograph disambiguation. Second, our comparative analysis of embeddings highlights their utility in different contexts, enriching the understanding of their capabilities. Third, by training and evaluating a spectrum of models, we extend valuable guidance for practitioners in selecting suitable strategies for homograph disambiguation tasks. In summary, our study unveils a new dataset, scrutinizes embeddings through diverse perspectives, and benchmarks various models for homograph disambiguation. These findings empower researchers and practitioners to navigate the intricate landscape of homograph-related challenges effectively.
Authors: Simon Akesson, Frances A. Santos
Abstract: Providing external knowledge to Large Language Models (LLMs) is a key point for using these models in real-world applications for several reasons, such as incorporating up-to-date content in a real-time manner, providing access to domain-specific knowledge, and contributing to hallucination prevention. The vector database-based Retrieval Augmented Generation (RAG) approach has been widely adopted to this end. Thus, any part of external knowledge can be retrieved and provided to some LLM as the input context. Despite RAG approach's success, it still might be unfeasible for some applications, because the context retrieved can demand a longer context window than the size supported by LLM. Even when the context retrieved fits into the context window size, the number of tokens might be expressive and, consequently, impact costs and processing time, becoming impractical for most applications. To address these, we propose CRAG, a novel approach able to effectively reduce the number of prompting tokens without degrading the quality of the response generated compared to a solution using RAG. Through our experiments, we show that CRAG can reduce the number of tokens by at least 46\%, achieving more than 90\% in some cases, compared to RAG. Moreover, the number of tokens with CRAG does not increase considerably when the number of reviews analyzed is higher, unlike RAG, where the number of tokens is almost 9x higher when there are 75 reviews compared to 4 reviews.
Authors: Hanjuan Huang (Dept. of Computer Science and Information Engineering National Taiwan University of Science and Technology, Taipei, Taiwan), Hao-Jia Song (Dept. of Computer Science and Information Engineering National Taiwan University of Science and Technology, Taipei, Taiwan), Hsing-Kuo Pao (Dept. of Computer Science and Information Engineering National Taiwan University of Science and Technology, Taipei, Taiwan)
Abstract: We surely enjoy the larger the better models for their superior performance in the last couple of years when both the hardware and software support the birth of such extremely huge models. The applied fields include text mining and others. In particular, the success of LLMs on text understanding and text generation draws attention from researchers who have worked on NLP and related areas for years or even decades. On the side, LLMs may suffer from problems like model overfitting, hallucination, and device limitation to name a few. In this work, we suggest a model pruning technique specifically focused on LLMs. The proposed methodology emphasizes the explainability of deep learning models. By having the theoretical foundation, we obtain a trustworthy deep model so that huge models with a massive number of model parameters become not quite necessary. A mutual information-based estimation is adopted to find neurons with redundancy to eliminate. Moreover, an estimator with well-tuned parameters helps to find precise estimation to guide the pruning procedure. At the same time, we also explore the difference between pruning on large-scale models vs. pruning on small-scale models. The choice of pruning criteria is sensitive in small models but not for large-scale models. It is a novel finding through this work. Overall, we demonstrate the superiority of the proposed model to the state-of-the-art models.
Authors: Achuth Chandrasekhar, Jonathan Chan, Francis Ogoke, Olabode Ajenifujah, Amir Barati Farimani
Abstract: Generalized large language models (LLMs) such as GPT-4 may not provide specific answers to queries formulated by materials science researchers. These models may produce a high-level outline but lack the capacity to return detailed instructions on manufacturing and material properties of novel alloys. Enhancing a smaller model with specialized domain knowledge may provide an advantage over large language models which cannot be retrained quickly enough to keep up with the rapid pace of research in metal additive manufacturing (AM). We introduce "AMGPT," a specialized LLM text generator designed for metal AM queries. The goal of AMGPT is to assist researchers and users in navigating the extensive corpus of literature in AM. Instead of training from scratch, we employ a pre-trained Llama2-7B model from Hugging Face in a Retrieval-Augmented Generation (RAG) setup, utilizing it to dynamically incorporate information from $\sim$50 AM papers and textbooks in PDF format. Mathpix is used to convert these PDF documents into TeX format, facilitating their integration into the RAG pipeline managed by LlamaIndex. Expert evaluations of this project highlight that specific embeddings from the RAG setup accelerate response times and maintain coherence in the generated text.
Authors: Ying Zhang, Xiaofeng Li, Zhaoyang Liu, Haipeng Zhang
Abstract: Notable people's life trajectories have been a focus of study -- the locations and times of various activities, such as birth, death, education, marriage, competition, work, delivering a speech, making a scientific discovery, finishing a masterpiece, and fighting a battle, and how these people interact with others, carry important messages for the broad research related to human dynamics. However, the scarcity of trajectory data in terms of volume, density, and inter-person interactions, limits relevant studies from being comprehensive and interactive. We mine millions of biography pages from Wikipedia and tackle the generalization problem stemming from the variety and heterogeneity of the trajectory descriptions. Our ensemble model COSMOS, which combines the idea of semi-supervised learning and contrastive learning, achieves an F1 score of 85.95%. For this task, we also create a hand-curated dataset, WikiLifeTrajectory, consisting of 8,852 (person, time, location) triplets as ground truth. Besides, we perform an empirical analysis on the trajectories of 8,272 historians to demonstrate the validity of the extracted results. To facilitate the research on trajectory extractions and help the analytical studies to construct grand narratives, we make our code, the million-level extracted trajectories, and the WikiLifeTrajectory dataset publicly available.
Authors: Sara Kemper, Justin Cui, Kai Dicarlantonio, Kathy Lin, Danjie Tang, Anton Korikov, Scott Sanner
Abstract: Conversational recommendation (ConvRec) systems must understand rich and diverse natural language (NL) expressions of user preferences and intents, often communicated in an indirect manner (e.g., "I'm watching my weight"). Such complex utterances make retrieving relevant items challenging, especially if only using often incomplete or out-of-date metadata. Fortunately, many domains feature rich item reviews that cover standard metadata categories and offer complex opinions that might match a user's interests (e.g., "classy joint for a date"). However, only recently have large language models (LLMs) let us unlock the commonsense connections between user preference utterances and complex language in user-generated reviews. Further, LLMs enable novel paradigms for semi-structured dialogue state tracking, complex intent and preference understanding, and generating recommendations, explanations, and question answers. We thus introduce a novel technology RA-Rec, a Retrieval-Augmented, LLM-driven dialogue state tracking system for ConvRec, showcased with a video, open source GitHub repository, and interactive Google Colab notebook.
Authors: Tianlong Wang, Xianfeng Jiao, Yifan He, Zhongzhi Chen, Yinghao Zhu, Xu Chu, Junyi Gao, Yasha Wang, Liantao Ma
Abstract: Recent studies have indicated that Large Language Models (LLMs) harbor an inherent understanding of truthfulness, yet often fail to express fully and generate false statements. This gap between "knowing" and "telling" poses a challenge for ensuring the truthfulness of generated content. To address this, we introduce Adaptive Activation Steering (ACT), a tuning-free method that adaptively shift LLM's activations in "truthful" direction during inference. ACT addresses diverse categories of hallucinations by utilizing diverse steering vectors and adjusting the steering intensity adaptively. Applied as an add-on across various models, ACT significantly improves truthfulness in LLaMA ($\uparrow$ 142\%), LLaMA2 ($\uparrow$ 24\%), Alpaca ($\uparrow$ 36\%), Vicuna ($\uparrow$ 28\%), and LLaMA2-Chat ($\uparrow$ 19\%). Furthermore, we verify ACT's scalability across larger models (13B, 33B, 65B), underscoring the adaptability of ACT to large-scale language models.
Authors: Yinghao Zhu, Changyu Ren, Zixiang Wang, Xiaochen Zheng, Shiyun Xie, Junlan Feng, Xi Zhu, Zhoujun Li, Liantao Ma, Chengwei Pan
Abstract: The integration of multimodal Electronic Health Records (EHR) data has notably advanced clinical predictive capabilities. However, current models that utilize clinical notes and multivariate time-series EHR data often lack the necessary medical context for precise clinical tasks. Previous methods using knowledge graphs (KGs) primarily focus on structured knowledge extraction. To address this, we propose EMERGE, a Retrieval-Augmented Generation (RAG) driven framework aimed at enhancing multimodal EHR predictive modeling. Our approach extracts entities from both time-series data and clinical notes by prompting Large Language Models (LLMs) and aligns them with professional PrimeKG to ensure consistency. Beyond triplet relationships, we include entities' definitions and descriptions to provide richer semantics. The extracted knowledge is then used to generate task-relevant summaries of patients' health statuses. These summaries are fused with other modalities utilizing an adaptive multimodal fusion network with cross-attention. Extensive experiments on the MIMIC-III and MIMIC-IV datasets for in-hospital mortality and 30-day readmission tasks demonstrate the superior performance of the EMERGE framework compared to baseline models. Comprehensive ablation studies and analyses underscore the efficacy of each designed module and the framework's robustness to data sparsity. EMERGE significantly enhances the use of multimodal EHR data in healthcare, bridging the gap with nuanced medical contexts crucial for informed clinical predictions.
Authors: Hongyu Yang, Liyang He, Min Hou, Shuanghong Shen, Rui Li, Jiahui Hou, Jianhui Ma, Junda Zhao
Abstract: Code Community Question Answering (CCQA) seeks to tackle programming-related issues, thereby boosting productivity in both software engineering and academic research. Recent advancements in Reinforcement Learning from Human Feedback (RLHF) have transformed the fine-tuning process of Large Language Models (LLMs) to produce responses that closely mimic human behavior. Leveraging LLMs with RLHF for practical CCQA applications has thus emerged as a promising area of study. Unlike standard code question-answering tasks, CCQA involves multiple possible answers, with varying user preferences for each response. Additionally, code communities often show a preference for new APIs. These challenges prevent LLMs from generating responses that cater to the diverse preferences of users in CCQA tasks. To address these issues, we propose a novel framework called Aligning LLMs through Multi-perspective User Preference Ranking-based Feedback for Programming Question Answering (ALMupQA) to create user-focused responses. Our approach starts with Multi-perspective Preference Ranking Alignment (MPRA), which synthesizes varied user preferences based on the characteristics of answers from code communities. We then introduce a Retrieval-augmented In-context Learning (RIL) module to mitigate the problem of outdated answers by retrieving responses to similar questions from a question bank. Due to the limited availability of high-quality, multi-answer CCQA datasets, we also developed a dataset named StaCCQA from real code communities. Extensive experiments demonstrated the effectiveness of the ALMupQA framework in terms of accuracy and user preference. Compared to the base model, ALMupQA showed nearly an 11% improvement in BLEU, with increases of 20% and 17.5% in BERTScore and CodeBERTScore, respectively.
Authors: Sanath Narayan, Yasser Abdelaziz Dahou Djilali, Ankit Singh, Eustache Le Bihan, Hakim Hacid
Abstract: This work presents an extensive and detailed study on Audio-Visual Speech Recognition (AVSR) for five widely spoken languages: Chinese, Spanish, English, Arabic, and French. We have collected large-scale datasets for each language except for English, and have engaged in the training of supervised learning models. Our model, ViSpeR, is trained in a multi-lingual setting, resulting in competitive performance on newly established benchmarks for each language. The datasets and models are released to the community with an aim to serve as a foundation for triggering and feeding further research work and exploration on Audio-Visual Speech Recognition, an increasingly important area of research. Code available at \href{https://github.com/YasserdahouML/visper}{https://github.com/YasserdahouML/visper}.
URLs: https://github.com/YasserdahouML/visper, https://github.com/YasserdahouML/visper
Authors: Subhankar Maity, Aniket Deroy, Sudeshna Sarkar
Abstract: Grammatical error correction (GEC) tools, powered by advanced generative artificial intelligence (AI), competently correct linguistic inaccuracies in user input. However, they often fall short in providing essential natural language explanations, which are crucial for learning languages and gaining a deeper understanding of the grammatical rules. There is limited exploration of these tools in low-resource languages such as Bengali. In such languages, grammatical error explanation (GEE) systems should not only correct sentences but also provide explanations for errors. This comprehensive approach can help language learners in their quest for proficiency. Our work introduces a real-world, multi-domain dataset sourced from Bengali speakers of varying proficiency levels and linguistic complexities. This dataset serves as an evaluation benchmark for GEE systems, allowing them to use context information to generate meaningful explanations and high-quality corrections. Various generative pre-trained large language models (LLMs), including GPT-4 Turbo, GPT-3.5 Turbo, Text-davinci-003, Text-babbage-001, Text-curie-001, Text-ada-001, Llama-2-7b, Llama-2-13b, and Llama-2-70b, are assessed against human experts for performance comparison. Our research underscores the limitations in the automatic deployment of current state-of-the-art generative pre-trained LLMs for Bengali GEE. Advocating for human intervention, our findings propose incorporating manual checks to address grammatical errors and improve feedback quality. This approach presents a more suitable strategy to refine the GEC tools in Bengali, emphasizing the educational aspect of language learning.
Authors: Krish Didwania, Dr. Durga Toshniwal, Amit Agarwal
Abstract: Legal documents are indispensable in every country for legal practices and serve as the primary source of information regarding previous cases and employed statutes. In today's world, with an increasing number of judicial cases, it is crucial to systematically categorize past cases into subgroups, which can then be utilized for upcoming cases and practices. Our primary focus in this endeavor was to annotate cases using topic modeling algorithms such as Latent Dirichlet Allocation, Non-Negative Matrix Factorization, and Bertopic for a collection of lengthy legal documents from India and the UK. This step is crucial for distinguishing the generated labels between the two countries, highlighting the differences in the types of cases that arise in each jurisdiction. Furthermore, an analysis of the timeline of cases from India was conducted to discern the evolution of dominant topics over the years.
Authors: Rui Guo, Greg Farnan, Niall McLaughlin, Barry Devereux
Abstract: The BioNLP ACL'24 Shared Task on Streamlining Discharge Documentation aims to reduce the administrative burden on clinicians by automating the creation of critical sections of patient discharge letters. This paper presents our approach using the Llama3 8B quantized model to generate the "Brief Hospital Course" and "Discharge Instructions" sections. We employ a zero-shot method combined with Retrieval-Augmented Generation (RAG) to produce concise, contextually accurate summaries. Our contributions include the development of a curated template-based approach to ensure reliability and consistency, as well as the integration of RAG for word count prediction. We also describe several unsuccessful experiments to provide insights into our pathway for the competition. Our results demonstrate the effectiveness and efficiency of our approach, achieving high scores across multiple evaluation metrics.
Authors: Xu Wang, Yuan Wu
Abstract: Adversarial training has been instrumental in advancing multi-domain text classification (MDTC). Traditionally, MDTC methods employ a shared-private paradigm, with a shared feature extractor for domain-invariant knowledge and individual private feature extractors for domain-specific knowledge. Despite achieving state-of-the-art results, these methods grapple with the escalating model parameters due to the continuous addition of new domains. To address this challenge, we introduce the Stochastic Adversarial Network (SAN), which innovatively models the parameters of the domain-specific feature extractor as a multivariate Gaussian distribution, as opposed to a traditional weight vector. This design allows for the generation of numerous domain-specific feature extractors without a substantial increase in model parameters, maintaining the model's size on par with that of a single domain-specific extractor. Furthermore, our approach integrates domain label smoothing and robust pseudo-label regularization to fortify the stability of adversarial training and to refine feature discriminability, respectively. The performance of our SAN, evaluated on two leading MDTC benchmarks, demonstrates its competitive edge against the current state-of-the-art methodologies. The code is available at https://github.com/wangxu0820/SAN.
Authors: Yuanpu Cao, Tianrong Zhang, Bochuan Cao, Ziyi Yin, Lu Lin, Fenglong Ma, Jinghui Chen
Abstract: Researchers have been studying approaches to steer the behavior of Large Language Models (LLMs) and build personalized LLMs tailored for various applications. While fine-tuning seems to be a direct solution, it requires substantial computational resources and may significantly affect the utility of the original LLM. Recent endeavors have introduced more lightweight strategies, focusing on extracting "steering vectors" to guide the model's output toward desired behaviors by adjusting activations within specific layers of the LLM's transformer architecture. However, such steering vectors are directly extracted from the activations of human preference data and thus often lead to suboptimal results and occasional failures, especially in alignment-related scenarios. This work proposes an innovative approach that could produce more effective steering vectors through bi-directional preference optimization. Our method is designed to allow steering vectors to directly influence the generation probability of contrastive human preference data pairs, thereby offering a more precise representation of the target behavior. By carefully adjusting the direction and magnitude of the steering vector, we enabled personalized control over the desired behavior across a spectrum of intensities. Extensive experimentation across various open-ended generation tasks, particularly focusing on steering AI personas, has validated the efficacy of our approach. Moreover, we comprehensively investigate critical alignment-concerning scenarios, such as managing truthfulness, mitigating hallucination, and addressing jailbreaking attacks. Remarkably, our method can still demonstrate outstanding steering effectiveness across these scenarios. Furthermore, we showcase the transferability of our steering vectors across different models/LoRAs and highlight the synergistic benefits of applying multiple vectors simultaneously.
Authors: Tong Chen, Danny Wang, Xurong Liang, Marten Risius, Gianluca Demartini, Hongzhi Yin
Abstract: To counter the side effect brought by the proliferation of social media platforms, hate speech detection (HSD) plays a vital role in halting the dissemination of toxic online posts at an early stage. However, given the ubiquitous topical communities on social media, a trained HSD classifier easily becomes biased towards specific targeted groups (e.g., female and black people), where a high rate of false positive/negative results can significantly impair public trust in the fairness of content moderation mechanisms, and eventually harm the diversity of online society. Although existing fairness-aware HSD methods can smooth out some discrepancies across targeted groups, they are mostly specific to a narrow selection of targets that are assumed to be known and fixed. This inevitably prevents those methods from generalizing to real-world use cases where new targeted groups constantly emerge over time. To tackle this defect, we propose Generalizable target-aware Fairness (GetFair), a new method for fairly classifying each post that contains diverse and even unseen targets during inference. To remove the HSD classifier's spurious dependence on target-related features, GetFair trains a series of filter functions in an adversarial pipeline, so as to deceive the discriminator that recovers the targeted group from filtered post embeddings. To maintain scalability and generalizability, we innovatively parameterize all filter functions via a hypernetwork that is regularized by the semantic affinity among targets. Taking a target's pretrained word embedding as input, the hypernetwork generates the weights used by each target-specific filter on-the-fly without storing dedicated filter parameters. Finally, comparative experiments on two HSD datasets have shown advantageous performance of GetFair on out-of-sample targets.
Authors: Francesco Cagnetta, Matthieu Wyart
Abstract: How much data is required to learn the structure of a language via next-token prediction? We study this question for synthetic datasets generated via a Probabilistic Context-Free Grammar (PCFG) -- a hierarchical generative model that captures the tree-like structure of natural languages. We determine token-token correlations analytically in our model and show that they can be used to build a representation of the grammar's hidden variables, the longer the range the deeper the variable. In addition, a finite training set limits the resolution of correlations to an effective range, whose size grows with that of the training set. As a result, a Language Model trained with increasingly many examples can build a deeper representation of the grammar's structure, thus reaching good performance despite the high dimensionality of the problem. We conjecture that the relationship between training set size and effective range of correlations holds beyond our synthetic datasets. In particular, our conjecture predicts how the scaling law for the test loss behaviour with training set size depends on the length of the context window, which we confirm empirically for a collection of lines from Shakespeare's plays.
Authors: Gon\c{c}alo R. A. Faria, Sweta Agrawal, Ant\'onio Farinhas, Ricardo Rei, Jos\'e G. C. de Souza, Andr\'e F. T. Martins
Abstract: An important challenge in machine translation (MT) is to generate high-quality and diverse translations. Prior work has shown that the estimated likelihood from the MT model correlates poorly with translation quality. In contrast, quality evaluation metrics (such as COMET or BLEURT) exhibit high correlations with human judgments, which has motivated their use as rerankers (such as quality-aware and minimum Bayes risk decoding). However, relying on a single translation with high estimated quality increases the chances of "gaming the metric''. In this paper, we address the problem of sampling a set of high-quality and diverse translations. We provide a simple and effective way to avoid over-reliance on noisy quality estimates by using them as the energy function of a Gibbs distribution. Instead of looking for a mode in the distribution, we generate multiple samples from high-density areas through the Metropolis-Hastings algorithm, a simple Markov chain Monte Carlo approach. The results show that our proposed method leads to high-quality and diverse outputs across multiple language pairs (English$\leftrightarrow${German, Russian}) with two strong decoder-only LLMs (Alma-7b, Tower-7b).
Authors: Xinyi Liu, Pinxin Liu, Hangfeng He
Abstract: In this study, we investigate the capabilities and inherent biases of advanced large language models (LLMs) such as GPT-3.5 and GPT-4 in the context of debate evaluation. We discover that LLM's performance exceeds humans and surpasses the performance of state-of-the-art methods fine-tuned on extensive datasets in debate evaluation. We additionally explore and analyze biases present in LLMs, including positional bias, lexical bias, order bias, which may affect their evaluative judgments. Our findings reveal a consistent bias in both GPT-3.5 and GPT-4 towards the second candidate response presented, attributed to prompt design. We also uncover lexical biases in both GPT-3.5 and GPT-4, especially when label sets carry connotations such as numerical or sequential, highlighting the critical need for careful label verbalizer selection in prompt design. Additionally, our analysis indicates a tendency of both models to favor the debate's concluding side as the winner, suggesting an end-of-discussion bias.
Authors: Suraj Anand, Michael A. Lepori, Jack Merullo, Ellie Pavlick
Abstract: Language models have the ability to perform in-context learning (ICL), allowing them to flexibly adapt their behavior based on context. This contrasts with in-weights learning, where information is statically encoded in model parameters from iterated observations of the data. Despite this apparent ability to learn in-context, language models are known to struggle when faced with unseen or rarely seen tokens. Hence, we study $\textbf{structural in-context learning}$, which we define as the ability of a model to execute in-context learning on arbitrary tokens -- so called because the model must generalize on the basis of e.g. sentence structure or task structure, rather than semantic content encoded in token embeddings. An ideal model would be able to do both: flexibly deploy in-weights operations (in order to robustly accommodate ambiguous or unknown contexts using encoded semantic information) and structural in-context operations (in order to accommodate novel tokens). We study structural in-context algorithms in a simple part-of-speech setting using both practical and toy models. We find that active forgetting, a technique that was recently introduced to help models generalize to new languages, forces models to adopt structural in-context learning solutions. Finally, we introduce $\textbf{temporary forgetting}$, a straightforward extension of active forgetting that enables one to control how much a model relies on in-weights vs. in-context solutions. Importantly, temporary forgetting allows us to induce a $\textit{dual process strategy}$ where in-context and in-weights solutions coexist within a single model.
Authors: Nick Alonso, Tom\'as Figliolia, Anthony Ndirango, Beren Millidge
Abstract: There has recently been growing interest in conversational agents with long-term memory which has led to the rapid development of language models that use retrieval-augmented generation (RAG). Until recently, most work on RAG has focused on information retrieval from large databases of texts, like Wikipedia, rather than information from long-form conversations. In this paper, we argue that effective retrieval from long-form conversational data faces two unique problems compared to static database retrieval: 1) time/event-based queries, which requires the model to retrieve information about previous conversations based on time or the order of a conversational event (e.g., the third conversation on Tuesday), and 2) ambiguous queries that require surrounding conversational context to understand. To better develop RAG-based agents that can deal with these challenges, we generate a new dataset of ambiguous and time-based questions that build upon a recent dataset of long-form, simulated conversations, and demonstrate that standard RAG based approaches handle such questions poorly. We then develop a novel retrieval model which combines chained-of-table search methods, standard vector-database retrieval, and a prompting method to disambiguate queries, and demonstrate that this approach substantially improves over current methods at solving these tasks. We believe that this new dataset and more advanced RAG agent can act as a key benchmark and stepping stone towards effective memory augmented conversational agents that can be used in a wide variety of AI applications.
Authors: Yechen Xu, Xinhao Kong, Tingjun Chen, Danyang Zhuo
Abstract: The complexity of large language model (LLM) serving workloads has substantially increased due to the integration with external tool invocations, such as ChatGPT plugins. In this paper, we identify a new opportunity for efficient LLM serving for requests that trigger tools: tool partial execution alongside LLM decoding. To this end, we design Conveyor, an efficient LLM serving system optimized for handling requests involving external tools. We introduce a novel interface for tool developers to expose partial execution opportunities to the LLM serving system and a request scheduler that facilitates partial tool execution. Our results demonstrate that tool partial execution can improve request completion latency by up to 38.8%.
Authors: Congchao Wang, Sean Augenstein, Keith Rush, Wittawat Jitkrittum, Harikrishna Narasimhan, Ankit Singh Rawat, Aditya Krishna Menon, Alec Go
Abstract: Reducing serving cost and latency is a fundamental concern for the deployment of language models (LMs) in business applications. To address this, cascades of LMs offer an effective solution that conditionally employ smaller models for simpler queries. Cascaded systems are typically built with independently trained models, neglecting the advantages of considering inference-time interactions of the cascaded LMs during training. In this paper, we present cascade-aware training(CAT), an approach to optimizing the overall quality-cost performance tradeoff of a cascade of LMs. We achieve inference-time benefits by training the small LM with awareness of its place in a cascade and downstream capabilities. We demonstrate the value of the proposed method with over 60 LM tasks of the SuperGLUE, WMT22, and FLAN2021 datasets.
Authors: David Pissarra, Isabel Curioso, Jo\~ao Alveira, Duarte Pereira, Bruno Ribeiro, Tom\'as Souper, Vasco Gomes, Andr\'e V. Carreiro, Vitor Rolla
Abstract: Automated clinical text anonymization has the potential to unlock the widespread sharing of textual health data for secondary usage while assuring patient privacy and safety. Despite the proposal of many complex and theoretically successful anonymization solutions in literature, these techniques remain flawed. As such, clinical institutions are still reluctant to apply them for open access to their data. Recent advances in developing Large Language Models (LLMs) pose a promising opportunity to further the field, given their capability to perform various tasks. This paper proposes six new evaluation metrics tailored to the challenges of generative anonymization with LLMs. Moreover, we present a comparative study of LLM-based methods, testing them against two baseline techniques. Our results establish LLM-based models as a reliable alternative to common approaches, paving the way toward trustworthy anonymization of clinical text.
Authors: Chengwei Wei, Kee Kiat Koo, Amir Tavanaei, Karim Bouyarmane
Abstract: Large Language Models (LLMs) have facilitated structured data generation, with applications in domains like tabular data, document databases, product catalogs, etc. However, concerns persist about generation veracity due to incorrect references or hallucinations, necessitating the incorporation of some form of model confidence for mitigation. Existing confidence estimation methods on LLM generations primarily focus on the confidence at the individual token level or the entire output sequence level, limiting their applicability to structured data generation, which consists of an intricate mix of both independent and correlated entries at the sub-structure level. In this paper, we first investigate confidence estimation methods for generated sub-structure-level data. We introduce the concept of Confidence Network that applies on the hidden state of the LLM transformer, as a more targeted estimate than the traditional token conditional probability. We further propose Confidence-Aware sub-structure Beam Search (CABS), a novel decoding method operating at the sub-structure level in structured data generation. CABS enhances the faithfulness of structured data generation by considering confidence scores from the Confidence Network for each sub-structure-level data and iteratively refining the prompts. Results show that CABS outperforms traditional token-level beam search for structured data generation by 16.7% Recall at 90% precision averagely on the problem of product attribute generation.
Authors: Giosue Baggio, Elliot Murphy
Abstract: In a recent paper, Mandelkern & Linzen (2024) - henceforth M&L - address the question of whether language models' (LMs) words refer. Their argument draws from the externalist tradition in philosophical semantics, which views reference as the capacity of words to "achieve 'word-to-world' connections". In the externalist framework, causally uninterrupted chains of usage, tracing every occurrence of a name back to its bearer, guarantee that, for example, 'Peano' refers to the individual Peano (Kripke 1980). This account is externalist both because words pick out referents 'out there' in the world, and because what determines reference are coordinated linguistic actions by members of a community, and not individual mental states. The "central question to ask", for M&L, is whether LMs too belong to human linguistic communities, such that words by LMs may also trace back causally to their bearers. Their answer is a cautious "yes": inputs to LMs are linguistic "forms with particular histories of referential use"; "those histories ground the referents of those forms"; any occurrence of 'Peano' in LM outputs is as causally connected to the individual Peano as any other occurrence of the same proper name in human speech or text; therefore, occurrences of 'Peano' in LM outputs refer to Peano. In this commentary, we first qualify M&L's claim as applying to a narrow class of natural language expressions. Thus qualified, their claim is valid, and we emphasise an additional motivation for that in Section 2. Next, we discuss the actual scope of their claim, and we suggest that the way they formulate it may lead to unwarranted generalisations about reference in LMs. Our critique is likewise applicable to other externalist accounts of LMs (e.g., Lederman & Mahowald 2024; Mollo & Milliere 2023). Lastly, we conclude with a comment on the status of LMs as members of human linguistic communities.
Authors: Bernd Bohnet, Kevin Swersky, Rosanne Liu, Pranjal Awasthi, Azade Nova, Javier Snaider, Hanie Sedghi, Aaron T Parisi, Michael Collins, Angeliki Lazaridou, Orhan Firat, Noah Fiedel
Abstract: We explore the use of long-context capabilities in large language models to create synthetic reading comprehension data from entire books. Previous efforts to construct such datasets relied on crowd-sourcing, but the emergence of transformers with a context size of 1 million or more tokens now enables entirely automatic approaches. Our objective is to test the capabilities of LLMs to analyze, understand, and reason over problems that require a detailed comprehension of long spans of text, such as questions involving character arcs, broader themes, or the consequences of early actions later in the story. We propose a holistic pipeline for automatic data generation including question generation, answering, and model scoring using an ``Evaluator''. We find that a relative approach, comparing answers between models in a pairwise fashion and ranking with a Bradley-Terry model, provides a more consistent and differentiating scoring mechanism than an absolute scorer that rates answers individually. We also show that LLMs from different model families produce moderate agreement in their ratings. We ground our approach using the manually curated NarrativeQA dataset, where our evaluator shows excellent agreement with human judgement and even finds errors in the dataset. Using our automatic evaluation approach, we show that using an entire book as context produces superior reading comprehension performance compared to baseline no-context (parametric knowledge only) and retrieval-based approaches.
Authors: Qian Ruan, Ilia Kuznetsov, Iryna Gurevych
Abstract: Collaborative review and revision of textual documents is the core of knowledge work and a promising target for empirical analysis and NLP assistance. Yet, a holistic framework that would allow modeling complex relationships between document revisions, reviews and author responses is lacking. To address this gap, we introduce Re3, a framework for joint analysis of collaborative document revision. We instantiate this framework in the scholarly domain, and present Re3-Sci, a large corpus of aligned scientific paper revisions manually labeled according to their action and intent, and supplemented with the respective peer reviews and human-written edit summaries. We use the new data to provide first empirical insights into collaborative document revision in the academic domain, and to assess the capabilities of state-of-the-art LLMs at automating edit analysis and facilitating text-based collaboration. We make our annotation environment and protocols, the resulting data and experimental code publicly available.
Authors: Maximillian Chen, Ruoxi Sun, Sercan \"O. Ar{\i}k, Tomas Pfister
Abstract: Large language models (LLMs) aligned through reinforcement learning from human feedback (RLHF) have quickly become one of the dominant paradigms for building intelligent conversational assistant agents. However, despite their strong performance across many benchmarks, LLM-based agents still lack conversational skills such as disambiguation: when generalized assistants are faced with ambiguity, they often overhedge or implicitly guess users' ground-truth intents rather than asking clarification questions, and under task-specific settings, high-quality conversation samples are often limited, affecting models' ability to learn optimal dialogue action policies. We propose Action-Based Contrastive Self-Training (henceforth ACT), a quasi-online preference optimization algorithm based on Direct Preference Optimization (DPO) which allows for sample-efficient dialogue policy learning in multi-turn conversation. We demonstrate ACT's efficacy under sample-efficient conditions in three difficult conversational tasks: tabular-grounded question-answering, machine reading comprehension, and AmbigSQL, a novel task for disambiguating information-seeking requests for text-to-SQL generation. Additionally, we propose evaluating LLMs' ability to function as conversational agents by examining whether they can implicitly recognize and reason about ambiguity in conversation. ACT demonstrates substantial conversation modeling improvements over standard approaches to supervised fine-tuning and DPO.
Authors: William Hogan, Jingbo Shang
Abstract: Recent research efforts have explored the potential of leveraging natural language inference (NLI) techniques to enhance relation extraction (RE). In this vein, we introduce MetaEntail-RE, a novel adaptation method that harnesses NLI principles to enhance RE performance. Our approach follows past works by verbalizing relation classes into class-indicative hypotheses, aligning a traditionally multi-class classification task to one of textual entailment. We introduce three key enhancements: (1) Instead of labeling non-entailed premise-hypothesis pairs with the uninformative "neutral" entailment label, we introduce meta-class analysis, which provides additional context by analyzing overarching meta relationships between classes when assigning entailment labels; (2) Feasible hypothesis filtering, which removes unlikely hypotheses from consideration based on pairs of entity types; and (3) Group-based prediction selection, which further improves performance by selecting highly confident predictions. MetaEntail-RE is conceptually simple and empirically powerful, yielding significant improvements over conventional relation extraction techniques and other NLI formulations. Our experimental results underscore the versatility of MetaEntail-RE, demonstrating performance gains across both biomedical and general domains.
Authors: Nate Rahn, Pierluca D'Oro, Marc G. Bellemare
Abstract: The generality of pretrained large language models (LLMs) has prompted increasing interest in their use as in-context learning agents. To be successful, such agents must form beliefs about how to achieve their goals based on limited interaction with their environment, resulting in uncertainty about the best action to take at each step. In this paper, we study how LLM agents form and act on these beliefs by conducting experiments in controlled sequential decision-making tasks. To begin, we find that LLM agents are overconfident: They draw strong conclusions about what to do based on insufficient evidence, resulting in inadequately explorative behavior. We dig deeper into this phenomenon and show how it emerges from a collapse in the entropy of the action distribution implied by sampling from the LLM. We then demonstrate that existing token-level sampling techniques are by themselves insufficient to make the agent explore more. Motivated by this fact, we introduce Entropic Activation Steering (EAST), an activation steering method for in-context LLM agents. EAST computes a steering vector as an entropy-weighted combination of representations, and uses it to manipulate an LLM agent's uncertainty over actions by intervening on its activations during the forward pass. We show that EAST can reliably increase the entropy in an LLM agent's actions, causing more explorative behavior to emerge. Finally, EAST modifies the subjective uncertainty an LLM agent expresses, paving the way to interpreting and controlling how LLM agents represent uncertainty about their decisions.
Authors: Mohammed Saidul Islam, Raian Rahman, Ahmed Masry, Md Tahmid Rahman Laskar, Mir Tafseer Nayeem, Enamul Hoque
Abstract: Natural language is a powerful complementary modality of communication for data visualizations, such as bar and line charts. To facilitate chart-based reasoning using natural language, various downstream tasks have been introduced recently such as chart question answering, chart summarization, and fact-checking with charts. These tasks pose a unique challenge, demanding both vision-language reasoning and a nuanced understanding of chart data tables, visual encodings, and natural language prompts. Despite the recent success of Large Language Models (LLMs) across diverse NLP tasks, their abilities and limitations in the realm of data visualization remain under-explored, possibly due to their lack of multi-modal capabilities. To bridge the gap, this paper presents the first comprehensive evaluation of the recently developed large vision language models (LVLMs) for chart understanding and reasoning tasks. Our evaluation includes a comprehensive assessment of LVLMs, including GPT-4V and Gemini, across four major chart reasoning tasks. Furthermore, we perform a qualitative evaluation of LVLMs' performance on a diverse range of charts, aiming to provide a thorough analysis of their strengths and weaknesses. Our findings reveal that LVLMs demonstrate impressive abilities in generating fluent texts covering high-level data insights while also encountering common problems like hallucinations, factual errors, and data bias. We highlight the key strengths and limitations of chart comprehension tasks, offering insights for future research.
Authors: Long Hei Matthew Lam, Ehsan Shareghi
Abstract: Logical reasoning serves as a cornerstone for human cognition. Recently, the emergence of Large Language Models (LLMs) has demonstrated promising progress in solving logical reasoning tasks effectively. To improve this capability, recent studies have delved into integrating LLMs with various symbolic solvers using diverse techniques and methodologies. While some combinations excel on specific datasets, others fall short. However, it remains unclear whether the variance in performance stems from the methodologies employed or the specific symbolic solvers utilized. Therefore, there is a lack of consistent comparison between symbolic solvers and how they influence LLM's logical reasoning ability. We perform experiments on LLMs integrated with 3 symbolic solvers: Z3, Pyke, and Prover9, and compare their performance on 3 logical reasoning datasets: ProofWriter, PrOntoQA, and FOLIO. Our findings indicate that when combined with LLMs Pyke's performance is significantly inferior to that of Prover9 and Z3. Z3's overall accuracy performance slightly surpasses Prover9, but Prover9 could execute more questions.
Authors: Sangwon Ryu, Heejin Do, Yunsu Kim, Gary Geunbae Lee, Jungseul Ok
Abstract: The evaluation of summary quality encompasses diverse dimensions such as consistency, coherence, relevance, and fluency. However, existing summarization methods often target a specific dimension, facing challenges in generating well-balanced summaries across multiple dimensions. In this paper, we propose multi-objective reinforcement learning tailored to generate balanced summaries across all four dimensions. We introduce two multi-dimensional optimization (MDO) strategies for adaptive learning: 1) MDO_min, rewarding the current lowest dimension score, and 2) MDO_pro, optimizing multiple dimensions similar to multi-task learning, resolves conflicting gradients across dimensions through gradient projection. Unlike prior ROUGE-based rewards relying on reference summaries, we use a QA-based reward model that aligns with human preferences. Further, we discover the capability to regulate the length of summaries by adjusting the discount factor, seeking the generation of concise yet informative summaries that encapsulate crucial points. Our approach achieved substantial performance gains compared to baseline models on representative summarization datasets, particularly in the overlooked dimensions.
Authors: Sarthak Harne, Monjoy Narayan Choudhury, Madhav Rao, TK Srikanth, Seema Mehrotra, Apoorva Vashisht, Aarushi Basu, Manjit Sodhi
Abstract: The limited availability of psychologists necessitates efficient identification of individuals requiring urgent mental healthcare. This study explores the use of Natural Language Processing (NLP) pipelines to analyze text data from online mental health forums used for consultations. By analyzing forum posts, these pipelines can flag users who may require immediate professional attention. A crucial challenge in this domain is data privacy and scarcity. To address this, we propose utilizing readily available curricular texts used in institutes specializing in mental health for pre-training the NLP pipelines. This helps us mimic the training process of a psychologist. Our work presents two models: a discriminative BERT-based model called CASE-BERT that flags potential mental health disorders based on forum text, and a generative model called CASE-Gemma that extracts key features for a preliminary diagnosis. CASE-BERT demonstrates superior performance compared to existing methods, achieving an f1 score of 0.91 for Depression and 0.88 for Anxiety, two of the most commonly reported mental health disorders. CASE-Gemma can achieve a BERT Score of 0.849 on generating diagnoses based on forum text. The effectiveness of CASE-Gemma is evaluated through both human evaluation and qualitative methods, with the collaboration of clinical psychologists who provide us with a set of annotated data for fine-tuning and evaluation. Our code is available at https://github.com/sarthakharne/CASE
Authors: Millicent Ochieng, Varun Gumma, Sunayana Sitaram, Jindong Wang, Keshet Ronen, Kalika Bali, Jacki O'Neill
Abstract: The deployment of Large Language Models (LLMs) in real-world applications presents both opportunities and challenges, particularly in multilingual and code-mixed communication settings. This research evaluates the performance of seven leading LLMs in sentiment analysis on a dataset derived from multilingual and code-mixed WhatsApp chats, including Swahili, English and Sheng. Our evaluation includes both quantitative analysis using metrics like F1 score and qualitative assessment of LLMs' explanations for their predictions. We find that, while Mistral-7b and Mixtral-8x7b achieved high F1 scores, they and other LLMs such as GPT-3.5-Turbo, Llama-2-70b, and Gemma-7b struggled with understanding linguistic and contextual nuances, as well as lack of transparency in their decision-making process as observed from their explanations. In contrast, GPT-4 and GPT-4-Turbo excelled in grasping diverse linguistic inputs and managing various contextual information, demonstrating high consistency with human alignment and transparency in their decision-making process. The LLMs however, encountered difficulties in incorporating cultural nuance especially in non-English settings with GPT-4s doing so inconsistently. The findings emphasize the necessity of continuous improvement of LLMs to effectively tackle the challenges of culturally nuanced, low-resource real-world settings.
Authors: Md. Mostafizer Rahman, Ariful Islam Shiplu, Yutaka Watanobe, Md. Ashad Alam
Abstract: Effectively analyzing the comments to uncover latent intentions holds immense value in making strategic decisions across various domains. However, several challenges hinder the process of sentiment analysis including the lexical diversity exhibited in comments, the presence of long dependencies within the text, encountering unknown symbols and words, and dealing with imbalanced datasets. Moreover, existing sentiment analysis tasks mostly leveraged sequential models to encode the long dependent texts and it requires longer execution time as it processes the text sequentially. In contrast, the Transformer requires less execution time due to its parallel processing nature. In this work, we introduce a novel hybrid deep learning model, RoBERTa-BiLSTM, which combines the Robustly Optimized BERT Pretraining Approach (RoBERTa) with Bidirectional Long Short-Term Memory (BiLSTM) networks. RoBERTa is utilized to generate meaningful word embedding vectors, while BiLSTM effectively captures the contextual semantics of long-dependent texts. The RoBERTa-BiLSTM hybrid model leverages the strengths of both sequential and Transformer models to enhance performance in sentiment analysis. We conducted experiments using datasets from IMDb, Twitter US Airline, and Sentiment140 to evaluate the proposed model against existing state-of-the-art methods. Our experimental findings demonstrate that the RoBERTa-BiLSTM model surpasses baseline models (e.g., BERT, RoBERTa-base, RoBERTa-GRU, and RoBERTa-LSTM), achieving accuracies of 80.74%, 92.36%, and 82.25% on the Twitter US Airline, IMDb, and Sentiment140 datasets, respectively. Additionally, the model achieves F1-scores of 80.73%, 92.35%, and 82.25% on the same datasets, respectively.
Authors: Chujie Gao, Qihui Zhang, Dongping Chen, Yue Huang, Siyuan Wu, Zhengyan Fu, Yao Wan, Xiangliang Zhang, Lichao Sun
Abstract: Large Language Models (LLMs) have achieved remarkable success across various industries due to their exceptional generative capabilities. However, for safe and effective real-world deployments, ensuring honesty and helpfulness is critical. This paper addresses the question: Can we prioritize the helpfulness of LLMs while preserving their honesty? To begin with, we establish exhaustive principles aimed at guaranteeing the honesty of LLM. Additionally, we introduce a novel dataset, referred to as HoneSet, comprising 930 queries spanning six categories meticulously crafted to assess an LLM's capacity for maintaining honesty. Subsequently, we present two approaches to augmenting honesty and helpfulness in LLMs: a training-free enhancement and a fine-tuning-based improvement. The training-free approach, which is based on curiosity-driven prompting, empowers LLMs to articulate internal confusion and uncertainty regarding queries, thereby optimizing their responses. Conversely, the fine-tuning-based method employs a two-stage process inspired by curriculum learning: initially instructing LLMs to discern between honest and dishonest responses, then refining their training to enhance helpfulness. Experiments conducted on nine prominent LLMs demonstrate a significant improvement in alignment with honesty across all models through the implementation of our proposed enhancements. Particularly noteworthy is the 65.3% enhancement observed in Llama3-8b and the remarkable 124.7% improvement in Mistral-7b, as measured by the H$^{2}$ (honest and helpful) assessment. We believe that our work can pave the way for developing more trustworthy LLMs for real-world applications.
Authors: Raysa Benatti, Fabiana Severi, Sandra Avila, Esther Luna Colombini
Abstract: Data derived from the realm of the social sciences is often produced in digital text form, which motivates its use as a source for natural language processing methods. Researchers and practitioners have developed and relied on artificial intelligence techniques to collect, process, and analyze documents in the legal field, especially for tasks such as text summarization and classification. While increasing procedural efficiency is often the primary motivation behind natural language processing in the field, several works have proposed solutions for human rights-related issues, such as assessment of public policy and institutional social settings. One such issue is the presence of gender biases in court decisions, which has been largely studied in social sciences fields; biased institutional responses to gender-based violence are a violation of international human rights dispositions since they prevent gender minorities from accessing rights and hamper their dignity. Natural language processing-based approaches can help detect these biases on a larger scale. Still, the development and use of such tools require researchers and practitioners to be mindful of legal and ethical aspects concerning data sharing and use, reproducibility, domain expertise, and value-charged choices. In this work, we (a) present an experimental framework developed to automatically detect gender biases in court decisions issued in Brazilian Portuguese and (b) describe and elaborate on features we identify to be critical in such a technology, given its proposed use as a support tool for research and assessment of court~activity.
Authors: Shichao Sun, Ruifeng Yuan, Ziqiang Cao, Wenjie Li, Pengfei Liu
Abstract: Large language models (LLMs) have demonstrated the capacity to improve summary quality by mirroring a human-like iterative process of critique and refinement starting from the initial draft. Two strategies are designed to perform this iterative process: Prompt Chaining and Stepwise Prompt. Prompt chaining orchestrates the drafting, critiquing, and refining phases through a series of three discrete prompts, while Stepwise prompt integrates these phases within a single prompt. However, the relative effectiveness of the two methods has not been extensively studied. This paper is dedicated to examining and comparing these two methods in the context of text summarization to ascertain which method stands out as the most effective. Experimental results show that the prompt chaining method can produce a more favorable outcome. This might be because stepwise prompt might produce a simulated refinement process according to our various experiments. Since refinement is adaptable to diverse tasks, our conclusions have the potential to be extrapolated to other applications, thereby offering insights that may contribute to the broader development of LLMs.
Authors: Juyong Jiang, Fan Wang, Jiasi Shen, Sungju Kim, Sunghun Kim
Abstract: Large Language Models (LLMs) have garnered remarkable advancements across diverse code-related tasks, known as Code LLMs, particularly in code generation that generates source code with LLM from natural language descriptions. This burgeoning field has captured significant interest from both academic researchers and industry professionals due to its practical significance in software development, e.g., GitHub Copilot. Despite the active exploration of LLMs for a variety of code tasks, either from the perspective of natural language processing (NLP) or software engineering (SE) or both, there is a noticeable absence of a comprehensive and up-to-date literature review dedicated to LLM for code generation. In this survey, we aim to bridge this gap by providing a systematic literature review that serves as a valuable reference for researchers investigating the cutting-edge progress in LLMs for code generation. We introduce a taxonomy to categorize and discuss the recent developments in LLMs for code generation, covering aspects such as data curation, latest advances, performance evaluation, and real-world applications. In addition, we present a historical overview of the evolution of LLMs for code generation and offer an empirical comparison using the widely recognized HumanEval and MBPP benchmarks to highlight the progressive enhancements in LLM capabilities for code generation. We identify critical challenges and promising opportunities regarding the gap between academia and practical development. Furthermore, we have established a dedicated resource website (https://codellm.github.io) to continuously document and disseminate the most recent advances in the field.
Authors: Phoebe J. Wang, Max Kreminski
Abstract: Instruction-tuned large language models (LLMs) are capable of generating stories in response to open-ended user requests, but the resulting stories tend to be limited in their diversity. Older, symbolic approaches to story generation (such as planning) can generate substantially more diverse plot outlines, but are limited to producing stories that recombine a fixed set of hand-engineered character action templates. Can we combine the strengths of these approaches while mitigating their weaknesses? We propose to do so by using a higher-level and more abstract symbolic specification of high-level story structure -- implemented via answer set programming (ASP) -- to guide and diversify LLM-based story generation. Via semantic similarity analysis, we demonstrate that our approach produces more diverse stories than an unguided LLM, and via code excerpts, we demonstrate the improved compactness and flexibility of ASP-based outline generation over full-fledged narrative planning.
Authors: Heidi C. Zhang, Sina J. Semnani, Farhad Ghassemi, Jialiang Xu, Shicheng Liu, Monica S. Lam
Abstract: We introduce SPAGHETTI: Semantic Parsing Augmented Generation for Hybrid English information from Text Tables and Infoboxes, a hybrid question-answering (QA) pipeline that utilizes information from heterogeneous knowledge sources, including knowledge base, text, tables, and infoboxes. Our LLM-augmented approach achieves state-of-the-art performance on the Compmix dataset, the most comprehensive heterogeneous open-domain QA dataset, with 56.5% exact match (EM) rate. More importantly, manual analysis on a sample of the dataset suggests that SPAGHETTI is more than 90% accurate, indicating that EM is no longer suitable for assessing the capabilities of QA systems today.
Authors: Liang Zhao, Tianwen Wei, Liang Zeng, Cheng Cheng, Liu Yang, Peng Cheng, Lijie Wang, Chenxia Li, Xuejie Wu, Bo Zhu, Yimeng Gan, Rui Hu, Shuicheng Yan, Han Fang, Yahui Zhou
Abstract: We introduce LongSkywork, a long-context Large Language Model (LLM) capable of processing up to 200,000 tokens. We provide a training recipe for efficiently extending context length of LLMs. We identify that the critical element in enhancing long-context processing capability is to incorporate a long-context SFT stage following the standard SFT stage. A mere 200 iterations can convert the standard SFT model into a long-context model. To reduce the effort in collecting and annotating data for long-context language modeling, we develop two novel methods for creating synthetic data. These methods are applied during the continual pretraining phase as well as the Supervised Fine-Tuning (SFT) phase, greatly enhancing the training efficiency of our long-context LLMs. Our findings suggest that synthetic long-context SFT data can surpass the performance of data curated by humans to some extent. LongSkywork achieves outstanding performance on a variety of long-context benchmarks. In the Needle test, a benchmark for long-context information retrieval, our models achieved perfect accuracy across multiple context spans. Moreover, in realistic application scenarios, LongSkywork-13B demonstrates performance on par with Claude2.1, the leading long-context model, underscoring the effectiveness of our proposed methods.
Authors: Ke Ji, Junying Chen, Anningzhe Gao, Wenya Xie, Xiang Wan, Benyou Wang
Abstract: In the quest for super-human performance, Large Language Models (LLMs) have traditionally been tethered to human-annotated datasets and predefined training objectives-a process that is both labor-intensive and inherently limited. This paper presents a transformative approach: Autonomous Learning for LLMs, a self-sufficient learning paradigm that frees models from the constraints of human supervision. This method endows LLMs with the ability to self-educate through direct interaction with text, akin to a human reading and comprehending literature. Our approach eliminates the reliance on annotated data, fostering an Autonomous Learning environment where the model independently identifies and reinforces its knowledge gaps. Empirical results from our comprehensive experiments, which utilized a diverse array of learning materials and were evaluated against standard public quizzes, reveal that Autonomous Learning outstrips the performance of both Pre-training and Supervised Fine-Tuning (SFT), as well as retrieval-augmented methods. These findings underscore the potential of Autonomous Learning to not only enhance the efficiency and effectiveness of LLM training but also to pave the way for the development of more advanced, self-reliant AI systems.
Authors: Xun Liu, Zhengwei Ni
Abstract: Large language models (LLM) have demonstrated remarkable abilities in generating natural language, understanding user instruction, and mimicking human language use. These capabilities have garnered considerable interest in applications such as role-playing. However, the process of collecting individual role scripts (or profiles) data and manually evaluating the performance can be costly. We introduce a framework that uses prompts to leverage the state-of-the-art (SOTA) LLMs to construct role-playing dialogue datasets and evaluate the role-playing performance. Additionally, we employ recall-oriented evaluation Rouge-L metric to support the result of the LLM evaluator.
Authors: Garrett Crumrine, Izzat Alsmadi, Jesus Guerrero, Yuvaraj Munian
Abstract: Large language models (LLMs) have revolutionized how we interact with machines. However, this technological advancement has been paralleled by the emergence of "Mallas," malicious services operating underground that exploit LLMs for nefarious purposes. Such services create malware, phishing attacks, and deceptive websites, escalating the cyber security threats landscape. This paper delves into the proliferation of Mallas by examining the use of various pre-trained language models and their efficiency and vulnerabilities when misused. Building on a dataset from the Common Vulnerabilities and Exposures (CVE) program, it explores fine-tuning methodologies to generate code and explanatory text related to identified vulnerabilities. This research aims to shed light on the operational strategies and exploitation techniques of Mallas, leading to the development of more secure and trustworthy AI applications. The paper concludes by emphasizing the need for further research, enhanced safeguards, and ethical guidelines to mitigate the risks associated with the malicious application of LLMs.
Authors: Chen Chen, Yuchen Hu, Wen Wu, Helin Wang, Eng Siong Chng, Chao Zhang
Abstract: In recent years, text-to-speech (TTS) technology has witnessed impressive advancements, particularly with large-scale training datasets, showcasing human-level speech quality and impressive zero-shot capabilities on unseen speakers. However, despite human subjective evaluations, such as the mean opinion score (MOS), remaining the gold standard for assessing the quality of synthetic speech, even state-of-the-art TTS approaches have kept human feedback isolated from training that resulted in mismatched training objectives and evaluation metrics. In this work, we investigate a novel topic of integrating subjective human evaluation into the TTS training loop. Inspired by the recent success of reinforcement learning from human feedback, we propose a comprehensive sampling-annotating-learning framework tailored to TTS optimization, namely uncertainty-aware optimization (UNO). Specifically, UNO eliminates the need for a reward model or preference data by directly maximizing the utility of speech generations while considering the uncertainty that lies in the inherent variability in subjective human speech perception and evaluations. Experimental results of both subjective and objective evaluations demonstrate that UNO considerably improves the zero-shot performance of TTS models in terms of MOS, word error rate, and speaker similarity. Additionally, we present a remarkable ability of UNO that it can adapt to the desired speaking style in emotional TTS seamlessly and flexibly.
Authors: Xianghe Ma, Dominik Schlechtweg, Wei Zhao
Abstract: In this work, we outline the components and results of our system submitted to the AXOLOTL-24 shared task for Finnish, Russian and German languages. Our system is fully unsupervised. It leverages a graph-based clustering approach to predict mappings between unknown word usages and dictionary entries for Subtask 1, and generates dictionary-like definitions for those novel word usages through the state-of-the-art Large Language Models such as GPT-4 and LLaMA-3 for Subtask 2. In Subtask 1, our system outperforms the baseline system by a large margin, and it offers interpretability for the mapping results by distinguishing between matched and unmatched (novel) word usages through our graph-based clustering approach. Our system ranks first in Finnish and German, and ranks second in Russian on the Subtask 2 test-phase leaderboard. These results show the potential of our system in managing dictionary entries, particularly for updating dictionaries to include novel sense entries. Our code and data are made publicly available\footnote{\url{https://github.com/xiaohemaikoo/axolotl24-ABDN-NLP}}.
Authors: Tomoki Doi, Masaru Isonuma, Hitomi Yanaka
Abstract: As conventional topic models rely on word co-occurrence to infer latent topics, topic modeling for short texts has been a long-standing challenge. Large Language Models (LLMs) can potentially overcome this challenge by contextually learning the semantics of words via pretraining. This paper studies two approaches, parallel prompting and sequential prompting, to use LLMs for topic modeling. Due to the input length limitations, LLMs cannot process many texts at once. By splitting the texts into smaller subsets and processing them parallelly or sequentially, an arbitrary number of texts can be handled by LLMs. Experimental results demonstrated that our methods can identify more coherent topics than existing ones while maintaining the diversity of the induced topics. Furthermore, we found that the inferred topics adequately covered the input texts, while hallucinated topics were hardly generated.
Authors: Zhu Liu
Abstract: Modern neural networks (NNs), trained on extensive raw sentence data, construct distributed representations by compressing individual words into dense, continuous, high-dimensional vectors. These representations are specifically designed to capture the varied meanings, including ambiguity, of word occurrences within context. In this thesis, our objective is to examine the efficacy of distributed representations from NNs in encoding lexical meaning. Initially, we identify four sources of ambiguity - homonymy, polysemy, semantic roles, and multifunctionality - based on the relatedness and similarity of meanings influenced by context. Subsequently, we aim to evaluate these sources by collecting or constructing multilingual datasets, leveraging various language models, and employing linguistic analysis tools.
Authors: Xiaoyuan Li, Wenjie Wang, Moxin Li, Junrong Guo, Yang Zhang, Fuli Feng
Abstract: The rapid advancement of Large Language Models (LLMs) in the realm of mathematical reasoning necessitates comprehensive evaluations to gauge progress and inspire future directions. Existing assessments predominantly focus on problem-solving from the examinee perspective, overlooking a dual perspective of examiner regarding error identification and correction. From the examiner perspective, we define four evaluation tasks for error identification and correction along with a new dataset with annotated error types and steps. We also design diverse prompts to thoroughly evaluate eleven representative LLMs. Our principal findings indicate that GPT-4 outperforms all models, while open-source model LLaMA-2-7B demonstrates comparable abilities to closed-source models GPT-3.5 and Gemini Pro. Notably, calculation error proves the most challenging error type. Moreover, prompting LLMs with the error types can improve the average correction accuracy by 47.9\%. These results reveal potential directions for developing the mathematical reasoning abilities of LLMs. Our code and dataset is available on https://github.com/LittleCirc1e/EIC.
Authors: Weihao Zeng, Can Xu, Yingxiu Zhao, Jian-Guang Lou, Weizhu Chen
Abstract: Fine-tuning large pre-trained language models with Evol-Instruct has achieved encouraging results across a wide range of tasks. However, designing effective evolving methods for instruction evolution requires substantial human expertise. This paper proposes Auto Evol-Instruct, an end-to-end framework that evolves instruction datasets using large language models without any human effort. The framework automatically analyzes and summarizes suitable evolutionary strategies for the given instruction data and iteratively improves the evolving method based on issues exposed during the instruction evolution process. Our extensive experiments demonstrate that the best method optimized by Auto Evol-Instruct outperforms human-designed methods on various benchmarks, including MT-Bench, AlpacaEval, GSM8K, and HumanEval.
Authors: Bar Iluz, Yanai Elazar, Asaf Yehudai, Gabriel Stanovsky
Abstract: Most works on gender bias focus on intrinsic bias -- removing traces of information about a protected group from the model's internal representation. However, these works are often disconnected from the impact of such debiasing on downstream applications, which is the main motivation for debiasing in the first place. In this work, we systematically test how methods for intrinsic debiasing affect neural machine translation models, by measuring the extrinsic bias of such systems under different design choices. We highlight three challenges and mismatches between the debiasing techniques and their end-goal usage, including the choice of embeddings to debias, the mismatch between words and sub-word tokens debiasing, and the effect on different target languages. We find that these considerations have a significant impact on downstream performance and the success of debiasing.
Authors: Md Taimur Ahad
Abstract: Despite the observable benefit of Natural Language Processing (NLP) in processing a large amount of textual medical data within a limited time for information retrieval, a handful of research efforts have been devoted to uncovering novel data-cleaning methods. Data cleaning in NLP is at the centre point for extracting validated information. Another observed limitation in the NLP domain is having limited medical corpora that provide answers to a given medical question. Realising the limitations and challenges from two perspectives, this research aims to clean a medical dataset using ensemble techniques and to develop a corpus. The corpora expect that it will answer the question based on the semantic relationship of corpus sequences. However, the data cleaning method in this research suggests that the ensemble technique provides the highest accuracy (94%) compared to the single process, which includes vectorisation, exploratory data analysis, and feeding the vectorised data. The second aim of having an adequate corpus was realised by extracting answers from the dataset. This research is significant in machine learning, specifically data cleaning and the medical sector, but it also underscores the importance of NLP in the medical field, where accurate and timely information extraction can be a matter of life and death. It establishes text data processing using NLP as a powerful tool for extracting valuable information like image data.
Authors: Lin Gui, Cristina G\^arbacea, Victor Veitch
Abstract: This paper concerns the problem of aligning samples from large language models to human preferences using best-of-$n$ sampling, where we draw $n$ samples, rank them, and return the best one. We consider two fundamental problems. First: what is the relationship between best-of-$n$ and approaches to alignment that train LLMs to output samples with a high expected reward (e.g., RLHF or DPO)? To answer this, we embed both the best-of-$n$ distribution and the sampling distributions learned by alignment procedures in a common class of tiltings of the base LLM distribution. We then show that, within this class, best-of-$n$ is essentially optimal in terms of the trade-off between win-rate against the base model vs KL distance from the base model. That is, best-of-$n$ is the best choice of alignment distribution if the goal is to maximize win rate. However, best-of-$n$ requires drawing $n$ samples for each inference, a substantial cost. To avoid this, the second problem we consider is how to fine-tune a LLM to mimic the best-of-$n$ sampling distribution. We derive BoNBoN Alignment to achieve this by exploiting the special structure of the best-of-$n$ distribution. Experiments show that BoNBoN alignment yields substantial improvements in producing a model that is preferred to the base policy while minimally affecting off-target aspects.
Authors: Kaixin Lan, Tao Fang, Derek F. Wong, Yabo Xu, Lidia S. Chao, Cecilia G. Zhao
Abstract: Pre-trained Language Models (PLMs) have shown impressive results in various Natural Language Generation (NLG) tasks, such as powering chatbots and generating stories. However, an ethical concern arises due to their potential to produce verbatim copies of paragraphs from their training data. This is problematic as PLMs are trained on corpora constructed by human authors. As such, there is a pressing need for research to promote the generation of original content by these models. In this study, we introduce a unique "self-plagiarism" contrastive decoding strategy, aimed at boosting the originality of text produced by PLMs. Our method entails modifying prompts in LLMs to develop an amateur model and a professional model. Specifically, the amateur model is urged to plagiarize using three plagiarism templates we have designed, while the professional model maintains its standard language model status. This strategy employs prompts to stimulate the model's capacity to identify non-original candidate token combinations and subsequently impose penalties. The application of this strategy is integrated prior to the model's final layer, ensuring smooth integration with most existing PLMs (T5, GPT, LLaMA) without necessitating further adjustments. Implementing our strategy, we observe a significant decline in non-original sequences comprised of more than three words in the academic AASC dataset and the story-based ROCStories dataset.
Authors: Ori Ernst, Ori Shapira, Aviv Slobodkin, Sharon Adar, Mohit Bansal, Jacob Goldberger, Ran Levy, Ido Dagan
Abstract: Multi-document summarization (MDS) is a challenging task, often decomposed to subtasks of salience and redundancy detection, followed by text generation. In this context, alignment of corresponding sentences between a reference summary and its source documents has been leveraged to generate training data for some of the component tasks. Yet, this enabling alignment step has usually been applied heuristically on the sentence level on a limited number of subtasks. In this paper, we propose extending the summary-source alignment framework by (1) applying it at the more fine-grained proposition span level, (2) annotating alignment manually in a multi-document setup, and (3) revealing the great potential of summary-source alignments to yield several datasets for at least six different tasks. Specifically, for each of the tasks, we release a manually annotated test set that was derived automatically from the alignment annotation. We also release development and train sets in the same way, but from automatically derived alignments. Using the datasets, each task is demonstrated with baseline models and corresponding evaluation metrics to spur future research on this broad challenge.
Authors: Parastoo Falakaflaki, Mehrnoush Shamsfard
Abstract: This study explores the formality style transfer in Persian, particularly relevant in the face of the increasing prevalence of informal language on digital platforms, which poses challenges for existing Natural Language Processing (NLP) tools. The aim is to transform informal text into formal while retaining the original meaning, addressing both lexical and syntactic differences. We introduce a novel model, Fa-BERT2BERT, based on the Fa-BERT architecture, incorporating consistency learning and gradient-based dynamic weighting. This approach improves the model's understanding of syntactic variations, balancing loss components effectively during training. Our evaluation of Fa-BERT2BERT against existing methods employs new metrics designed to accurately measure syntactic and stylistic changes. Results demonstrate our model's superior performance over traditional techniques across various metrics, including BLEU, BERT score, Rouge-l, and proposed metrics underscoring its ability to adeptly navigate the complexities of Persian language style transfer. This study significantly contributes to Persian language processing by enhancing the accuracy and functionality of NLP models and thereby supports the development of more efficient and reliable NLP applications, capable of handling language style transformation effectively, thereby streamlining content moderation, enhancing data mining results, and facilitating cross-cultural communication.
Authors: Omar Shaikh, Michelle Lam, Joey Hejna, Yijia Shao, Michael Bernstein, Diyi Yang
Abstract: Language models are aligned to emulate the collective voice of many, resulting in outputs that align with no one in particular. Steering LLMs away from generic output is possible through supervised finetuning or RLHF, but requires prohibitively large datasets for new ad-hoc tasks. We argue that it is instead possible to align an LLM to a specific setting by leveraging a very small number ($<10$) of demonstrations as feedback. Our method, Demonstration ITerated Task Optimization (DITTO), directly aligns language model outputs to a user's demonstrated behaviors. Derived using ideas from online imitation learning, DITTO cheaply generates online comparison data by treating users' demonstrations as preferred over output from the LLM and its intermediate checkpoints. We evaluate DITTO's ability to learn fine-grained style and task alignment across domains such as news articles, emails, and blog posts. Additionally, we conduct a user study soliciting a range of demonstrations from participants ($N=16$). Across our benchmarks and user study, we find that win-rates for DITTO outperform few-shot prompting, supervised fine-tuning, and other self-play methods by an average of 19% points. By using demonstrations as feedback directly, DITTO offers a novel method for effective customization of LLMs.
Authors: Xinjian Li, Shinnosuke Takamichi, Takaaki Saeki, William Chen, Sayaka Shiota, Shinji Watanabe
Abstract: In this study, we introduce YODAS (YouTube-Oriented Dataset for Audio and Speech), a large-scale, multilingual dataset comprising currently over 500k hours of speech data in more than 100 languages, sourced from both labeled and unlabeled YouTube speech datasets. The labeled subsets, including manual or automatic subtitles, facilitate supervised model training. Conversely, the unlabeled subsets are apt for self-supervised learning applications. YODAS is distinctive as the first publicly available dataset of its scale, and it is distributed under a Creative Commons license. We introduce the collection methodology utilized for YODAS, which contributes to the large-scale speech dataset construction. Subsequently, we provide a comprehensive analysis of speech, text contained within the dataset. Finally, we describe the speech recognition baselines over the top-15 languages.
Authors: Shuyue Stella Li, Vidhisha Balachandran, Shangbin Feng, Jonathan Ilgen, Emma Pierson, Pang Wei Koh, Yulia Tsvetkov
Abstract: In high-stakes domains like medical reasoning, AI assistants powered by large language models (LLMs) are yet to be reliable and safe. We identify a key obstacle towards reliability: existing LLMs are trained to answer any question, even with incomplete context in the prompt or insufficient parametric knowledge. We propose to change this paradigm to develop more careful LLMs that ask follow-up questions to gather necessary and sufficient information and respond reliably. We introduce MEDIQ, a framework to simulate realistic clinical interactions, which incorporates a Patient System and an adaptive Expert System. The Patient may provide incomplete information in the beginning; the Expert refrains from making diagnostic decisions when unconfident, and instead elicits missing details from the Patient via follow-up questions. To evaluate MEDIQ, we convert MEDQA and CRAFT-MD -- medical benchmarks for diagnostic question answering -- into an interactive setup. We develop a reliable Patient system and prototype several Expert systems, first showing that directly prompting state-of-the-art LLMs to ask questions degrades the quality of clinical reasoning, indicating that adapting LLMs to interactive information-seeking settings is nontrivial. We then augment the Expert with a novel abstention module to better estimate model confidence and decide whether to ask more questions, thereby improving diagnostic accuracy by 20.3%; however, performance still lags compared to an (unrealistic in practice) upper bound when full information is given upfront. Further analyses reveal that interactive performance can be improved by filtering irrelevant contexts and reformatting conversations. Overall, our paper introduces a novel problem towards LLM reliability, a novel MEDIQ framework, and highlights important future directions to extend the information-seeking abilities of LLM assistants in critical domains.
Authors: Ji-Lun Peng, Sijia Cheng, Egil Diau, Yung-Yu Shih, Po-Heng Chen, Yen-Ting Lin, Yun-Nung Chen
Abstract: LLMs have gotten attention across various research domains due to their exceptional performance on a wide range of complex tasks. Therefore, refined methods to evaluate the capabilities of LLMs are needed to determine the tasks and responsibility they should undertake. Our study mainly discussed how LLMs, as useful tools, should be effectively assessed. We proposed the two-stage framework: from ``core ability'' to ``agent'', clearly explaining how LLMs can be applied based on their specific capabilities, along with the evaluation methods in each stage. Core ability refers to the capabilities that LLMs need in order to generate high-quality natural language texts. After confirming LLMs possess core ability, they can solve real-world and complex tasks as agent. In the "core ability" stage, we discussed the reasoning ability, societal impact, and domain knowledge of LLMs. In the ``agent'' stage, we demonstrated embodied action, planning, and tool learning of LLMs agent applications. Finally, we examined the challenges currently confronting the evaluation methods for LLMs, as well as the directions for future development.
Authors: Shicheng Xu, Liang Pang, Huawei Shen, Xueqi Cheng
Abstract: Retrieval-augmented generation (RAG) utilizes retrieved texts to enhance large language models (LLMs). However, studies show that RAG is not consistently effective and can even mislead LLMs due to noisy or incorrect retrieved texts. This suggests that RAG possesses a duality including both benefit and detriment. Although many existing methods attempt to address this issue, they lack a theoretical explanation for the duality in RAG. The benefit and detriment within this duality remain a black box that cannot be quantified or compared in an explainable manner. This paper takes the first step in theoretically giving the essential explanation of benefit and detriment in RAG by: (1) decoupling and formalizing them from RAG prediction, (2) approximating the gap between their values by representation similarity and (3) establishing the trade-off mechanism between them, to make them explainable, quantifiable, and comparable. We demonstrate that the distribution difference between retrieved texts and LLMs' knowledge acts as double-edged sword, bringing both benefit and detriment. We also prove that the actual effect of RAG can be predicted at token level. Based on our theory, we propose a practical novel method, X-RAG, which achieves collaborative generation between pure LLM and RAG at token level to preserve benefit and avoid detriment. Experiments in real-world tasks based on LLMs including OPT, LLaMA-2, and Mistral show the effectiveness of our method and support our theoretical results.
Authors: Shiqi Liu, Sannyuya Liu, Lele Sha, Zijie Zeng, Dragan Gasevic, Zhi Liu
Abstract: Various machine learning approaches have gained significant popularity for the automated classification of educational text to identify indicators of learning engagement -- i.e. learning engagement classification (LEC). LEC can offer comprehensive insights into human learning processes, attracting significant interest from diverse research communities, including Natural Language Processing (NLP), Learning Analytics, and Educational Data Mining. Recently, Large Language Models (LLMs), such as ChatGPT, have demonstrated remarkable performance in various NLP tasks. However, their comprehensive evaluation and improvement approaches in LEC tasks have not been thoroughly investigated. In this study, we propose the Annotation Guidelines-based Knowledge Augmentation (AGKA) approach to improve LLMs. AGKA employs GPT 4.0 to retrieve label definition knowledge from annotation guidelines, and then applies the random under-sampler to select a few typical examples. Subsequently, we conduct a systematic evaluation benchmark of LEC, which includes six LEC datasets covering behavior classification (question and urgency level), emotion classification (binary and epistemic emotion), and cognition classification (opinion and cognitive presence). The study results demonstrate that AGKA can enhance non-fine-tuned LLMs, particularly GPT 4.0 and Llama 3 70B. GPT 4.0 with AGKA few-shot outperforms full-shot fine-tuned models such as BERT and RoBERTa on simple binary classification datasets. However, GPT 4.0 lags in multi-class tasks that require a deep understanding of complex semantic information. Notably, Llama 3 70B with AGKA is a promising combination based on open-source LLM, because its performance is on par with closed-source GPT 4.0 with AGKA. In addition, LLMs struggle to distinguish between labels with similar names in multi-class classification.
Authors: Nikhil Mehta, Dan Goldwasser
Abstract: The large scale usage of social media, combined with its significant impact, has made it increasingly important to understand it. In particular, identifying user communities, can be helpful for many downstream tasks. However, particularly when models are trained on past data and tested on future, doing this is difficult. In this paper, we hypothesize to take advantage of Large Language Models (LLMs), to better identify user communities. Due to the fact that many LLMs, such as ChatGPT, are fixed and must be treated as black-boxes, we propose an approach to better prompt them, by training a smaller LLM to do this. We devise strategies to train this smaller model, showing how it can improve the larger LLMs ability to detect communities. Experimental results show improvements on Reddit and Twitter data, on the tasks of community detection, bot detection, and news media profiling.
Authors: Masha Belyi, Robert Friel, Shuai Shao, Atindriyo Sanyal
Abstract: Retriever Augmented Generation (RAG) systems have become pivotal in enhancing the capabilities of language models by incorporating external knowledge retrieval mechanisms. However, a significant challenge in deploying these systems in industry applications is the detection and mitigation of hallucinations: instances where the model generates information that is not grounded in the retrieved context. Addressing this issue is crucial for ensuring the reliability and accuracy of responses generated by large language models (LLMs) in diverse industry settings. Current hallucination detection techniques fail to deliver accuracy, low latency, and low cost simultaneously. We introduce Luna: a DeBERTA-large (440M) encoder, finetuned for hallucination detection in RAG settings. We demonstrate that Luna outperforms GPT-3.5 and commercial evaluation frameworks on the hallucination detection task, with 97% and 96% reduction in cost and latency, respectively. Luna is lightweight and generalizes across multiple industry verticals and out-of-domain data, making it an ideal candidate for industry LLM applications.
Authors: Yongxin Zhu, Dan Su, Liqiang He, Linli Xu, Dong Yu
Abstract: While recent advancements in speech language models have achieved significant progress, they face remarkable challenges in modeling the long acoustic sequences of neural audio codecs. In this paper, we introduce \textbf{G}enerative \textbf{P}re-trained \textbf{S}peech \textbf{T}ransformer (GPST), a hierarchical transformer designed for efficient speech language modeling. GPST quantizes audio waveforms into two distinct types of discrete speech representations and integrates them within a hierarchical transformer architecture, allowing for a unified one-stage generation process and enhancing Hi-Res audio generation capabilities. By training on large corpora of speeches in an end-to-end unsupervised manner, GPST can generate syntactically consistent speech with diverse speaker identities. Given a brief 3-second prompt, GPST can produce natural and coherent personalized speech, demonstrating in-context learning abilities. Moreover, our approach can be easily extended to spoken cross-lingual speech generation by incorporating multi-lingual semantic tokens and universal acoustic tokens. Experimental results indicate that GPST significantly outperforms the existing speech language models in terms of word error rate, speech quality, and speaker similarity. See \url{https://youngsheen.github.io/GPST/demo} for demo samples.
Authors: Julian Martin Eisenschlos, Hern\'an Maina, Guido Ivetta, Luciana Benotti
Abstract: Recently, large multi-modal models (LMMs) have emerged with the capacity to perform vision tasks such as captioning and visual question answering (VQA) with unprecedented accuracy. Applications such as helping the blind or visually impaired have a critical need for precise answers. It is specially important for models to be well calibrated and be able to quantify their uncertainty in order to selectively decide when to answer and when to abstain or ask for clarifications. We perform the first in-depth analysis of calibration methods and metrics for VQA with in-context learning LMMs. Studying VQA on two answerability benchmarks, we show that the likelihood score of visually grounded models is better calibrated than in their text-only counterparts for in-context learning, where sampling based methods are generally superior, but no clear winner arises. We propose Avg BLEU, a calibration score combining the benefits of both sampling and likelihood methods across modalities.
Authors: Junyu Lu, Bo Xu, Xiaokun Zhang, Kaiyuan Liu, Dongyu Zhang, Liang Yang, Hongfei Lin
Abstract: Current methods of toxic language detection (TLD) typically rely on specific tokens to conduct decisions, which makes them suffer from lexical bias, leading to inferior performance and generalization. Lexical bias has both "useful" and "misleading" impacts on understanding toxicity. Unfortunately, instead of distinguishing between these impacts, current debiasing methods typically eliminate them indiscriminately, resulting in a degradation in the detection accuracy of the model. To this end, we propose a Counterfactual Causal Debiasing Framework (CCDF) to mitigate lexical bias in TLD. It preserves the "useful impact" of lexical bias and eliminates the "misleading impact". Specifically, we first represent the total effect of the original sentence and biased tokens on decisions from a causal view. We then conduct counterfactual inference to exclude the direct causal effect of lexical bias from the total effect. Empirical evaluations demonstrate that the debiased TLD model incorporating CCDF achieves state-of-the-art performance in both accuracy and fairness compared to competitive baselines applied on several vanilla models. The generalization capability of our model outperforms current debiased models for out-of-distribution data.
Authors: Hiroaki Yamagiwa, Ryoma Hashimoto, Kiwamu Arakane, Ken Murakami, Shou Soeda, Momose Oyama, Mariko Okada, Hidetoshi Shimodaira
Abstract: Natural language processing (NLP) is utilized in a wide range of fields, where words in text are typically transformed into feature vectors called embeddings. BioConceptVec is a specific example of embeddings tailored for biology, trained on approximately 30 million PubMed abstracts using models such as skip-gram. Generally, word embeddings are known to solve analogy tasks through simple vector arithmetic. For instance, $\mathrm{\textit{king}} - \mathrm{\textit{man}} + \mathrm{\textit{woman}}$ predicts $\mathrm{\textit{queen}}$. In this study, we demonstrate that BioConceptVec embeddings, along with our own embeddings trained on PubMed abstracts, contain information about drug-gene relations and can predict target genes from a given drug through analogy computations. We also show that categorizing drugs and genes using biological pathways improves performance. Furthermore, we illustrate that vectors derived from known relations in the past can predict unknown future relations in datasets divided by year.
Authors: Yangruibo Ding, Jinjun Peng, Marcus J. Min, Gail Kaiser, Junfeng Yang, Baishakhi Ray
Abstract: Code Large Language Models (Code LLMs) have excelled at tasks like code completion but often miss deeper semantics such as execution effects and dynamic states. This paper aims to bridge the gap between Code LLMs' reliance on static text data and the need for thorough semantic understanding for complex tasks like debugging and program repair. We introduce a novel strategy to train Code LLMs with comprehensive semantics, encompassing high-level functional descriptions, local execution effects of individual statements, and overall input/output behavior, thereby linking static code text with dynamic execution states. We begin by collecting PyX, a clean code corpus of fully executable samples with functional descriptions and execution tracing. We propose training Code LLMs to write code and represent and reason about execution behaviors using natural language, mimicking human verbal debugging. This approach led to the development of SemCoder, a Code LLM with only 6.7B parameters, which shows competitive performance with GPT-3.5-turbo on code generation and execution reasoning tasks. SemCoder achieves 81.1% on HumanEval (GPT-3.5-turbo: 76.8%) and 54.5% on CRUXEval-I (GPT-3.5-turbo: 50.3%). We also study the effectiveness of SemCoder's monologue-style execution reasoning compared to concrete scratchpad reasoning, showing that our approach integrates semantics from multiple dimensions more smoothly. Finally, we demonstrate the potential of applying learned semantics to improve Code LLMs' debugging and self-refining capabilities.
Authors: Junyang Wang, Haiyang Xu, Haitao Jia, Xi Zhang, Ming Yan, Weizhou Shen, Ji Zhang, Fei Huang, Jitao Sang
Abstract: Mobile device operation tasks are increasingly becoming a popular multi-modal AI application scenario. Current Multi-modal Large Language Models (MLLMs), constrained by their training data, lack the capability to function effectively as operation assistants. Instead, MLLM-based agents, which enhance capabilities through tool invocation, are gradually being applied to this scenario. However, the two major navigation challenges in mobile device operation tasks, task progress navigation and focus content navigation, are significantly complicated under the single-agent architecture of existing work. This is due to the overly long token sequences and the interleaved text-image data format, which limit performance. To address these navigation challenges effectively, we propose Mobile-Agent-v2, a multi-agent architecture for mobile device operation assistance. The architecture comprises three agents: planning agent, decision agent, and reflection agent. The planning agent generates task progress, making the navigation of history operations more efficient. To retain focus content, we design a memory unit that updates with task progress. Additionally, to correct erroneous operations, the reflection agent observes the outcomes of each operation and handles any mistakes accordingly. Experimental results indicate that Mobile-Agent-v2 achieves over a 30% improvement in task completion compared to the single-agent architecture of Mobile-Agent. The code is open-sourced at https://github.com/X-PLUG/MobileAgent.
Authors: Emily Ohman, Riikka Rossi
Abstract: What can we learn from the classics of Finnish literature by using computational emotion analysis? This article tries to answer this question by examining how computational methods of sentiment analysis can be used in the study of literary works in conjunction with a qualitative or more 'traditional' approach to literature and affect. We present and develop a simple but robust computational approach of affect analysis that uses a carefully curated emotion lexicon adapted to Finnish turn-of-the-century literary texts combined with word embeddings to map out the semantic emotional spaces of seminal works of Finnish literature. We focus our qualitative analysis on selected case studies: four works by Juhani Aho, Minna Canth, Maria Jotuni, and F. E. Sillanp\"a\"a, but provide emotion arcs for a total of 975 Finnish novels. We argue that a computational analysis of a text's lexicon can be valuable in evaluating the large distribution of the emotional valence in a text and provide guidelines to help other researchers replicate our findings. We show that computational approaches have a place in traditional studies on affect in literature as a support tool for close-reading-based analyses, but also allowing for large-scale comparison between, for example, genres or national canons.
Authors: Mengge Xue, Zhenyu Hu, Meng Zhao, Liqun Liu, Kuo Liao, Shuang Li, Honglin Han, Chengguo Yin
Abstract: Multiple-Choice Questions (MCQs) constitute a critical area of research in the study of Large Language Models (LLMs). Previous works have investigated the selection bias problem in MCQs within few-shot scenarios, in which the LLM's performance may be influenced by the presentation of answer choices, leaving the selection bias during Supervised Fine-Tuning (SFT) unexplored. In this paper, we reveal that selection bias persists in the SFT phase , primarily due to the LLM's inadequate Multiple Choice Symbol Binding (MCSB) ability. This limitation implies that the model struggles to associate the answer options with their corresponding symbols (e.g., A/B/C/D) effectively. To enhance the model's MCSB capability, we first incorporate option contents into the loss function and subsequently adjust the weights of the option symbols and contents, guiding the model to understand the option content of the current symbol. Based on this, we introduce an efficient SFT algorithm for MCQs, termed Point-wise Intelligent Feedback (PIF). PIF constructs negative instances by randomly combining the incorrect option contents with all candidate symbols, and proposes a point-wise loss to provide feedback on these negative samples into LLMs. Our experimental results demonstrate that PIF significantly reduces the model's selection bias by improving its MCSB capability. Remarkably, PIF exhibits a substantial enhancement in the accuracy for MCQs.
Authors: Fatemeh Shiri, Van Nguyen, Farhad Moghimifar, John Yoo, Gholamreza Haffari, Yuan-Fang Li
Abstract: Large Language Models (LLMs) demonstrate significant capabilities in processing natural language data, promising efficient knowledge extraction from diverse textual sources to enhance situational awareness and support decision-making. However, concerns arise due to their susceptibility to hallucination, resulting in contextually inaccurate content. This work focuses on harnessing LLMs for automated Event Extraction, introducing a new method to address hallucination by decomposing the task into Event Detection and Event Argument Extraction. Moreover, the proposed method integrates dynamic schema-aware augmented retrieval examples into prompts tailored for each specific inquiry, thereby extending and adapting advanced prompting techniques such as Retrieval-Augmented Generation. Evaluation findings on prominent event extraction benchmarks and results from a synthesized benchmark illustrate the method's superior performance compared to baseline approaches.
Authors: Jiangming Liu
Abstract: Discourse Representation Structure (DRS) is an innovative semantic representation designed to capture the meaning of texts with arbitrary lengths across languages. The semantic representation parsing is essential for achieving natural language understanding through logical forms. Nevertheless, the performance of DRS parsing models remains constrained when trained exclusively on monolingual data. To tackle this issue, we introduce a cross-lingual training strategy. The proposed method is model-agnostic yet highly effective. It leverages cross-lingual training data and fully exploits the alignments between languages encoded in pre-trained language models. The experiments conducted on the standard benchmarks demonstrate that models trained using the cross-lingual training method exhibit significant improvements in DRS clause and graph parsing in English, German, Italian and Dutch. Comparing our final models to previous works, we achieve state-of-the-art results in the standard benchmarks. Furthermore, the detailed analysis provides deep insights into the performance of the parsers, offering inspiration for future research in DRS parsing. We keep updating new results on benchmarks to the appendix.
Authors: Jun Gao, Ziqiang Cao, Shaoyao Huang, Luozheng Qin, Chunhui Ai
Abstract: ChatGPT is instruct-tuned to generate general and human-expected content to align with human preference through Reinforcement Learning from Human Feedback (RLHF), meanwhile resulting in generated responses not salient enough. Therefore, in this case, ChatGPT may fail to satisfy domain requirements in zero-shot settings, leading to poor ROUGE scores. Inspired by the In-Context Learning (ICL) and retelling ability of ChatGPT, this paper proposes PADS, a \textbf{P}ipeline for \textbf{A}ssisting ChatGPT in \textbf{D}omain \textbf{S}ummarization. PADS consists of a retriever to retrieve similar examples from corpora and a rank model to rerank the multiple candidate summaries generated by ChatGPT. Specifically, given an inference document, we first retrieve an in-context demonstration via the retriever. Then, we require ChatGPT to generate $k$ candidate summaries for the inference document at a time under the guidance of the retrieved demonstration. Finally, the rank model independently scores the $k$ candidate summaries according to their quality and selects the optimal one. We extensively explore dense and sparse retrieval methods to select effective demonstrations for reference and efficiently train the rank model to reflect the quality of candidate summaries for each given summarized document. Additionally, PADS contains merely 400M trainable parameters originating from the rank model and we merely collect 2.5k data to train it. We evaluate PADS on five datasets from different domains, and the result indicates that each module in PADS is committed to effectively guiding ChatGPT to generate salient summaries fitting different domain requirements. Specifically, in the popular summarization dataset Gigaword, PADS achieves over +8 gain on ROUGE-L, compared with the naive ChatGPT in the zero-shot setting. \footnote{Our code are available at \url{https://github.com/jungao1106/PADS}}
Authors: Wrick Talukdar, Anjanava Biswas
Abstract: While supervised learning models have shown remarkable performance in various natural language processing (NLP) tasks, their success heavily relies on the availability of large-scale labeled datasets, which can be costly and time-consuming to obtain. Conversely, unsupervised learning techniques can leverage abundant unlabeled text data to learn rich representations, but they do not directly optimize for specific NLP tasks. This paper presents a novel hybrid approach that synergizes unsupervised and supervised learning to improve the accuracy of NLP task modeling. While supervised models excel at specific tasks, they rely on large labeled datasets. Unsupervised techniques can learn rich representations from abundant unlabeled text but don't directly optimize for tasks. Our methodology integrates an unsupervised module that learns representations from unlabeled corpora (e.g., language models, word embeddings) and a supervised module that leverages these representations to enhance task-specific models. We evaluate our approach on text classification and named entity recognition (NER), demonstrating consistent performance gains over supervised baselines. For text classification, contextual word embeddings from a language model pretrain a recurrent or transformer-based classifier. For NER, word embeddings initialize a BiLSTM sequence labeler. By synergizing techniques, our hybrid approach achieves SOTA results on benchmark datasets, paving the way for more data-efficient and robust NLP systems.
Authors: Wenjing Yue, Xiaoling Wang, Wei Zhu, Ming Guan, Huanran Zheng, Pengfei Wang, Changzhi Sun, Xin Ma
Abstract: Large language models (LLMs) have performed remarkably well in various natural language processing tasks by benchmarking, including in the Western medical domain. However, the professional evaluation benchmarks for LLMs have yet to be covered in the traditional Chinese medicine(TCM) domain, which has a profound history and vast influence. To address this research gap, we introduce TCM-Bench, an comprehensive benchmark for evaluating LLM performance in TCM. It comprises the TCM-ED dataset, consisting of 5,473 questions sourced from the TCM Licensing Exam (TCMLE), including 1,300 questions with authoritative analysis. It covers the core components of TCMLE, including TCM basis and clinical practice. To evaluate LLMs beyond accuracy of question answering, we propose TCMScore, a metric tailored for evaluating the quality of answers generated by LLMs for TCM related questions. It comprehensively considers the consistency of TCM semantics and knowledge. After conducting comprehensive experimental analyses from diverse perspectives, we can obtain the following findings: (1) The unsatisfactory performance of LLMs on this benchmark underscores their significant room for improvement in TCM. (2) Introducing domain knowledge can enhance LLMs' performance. However, for in-domain models like ZhongJing-TCM, the quality of generated analysis text has decreased, and we hypothesize that their fine-tuning process affects the basic LLM capabilities. (3) Traditional metrics for text generation quality like Rouge and BertScore are susceptible to text length and surface semantic ambiguity, while domain-specific metrics such as TCMScore can further supplement and explain their evaluation results. These findings highlight the capabilities and limitations of LLMs in the TCM and aim to provide a more profound assistance to medical research.
Authors: Guangyi Liu, Yongqi Zhang, Yong Li, Quanming Yao
Abstract: The task of reasoning over Knowledge Graphs (KGs) poses a significant challenge for Large Language Models (LLMs) due to the complex structure and large amounts of irrelevant information. Existing LLM reasoning methods overlook the importance of compositional learning on KG to supply with precise knowledge. Besides, the fine-tuning and frequent interaction with LLMs incur substantial time and resource costs. This paper focuses on the Question Answering over Knowledge Graph (KGQA) task and proposes an Explore-then-Determine (EtD) framework that synergizes LLMs with graph neural networks (GNNs) for reasoning over KGs. The Explore stage employs a lightweight GNN to explore promising candidates and relevant fine-grained knowledge to the questions, while the Determine stage utilizes the explored information to construct a knowledge-enhanced multiple-choice prompt, guiding a frozen LLM to determine the final answer. Extensive experiments on three benchmark KGQA datasets demonstrate that EtD achieves state-of-the-art performance and generates faithful reasoning results.
Authors: Yu-Min Tseng, Yu-Chao Huang, Teng-Yun Hsiao, Yu-Ching Hsu, Jia-Yin Foo, Chao-Wei Huang, Yun-Nung Chen
Abstract: Recently, methods investigating how to adapt large language models (LLMs) for specific scenarios have gained great attention. Particularly, the concept of \textit{persona}, originally adopted in dialogue literature, has re-surged as a promising avenue. However, the growing research on persona is relatively disorganized, lacking a systematic overview. To close the gap, we present a comprehensive survey to categorize the current state of the field. We identify two lines of research, namely (1) LLM Role-Playing, where personas are assigned to LLMs, and (2) LLM Personalization, where LLMs take care of user personas. To the best of our knowledge, we present the first survey tailored for LLM role-playing and LLM personalization under the unified view of persona, including taxonomy, current challenges, and potential directions. To foster future endeavors, we actively maintain a paper collection available to the community: https://github.com/MiuLab/PersonaLLM-Survey
Authors: Guanhua Huang, Yuchen Zhang, Zhe Li, Yongjian You, Mingze Wang, Zhouwang Yang
Abstract: The widespread use of large language models (LLMs) has sparked concerns about the potential misuse of AI-generated text, as these models can produce content that closely resembles human-generated text. Current detectors for AI-generated text (AIGT) lack robustness against adversarial perturbations, with even minor changes in characters or words causing a reversal in distinguishing between human-created and AI-generated text. This paper investigates the robustness of existing AIGT detection methods and introduces a novel detector, the Siamese Calibrated Reconstruction Network (SCRN). The SCRN employs a reconstruction network to add and remove noise from text, extracting a semantic representation that is robust to local perturbations. We also propose a siamese calibration technique to train the model to make equally confidence predictions under different noise, which improves the model's robustness against adversarial perturbations. Experiments on four publicly available datasets show that the SCRN outperforms all baseline methods, achieving 6.5\%-18.25\% absolute accuracy improvement over the best baseline method under adversarial attacks. Moreover, it exhibits superior generalizability in cross-domain, cross-genre, and mixed-source scenarios. The code is available at \url{https://github.com/CarlanLark/Robust-AIGC-Detector}.
Authors: Kun Sun, Rong Wang
Abstract: Automated essay scoring (AES) involves predicting a score that reflects the writing quality of an essay. Most existing AES systems produce only a single overall score. However, users and L2 learners expect scores across different dimensions (e.g., vocabulary, grammar, coherence) for English essays in real-world applications. To address this need, we have developed two models that automatically score English essays across multiple dimensions by employing fine-tuning and other strategies on two large datasets. The results demonstrate that our systems achieve impressive performance in evaluation using three criteria: precision, F1 score, and Quadratic Weighted Kappa. Furthermore, our system outperforms existing methods in overall scoring.
Authors: Zhuojun Ding, Wei Wei, Xiaoye Qu, Dangyang Chen
Abstract: Cross-lingual named entity recognition (NER) aims to train an NER model for the target language leveraging only labeled source language data and unlabeled target language data. Prior approaches either perform label projection on translated source language data or employ a source model to assign pseudo labels for target language data and train a target model on these pseudo-labeled data to generalize to the target language. However, these automatic labeling procedures inevitably introduce noisy labels, thus leading to a performance drop. In this paper, we propose a Global-Local Denoising framework (GLoDe) for cross-lingual NER. Specifically, GLoDe introduces a progressive denoising strategy to rectify incorrect pseudo labels by leveraging both global and local distribution information in the semantic space. The refined pseudo-labeled target language data significantly improves the model's generalization ability. Moreover, previous methods only consider improving the model with language-agnostic features, however, we argue that target language-specific features are also important and should never be ignored. To this end, we employ a simple auxiliary task to achieve this goal. Experimental results on two benchmark datasets with six target languages demonstrate that our proposed GLoDe significantly outperforms current state-of-the-art methods.
Authors: Yi Su, Yunpeng Tai, Yixin Ji, Juntao Li, Bowen Yan, Min Zhang
Abstract: Large Language Models (LLMs) have demonstrated an impressive capability known as In-context Learning (ICL), which enables them to acquire knowledge from textual demonstrations without the need for parameter updates. However, many studies have highlighted that the model's performance is sensitive to the choice of demonstrations, presenting a significant challenge for practical applications where we lack prior knowledge of user queries. Consequently, we need to construct an extensive demonstration pool and incorporate external databases to assist the model, leading to considerable time and financial costs. In light of this, some recent research has shifted focus towards zero-shot ICL, aiming to reduce the model's reliance on external information by leveraging their inherent generative capabilities. Despite the effectiveness of these approaches, the content generated by the model may be unreliable, and the generation process is time-consuming. To address these issues, we propose Demonstration Augmentation for In-context Learning (DAIL), which employs the model's previously predicted historical samples as demonstrations for subsequent ones. DAIL brings no additional inference cost and does not rely on the model's generative capabilities. Our experiments reveal that DAIL can significantly improve the model's performance over direct zero-shot inference and can even outperform few-shot ICL without any external information.
Authors: Zixuan Dong, Baoyun Peng, Yufei Wang, Jia Fu, Xiaodong Wang, Yongxue Shan, Xin Zhou
Abstract: While large language models (LLMs) have shown remarkable capabilities in natural language processing, they struggle with complex, multi-step reasoning tasks involving knowledge graphs (KGs). Existing approaches that integrate LLMs and KGs either underutilize the reasoning abilities of LLMs or suffer from prohibitive computational costs due to tight coupling. To address these limitations, we propose a novel collaborative framework named EffiQA that can strike a balance between performance and efficiency via an iterative paradigm. EffiQA consists of three stages: global planning, efficient KG exploration, and self-reflection. Specifically, EffiQA leverages the commonsense capability of LLMs to explore potential reasoning pathways through global planning. Then, it offloads semantic pruning to a small plug-in model for efficient KG exploration. Finally, the exploration results are fed to LLMs for self-reflection to further improve the global planning and efficient KG exploration. Empirical evidence on multiple KBQA benchmarks shows EffiQA's effectiveness, achieving an optimal balance between reasoning accuracy and computational costs. We hope the proposed new framework will pave the way for efficient, knowledge-intensive querying by redefining the integration of LLMs and KGs, fostering future research on knowledge-based question answering.
Authors: Boxi Cao, Keming Lu, Xinyu Lu, Jiawei Chen, Mengjie Ren, Hao Xiang, Peilin Liu, Yaojie Lu, Ben He, Xianpei Han, Le Sun, Hongyu Lin, Bowen Yu
Abstract: Alignment is the most critical step in building large language models (LLMs) that meet human needs. With the rapid development of LLMs gradually surpassing human capabilities, traditional alignment methods based on human-annotation are increasingly unable to meet the scalability demands. Therefore, there is an urgent need to explore new sources of automated alignment signals and technical approaches. In this paper, we systematically review the recently emerging methods of automated alignment, attempting to explore how to achieve effective, scalable, automated alignment once the capabilities of LLMs exceed those of humans. Specifically, we categorize existing automated alignment methods into 4 major categories based on the sources of alignment signals and discuss the current status and potential development of each category. Additionally, we explore the underlying mechanisms that enable automated alignment and discuss the essential factors that make automated alignment technologies feasible and effective from the fundamental role of alignment.
Authors: Zhenya Huang, Yuting Ning, Longhu Qin, Shiwei Tong, Shangzi Xue, Tong Xiao, Xin Lin, Jiayu Liu, Qi Liu, Enhong Chen, Shijing Wang
Abstract: Educational resource understanding is vital to online learning platforms, which have demonstrated growing applications recently. However, researchers and developers always struggle with using existing general natural language toolkits or domain-specific models. The issue raises a need to develop an effective and easy-to-use one that benefits AI education-related research and applications. To bridge this gap, we present a unified, modularized, and extensive library, \textit{EduNLP}, focusing on educational resource understanding. In the library, we decouple the whole workflow to four key modules with consistent interfaces including data configuration, processing, model implementation, and model evaluation. We also provide a configurable pipeline to unify the data usage and model usage in standard ways, where users can customize their own needs. For the current version, we primarily provide 10 typical models from four categories, and 5 common downstream-evaluation tasks in the education domain on 8 subjects for users' usage. The project is released at: https://github.com/bigdata-ustc/EduNLP.
Authors: Jungmin Yun, Mihyeon Kim, Youngbin Kim
Abstract: Transformer-based models have achieved dominant performance in numerous NLP tasks. Despite their remarkable successes, pre-trained transformers such as BERT suffer from a computationally expensive self-attention mechanism that interacts with all tokens, including the ones unfavorable to classification performance. To overcome these challenges, we propose integrating two strategies: token pruning and token combining. Token pruning eliminates less important tokens in the attention mechanism's key and value as they pass through the layers. Additionally, we adopt fuzzy logic to handle uncertainty and alleviate potential mispruning risks arising from an imbalanced distribution of each token's importance. Token combining, on the other hand, condenses input sequences into smaller sizes in order to further compress the model. By integrating these two approaches, we not only improve the model's performance but also reduce its computational demands. Experiments with various datasets demonstrate superior performance compared to baseline models, especially with the best improvement over the existing BERT model, achieving +5%p in accuracy and +5.6%p in F1 score. Additionally, memory cost is reduced to 0.61x, and a speedup of 1.64x is achieved.
Authors: Xiaosen Zheng, Tianyu Pang, Chao Du, Qian Liu, Jing Jiang, Min Lin
Abstract: Recently, Anil et al. (2024) show that many-shot (up to hundreds of) demonstrations can jailbreak state-of-the-art LLMs by exploiting their long-context capability. Nevertheless, is it possible to use few-shot demonstrations to efficiently jailbreak LLMs within limited context sizes? While the vanilla few-shot jailbreaking may be inefficient, we propose improved techniques such as injecting special system tokens like [/INST] and employing demo-level random search from a collected demo pool. These simple techniques result in surprisingly effective jailbreaking against aligned LLMs (even with advanced defenses). For examples, our method achieves >80% (mostly >95%) ASRs on Llama-2-7B and Llama-3-8B without multiple restarts, even if the models are enhanced by strong defenses such as perplexity detection and/or SmoothLLM, which is challenging for suffix-based jailbreaking. In addition, we conduct comprehensive and elaborate (e.g., making sure to use correct system prompts) evaluations against other aligned LLMs and advanced defenses, where our method consistently achieves nearly 100% ASRs. Our code is available at https://github.com/sail-sg/I-FSJ.
Authors: Ryo Kamoi, Yusen Zhang, Nan Zhang, Jiawei Han, Rui Zhang
Abstract: Self-correction is an approach to improving responses from large language models (LLMs) by refining the responses using LLMs during inference. Prior work has proposed various self-correction frameworks using different sources of feedback, including self-evaluation and external feedback. However, there is still no consensus on the question of when LLMs can correct their own mistakes, as recent studies also report negative results. In this work, we critically survey broad papers and discuss the conditions required for successful self-correction. We first find that prior studies often do not define their research questions in detail and involve impractical frameworks or unfair evaluations that over-evaluate self-correction. To tackle these issues, we categorize research questions in self-correction research and provide a checklist for designing appropriate experiments. Our critical survey based on the newly categorized research questions shows that (1) no prior work demonstrates successful self-correction with feedback from prompted LLMs in general tasks, (2) self-correction works well in tasks that can use reliable external feedback, and (3) large-scale fine-tuning enables self-correction.
Authors: Dong Chen, Shaoxin Lin, Muhan Zeng, Daoguang Zan, Jian-Gang Wang, Anton Cheshkov, Jun Sun, Hao Yu, Guoliang Dong, Artem Aliev, Jie Wang, Xiao Cheng, Guangtai Liang, Yuchi Ma, Pan Bian, Tao Xie, Qianxiang Wang
Abstract: GitHub issue resolving recently has attracted significant attention from academia and industry. SWE-bench is proposed to measure the performance in resolving issues. In this paper, we propose CodeR, which adopts a multi-agent framework and pre-defined task graphs to Repair & Resolve reported bugs and add new features within code Repository. On SWE-bench lite, CodeR is able to solve 28.00% of issues, in the case of submitting only once for each issue. We examine the performance impact of each design of CodeR and offer insights to advance this research direction.
Authors: Fanyi Qu, Hao Sun, Yunfang Wu
Abstract: Within the context of reading comprehension, the task of Distractor Generation (DG) aims to generate several incorrect options to confuse readers. Traditional supervised methods for DG rely heavily on expensive human-annotated distractor labels. In this paper, we propose an unsupervised DG framework, leveraging Large Language Models (LLMs) as cost-effective annotators to enhance the DG capability of smaller student models. Specially, to perform knowledge distilling, we propose a dual task training strategy that integrates pseudo distractors from LLMs and the original answer in-formation as the objective targets with a two-stage training process. Moreover, we devise a counterfactual contrastive decoding mechanism for increasing the distracting capability of the DG model. Experiments show that our unsupervised generation method with Bart-base greatly surpasses GPT-3.5-turbo performance with only 200 times fewer model parameters. Our proposed unsupervised DG method offers a cost-effective framework for practical reading comprehension applications, without the need of laborious distractor annotation and costly large-size models
Authors: Sushant Gautam
Abstract: Fact-checking is a crucial natural language processing (NLP) task that verifies the truthfulness of claims by considering reliable evidence. Traditional methods are often limited by labour-intensive data curation and rule-based approaches. In this paper, we present FactGenius, a novel method that enhances fact-checking by combining zero-shot prompting of large language models (LLMs) with fuzzy text matching on knowledge graphs (KGs). Leveraging DBpedia, a structured linked data dataset derived from Wikipedia, FactGenius refines LLM-generated connections using similarity measures to ensure accuracy. The evaluation of FactGenius on the FactKG, a benchmark dataset for fact verification, demonstrates that it significantly outperforms existing baselines, particularly when fine-tuning RoBERTa as a classifier. The two-stage approach of filtering and validating connections proves crucial, achieving superior performance across various reasoning types and establishing FactGenius as a promising tool for robust fact-checking. The code and materials are available at https://github.com/SushantGautam/FactGenius.
Authors: Zhenhua Liu, Tong Zhu, Chuanyuan Tan, Haonan Lu, Bing Liu, Wenliang Chen
Abstract: Large Language Models (LLMs) have shown their impressive capabilities, while also raising concerns about the data contamination problems due to privacy issues and leakage of benchmark datasets in the pre-training phase. Therefore, it is vital to detect the contamination by checking whether an LLM has been pre-trained on the target texts. Recent studies focus on the generated texts and compute perplexities, which are superficial features and not reliable. In this study, we propose to utilize the probing technique for pre-training data detection by examining the model's internal activations. Our method is simple and effective and leads to more trustworthy pre-training data detection. Additionally, we propose ArxivMIA, a new challenging benchmark comprising arxiv abstracts from Computer Science and Mathematics categories. Our experiments demonstrate that our method outperforms all baselines, and achieves state-of-the-art performance on both WikiMIA and ArxivMIA, with additional experiments confirming its efficacy (Our code and dataset are available at https://github.com/zhliu0106/probing-lm-data).
Authors: Ken Deng, Jiaheng Liu, He Zhu, Congnan Liu, Jingxin Li, Jiakai Wang, Peng Zhao, Chenchen Zhang, Yanan Wu, Xueqiao Yin, Yuanxing Zhang, Wenbo Su, Bangyu Xiang, Tiezheng Ge, Bo Zheng
Abstract: Code completion models have made significant progress in recent years. Recently, repository-level code completion has drawn more attention in modern software development, and several baseline methods and benchmarks have been proposed. However, existing repository-level code completion methods often fall short of fully using the extensive context of a project repository, such as the intricacies of relevant files and class hierarchies. Besides, the existing benchmarks usually focus on limited code completion scenarios, which cannot reflect the repository-level code completion abilities well of existing methods. To address these limitations, we propose the R2C2-Coder to enhance and benchmark the real-world repository-level code completion abilities of code Large Language Models, where the R2C2-Coder includes a code prompt construction method R2C2-Enhance and a well-designed benchmark R2C2-Bench. Specifically, first, in R2C2-Enhance, we first construct the candidate retrieval pool and then assemble the completion prompt by retrieving from the retrieval pool for each completion cursor position. Second, based on R2C2 -Enhance, we can construct a more challenging and diverse R2C2-Bench with training, validation and test splits, where a context perturbation strategy is proposed to simulate the real-world repository-level code completion well. Extensive results on multiple benchmarks demonstrate the effectiveness of our R2C2-Coder.
Authors: Sichun Luo, Wei Shao, Yuxuan Yao, Jian Xu, Mingyang Liu, Qintong Li, Bowei He, Maolin Wang, Guanzhi Deng, Hanxu Hou, Xinyi Zhang, Linqi Song
Abstract: Nowadays, large language models (LLMs) have been integrated with conventional recommendation models to improve recommendation performance. However, while most of the existing works have focused on improving the model performance, the privacy issue has only received comparatively less attention. In this paper, we review recent advancements in privacy within LLM-based recommendation, categorizing them into privacy attacks and protection mechanisms. Additionally, we highlight several challenges and propose future directions for the community to address these critical problems.
Authors: Cem Bozsahin
Abstract: TheBench is a tool to study monadic structures in natural language. It is for writing monadic grammars to explore analyses, compare diverse languages through their categories, and to train models of grammar from form-meaning pairs where syntax is latent variable. Monadic structures are binary combinations of elements that employ semantics of composition only. TheBench is essentially old-school categorial grammar to syntacticize the idea, with the implication that although syntax is autonomous (recall \emph{colorless green ideas sleep furiously}), the treasure is in the baggage it carries at every step, viz. semantics, more narrowly, predicate-argument structures indicating choice of categorial reference and its consequent placeholders for decision in such structures. There is some new thought in old school. Unlike traditional categorial grammars, application is turned into composition in monadic analysis. Moreover, every correspondence requires specifying two command relations, one on syntactic command and the other on semantic command. A monadic grammar of TheBench contains only synthetic elements (called `objects' in category theory of mathematics) that are shaped by this analytic invariant, viz. composition. Both ingredients (command relations) of any analytic step must therefore be functions (`arrows' in category theory). TheBench is one implementation of the idea for iterative development of such functions along with grammar of synthetic elements.
Authors: Haoran Que, Jiaheng Liu, Ge Zhang, Chenchen Zhang, Xingwei Qu, Yinghao Ma, Feiyu Duan, Zhiqi Bai, Jiakai Wang, Yuanxing Zhang, Xu Tan, Jie Fu, Wenbo Su, Jiamang Wang, Lin Qu, Bo Zheng
Abstract: Continual Pre-Training (CPT) on Large Language Models (LLMs) has been widely used to expand the model's fundamental understanding of specific downstream domains (e.g., math and code). For the CPT on domain-specific LLMs, one important question is how to choose the optimal mixture ratio between the general-corpus (e.g., Dolma, Slim-pajama) and the downstream domain-corpus. Existing methods usually adopt laborious human efforts by grid-searching on a set of mixture ratios, which require high GPU training consumption costs. Besides, we cannot guarantee the selected ratio is optimal for the specific domain. To address the limitations of existing methods, inspired by the Scaling Law for performance prediction, we propose to investigate the Scaling Law of the Domain-specific Continual Pre-Training (D-CPT Law) to decide the optimal mixture ratio with acceptable training costs for LLMs of different sizes. Specifically, by fitting the D-CPT Law, we can easily predict the general and downstream performance of arbitrary mixture ratios, model sizes, and dataset sizes using small-scale training costs on limited experiments. Moreover, we also extend our standard D-CPT Law on cross-domain settings and propose the Cross-Domain D-CPT Law to predict the D-CPT law of target domains, where very small training costs (about 1% of the normal training costs) are needed for the target domains. Comprehensive experimental results on six downstream domains demonstrate the effectiveness and generalizability of our proposed D-CPT Law and Cross-Domain D-CPT Law.
Authors: Keyon Vafa, Ashesh Rambachan, Sendhil Mullainathan
Abstract: What makes large language models (LLMs) impressive is also what makes them hard to evaluate: their diversity of uses. To evaluate these models, we must understand the purposes they will be used for. We consider a setting where these deployment decisions are made by people, and in particular, people's beliefs about where an LLM will perform well. We model such beliefs as the consequence of a human generalization function: having seen what an LLM gets right or wrong, people generalize to where else it might succeed. We collect a dataset of 19K examples of how humans make generalizations across 79 tasks from the MMLU and BIG-Bench benchmarks. We show that the human generalization function can be predicted using NLP methods: people have consistent structured ways to generalize. We then evaluate LLM alignment with the human generalization function. Our results show that -- especially for cases where the cost of mistakes is high -- more capable models (e.g. GPT-4) can do worse on the instances people choose to use them for, exactly because they are not aligned with the human generalization function.
Authors: Da Ma, Lu Chen, Pengyu Wang, Hongshen Xu, Hanqi Li, Liangtai Sun, Su Zhu, Shuai Fan, Kai Yu
Abstract: Large language models (LLMs) have demonstrated proficiency across various natural language processing (NLP) tasks but often require additional training, such as continual pre-training and supervised fine-tuning. However, the costs associated with this, primarily due to their large parameter count, remain high. This paper proposes leveraging \emph{sparsity} in pre-trained LLMs to expedite this training process. By observing sparsity in activated neurons during forward iterations, we identify the potential for computational speed-ups by excluding inactive neurons. We address associated challenges by extending existing neuron importance evaluation metrics and introducing a ladder omission rate scheduler. Our experiments on Llama-2 demonstrate that Sparsity-Accelerated Training (SAT) achieves comparable or superior performance to standard training while significantly accelerating the process. Specifically, SAT achieves a $45\%$ throughput improvement in continual pre-training and saves $38\%$ training time in supervised fine-tuning in practice. It offers a simple, hardware-agnostic, and easily deployable framework for additional LLM training. Our code is available at https://github.com/OpenDFM/SAT.
Authors: Martin J. Hetz, Nicolas Carl, Sarah Haggenm\"uller, Christoph Wies, Maurice Stephan Michel, Frederik Wessels, Titus J. Brinker
Abstract: Large Language Models (LLMs) are revolutionizing medical Question-Answering (medQA) through extensive use of medical literature. However, their performance is often hampered by outdated training data and a lack of explainability, which limits clinical applicability. This study aimed to create and assess UroBot, a urology-specialized chatbot, by comparing it with state-of-the-art models and the performance of urologists on urological board questions, ensuring full clinician-verifiability. UroBot was developed using OpenAI's GPT-3.5, GPT-4, and GPT-4o models, employing retrieval-augmented generation (RAG) and the latest 2023 guidelines from the European Association of Urology (EAU). The evaluation included ten runs of 200 European Board of Urology (EBU) In-Service Assessment (ISA) questions, with performance assessed by the mean Rate of Correct Answers (RoCA). UroBot-4o achieved an average RoCA of 88.4%, surpassing GPT-4o by 10.8%, with a score of 77.6%. It was also clinician-verifiable and exhibited the highest run agreement as indicated by Fleiss' Kappa (k = 0.979). By comparison, the average performance of urologists on board questions, as reported in the literature, is 68.7%. UroBot's clinician-verifiable nature and superior accuracy compared to both existing models and urologists on board questions highlight its potential for clinical integration. The study also provides the necessary code and instructions for further development of UroBot.
Authors: Cheng-Hsun Hsueh, Paul Kuo-Ming Huang, Tzu-Han Lin, Che-Wei Liao, Hung-Chieh Fang, Chao-Wei Huang, Yun-Nung Chen
Abstract: Knowledge editing is a rising technique for efficiently updating factual knowledge in Large Language Models (LLMs) with minimal alteration of parameters. However, recent studies have identified concerning side effects, such as knowledge distortion and the deterioration of general abilities, that have emerged after editing. This survey presents a comprehensive study of these side effects, providing a unified view of the challenges associated with knowledge editing in LLMs. We discuss related works and summarize potential research directions to overcome these limitations. Our work highlights the limitations of current knowledge editing methods, emphasizing the need for deeper understanding of inner knowledge structures of LLMs and improved knowledge editing methods. To foster future research, we have released the complementary materials such as paper collection publicly at https://github.com/MiuLab/EditLLM-Survey
Authors: Yongjing Yin, Jiali Zeng, Yafu Li, Fandong Meng, Yue Zhang
Abstract: The fine-tuning of open-source large language models (LLMs) for machine translation has recently received considerable attention, marking a shift towards data-centric research from traditional neural machine translation. However, the area of data collection for instruction fine-tuning in machine translation remains relatively underexplored. In this paper, we present LexMatcher, a simple yet effective method for data collection that leverages bilingual dictionaries to generate a dataset, the design of which is driven by the coverage of senses found in these dictionaries. The dataset comprises a subset retrieved from an existing corpus and a smaller synthesized subset which supplements the infrequent senses of polysemous words. Utilizing LLaMA2 as our base model, our approach outperforms the established baselines on the WMT2022 test sets and also exhibits significant performance improvements in tasks related to word sense disambiguation and specialized terminology translation. These results underscore the effectiveness of LexMatcher in enhancing LLM-based machine translation.
Authors: Ara Yeroyan (Data Science Department, American University of Armenia), Nikolay Karpov (Nvidia, NeMo Conversational AI team)
Abstract: In recent years, automatic speech recognition (ASR) systems have significantly improved, especially in languages with a vast amount of transcribed speech data. However, ASR systems tend to perform poorly for low-resource languages with fewer resources, such as minority and regional languages. This study introduces a novel pipeline designed to generate ASR training datasets from audiobooks, which typically feature a single transcript associated with hours-long audios. The common structure of these audiobooks poses a unique challenge due to the extensive length of audio segments, whereas optimal ASR training requires segments ranging from 4 to 15 seconds. To address this, we propose a method for effectively aligning audio with its corresponding text and segmenting it into lengths suitable for ASR training. Our approach simplifies data preparation for ASR systems in low-resource languages and demonstrates its application through a case study involving the Armenian language. Our method, which is "portable" to many low-resource languages, not only mitigates the issue of data scarcity but also enhances the performance of ASR models for underrepresented languages.
Authors: Hakaze Cho, Yoshihiro Sakai, Kenshiro Tanaka, Mariko Kato, Naoya Inoue
Abstract: In this paper, we investigate the output token probability information in the output embedding of language models. We provide an approximate common log-linear encoding of output token probabilities within the output embedding vectors and demonstrate that it is accurate and sparse when the output space is large and output logits are concentrated. Based on such findings, we edit the encoding in output embedding to modify the output probability distribution accurately. Moreover, the sparsity we find in output probability encoding suggests that a large number of dimensions in the output embedding do not contribute to causal language modeling. Therefore, we attempt to delete the output-unrelated dimensions and find more than 30% of the dimensions can be deleted without significant movement in output distribution and degeneration on sequence generation. Additionally, in training dynamics, we use such encoding as a probe and find that the output embeddings capture token frequency information in early steps, even before an obvious convergence starts.
Authors: Zi-Yi Dou, Cheng-Fu Yang, Xueqing Wu, Kai-Wei Chang, Nanyun Peng
Abstract: Self-training can potentially improve the performance of language agents without relying on demonstrations from humans or stronger models. The general process involves generating samples from a model, evaluating their quality, and updating the model by training on high-quality samples. However, self-training can face limitations because achieving good performance requires a good amount of high-quality samples, yet relying solely on model sampling for obtaining such samples can be inefficient. In addition, these methods often disregard low-quality samples, failing to leverage them effectively. To address these limitations, we present Reflection-Reinforced Self-Training (Re-ReST), which leverages a reflection model to refine low-quality samples and subsequently uses these improved samples to augment self-training. The reflection model takes both the model output and feedback from an external environment (e.g., unit test results in code generation) as inputs and produces improved samples as outputs. By employing this technique, we effectively enhance the quality of inferior samples, and enrich the self-training dataset with higher-quality samples efficiently. We perform extensive experiments on open-source language agents across tasks, including multi-hop question answering, sequential decision-making, code generation, visual question answering, and text-to-image generation. Results demonstrate improvements over self-training baselines across settings. Moreover, ablation studies confirm the reflection model's efficiency in generating quality self-training samples and its compatibility with self-consistency decoding.
Authors: Kiho Park, Yo Joong Choe, Yibo Jiang, Victor Veitch
Abstract: Understanding how semantic meaning is encoded in the representation spaces of large language models is a fundamental problem in interpretability. In this paper, we study the two foundational questions in this area. First, how are categorical concepts, such as {'mammal', 'bird', 'reptile', 'fish'}, represented? Second, how are hierarchical relations between concepts encoded? For example, how is the fact that 'dog' is a kind of 'mammal' encoded? We show how to extend the linear representation hypothesis to answer these questions. We find a remarkably simple structure: simple categorical concepts are represented as simplices, hierarchically related concepts are orthogonal in a sense we make precise, and (in consequence) complex concepts are represented as polytopes constructed from direct sums of simplices, reflecting the hierarchical structure. We validate these theoretical results on the Gemma large language model, estimating representations for 957 hierarchically related concepts using data from WordNet.
Authors: Yiqian Yang, Hyejeong Jo, Yiqun Duan, Qiang Zhang, Jinni Zhou, Won Hee Lee, Renjing Xu, Hui Xiong
Abstract: Deciphering language from brain activity is a crucial task in brain-computer interface (BCI) research. Non-invasive cerebral signaling techniques including electroencephalography (EEG) and magnetoencephalography (MEG) are becoming increasingly popular due to their safety and practicality, avoiding invasive electrode implantation. However, current works under-investigated three points: 1) a predominant focus on EEG with limited exploration of MEG, which provides superior signal quality; 2) poor performance on unseen text, indicating the need for models that can better generalize to diverse linguistic contexts; 3) insufficient integration of information from other modalities, which could potentially constrain our capacity to comprehensively understand the intricate dynamics of brain activity. This study presents a novel approach for translating MEG signals into text using a speech-decoding framework with multiple alignments. Our method is the first to introduce an end-to-end multi-alignment framework for totally unseen text generation directly from MEG signals. We achieve an impressive BLEU-1 score on the $\textit{GWilliams}$ dataset, significantly outperforming the baseline from 5.49 to 10.44 on the BLEU-1 metric. This improvement demonstrates the advancement of our model towards real-world applications and underscores its potential in advancing BCI research. Code is available at $\href{https://github.com/NeuSpeech/MAD-MEG2text}{https://github.com/NeuSpeech/MAD-MEG2text}$.
URLs: https://github.com/NeuSpeech/MAD-MEG2text, https://github.com/NeuSpeech/MAD-MEG2text
Authors: Haozheng Luo, Jiahao Yu, Wenxin Zhang, Jialong Li, Jerry Yao-Chieh Hu, Xingyu Xin, Han Liu
Abstract: We introduce a low-resource safety enhancement method for aligning large language models (LLMs) without the need for supervised fine-tuning (SFT) or reinforcement learning from human feedback (RLHF). Our main idea is to exploit knowledge distillation to extract the alignment information from existing well-aligned LLMs and integrate it into unaligned LLMs in a plug-and-play fashion. Methodology, we employ delta debugging to identify the critical components of knowledge necessary for effective distillation. On the harmful question dataset, our method significantly enhances the average defense success rate by approximately 14.41%, reaching as high as 51.39%, in 17 unaligned pre-trained LLMs, without compromising performance.
Authors: Ebrahim Feghhi, Nima Hadidi, Bryan Song, Idan A. Blank, Jonathan C. Kao
Abstract: Given the remarkable capabilities of large language models (LLMs), there has been a growing interest in evaluating their similarity to the human brain. One approach towards quantifying this similarity is by measuring how well a model predicts neural signals, also called "brain score". Internal representations from LLMs achieve state-of-the-art brain scores, leading to speculation that they share computational principles with human language processing. This inference is only valid if the subset of neural activity predicted by LLMs reflects core elements of language processing. Here, we question this assumption by analyzing three neural datasets used in an impactful study on LLM-to-brain mappings, with a particular focus on an fMRI dataset where participants read short passages. We first find that when using shuffled train-test splits, as done in previous studies with these datasets, a trivial feature that encodes temporal autocorrelation not only outperforms LLMs but also accounts for the majority of neural variance that LLMs explain. We therefore use contiguous splits moving forward. Second, we explain the surprisingly high brain scores of untrained LLMs by showing they do not account for additional neural variance beyond two simple features: sentence length and sentence position. This undermines evidence used to claim that the transformer architecture biases computations to be more brain-like. Third, we find that brain scores of trained LLMs on this dataset can largely be explained by sentence length, position, and pronoun-dereferenced static word embeddings; a small, additional amount is explained by sense-specific embeddings and contextual representations of sentence structure. We conclude that over-reliance on brain scores can lead to over-interpretations of similarity between LLMs and brains, and emphasize the importance of deconstructing what LLMs are mapping to in neural signals.
Authors: Kun Zhu, Xiaocheng Feng, Xiyuan Du, Yuxuan Gu, Weijiang Yu, Haotian Wang, Qianglong Chen, Zheng Chu, Jingchang Chen, Bing Qin
Abstract: Retrieval-augmented generation integrates the capabilities of large language models with relevant information retrieved from an extensive corpus, yet encounters challenges when confronted with real-world noisy data. One recent solution is to train a filter module to find relevant content but only achieve suboptimal noise compression. In this paper, we propose to introduce the information bottleneck theory into retrieval-augmented generation. Our approach involves the filtration of noise by simultaneously maximizing the mutual information between compression and ground output, while minimizing the mutual information between compression and retrieved passage. In addition, we derive the formula of information bottleneck to facilitate its application in novel comprehensive evaluations, the selection of supervised fine-tuning data, and the construction of reinforcement learning rewards. Experimental results demonstrate that our approach achieves significant improvements across various question answering datasets, not only in terms of the correctness of answer generation but also in the conciseness with $2.5\%$ compression rate.
Authors: Fangcong Yin, Xi Ye, Greg Durrett
Abstract: Recent work in interpretability shows that large language models (LLMs) can be adapted for new tasks in a learning-free way: it is possible to intervene on LLM representations to elicit desired behaviors for alignment. For instance, adding certain bias vectors to the outputs of certain attention heads is reported to boost the truthfulness of models. In this work, we show that localized fine-tuning serves as an effective alternative to such representation intervention methods. We introduce a framework called Localized Fine-Tuning on LLM Representations (LoFiT), which identifies a subset of attention heads that are most important for learning a specific task, then trains offset vectors to add to the model's hidden representations at those selected heads. LoFiT localizes to a sparse set of heads (3%) and learns the offset vectors from limited training data, comparable to the settings used for representation intervention. For truthfulness and reasoning tasks, we find that LoFiT's intervention vectors are more effective for LLM adaptation than vectors from representation intervention methods such as Inference-time Intervention. We also find that the localization step is important: selecting a task-specific set of attention heads can lead to higher performance than intervening on heads selected for a different task. Finally, for the tasks we study, LoFiT achieves comparable performance to other parameter-efficient fine-tuning methods such as LoRA, despite modifying 20x-200x fewer parameters than these methods.
Authors: Yubo Wang, Xueguang Ma, Ge Zhang, Yuansheng Ni, Abhranil Chandra, Shiguang Guo, Weiming Ren, Aaran Arulraj, Xuan He, Ziyan Jiang, Tianle Li, Max Ku, Kai Wang, Alex Zhuang, Rongqi Fan, Xiang Yue, Wenhu Chen
Abstract: In the age of large-scale language models, benchmarks like the Massive Multitask Language Understanding (MMLU) have been pivotal in pushing the boundaries of what AI can achieve in language comprehension and reasoning across diverse domains. However, as models continue to improve, their performance on these benchmarks has begun to plateau, making it increasingly difficult to discern differences in model capabilities. This paper introduces MMLU-Pro, an enhanced dataset designed to extend the mostly knowledge-driven MMLU benchmark by integrating more challenging, reasoning-focused questions and expanding the choice set from four to ten options. Additionally, MMLU-Pro eliminates the trivial and noisy questions in MMLU. Our experimental results show that MMLU-Pro not only raises the challenge, causing a significant drop in accuracy by 16% to 33% compared to MMLU but also demonstrates greater stability under varying prompts. With 24 different prompt styles tested, the sensitivity of model scores to prompt variations decreased from 4-5% in MMLU to just 2% in MMLU-Pro. Additionally, we found that models utilizing Chain of Thought (CoT) reasoning achieved better performance on MMLU-Pro compared to direct answering, which is in stark contrast to the findings on the original MMLU, indicating that MMLU-Pro includes more complex reasoning questions. Our assessments confirm that MMLU-Pro is a more discriminative benchmark to better track progress in the field.
Authors: Yaohua Liu
Abstract: With the rapid development of drone technology, the application of multi-drones is becoming increasingly widespread in various fields. However, the task planning technology for multi-drones still faces challenges such as the complexity of remote operation and the convenience of human-machine interaction. To address these issues, this paper proposes a prompt-driven task planning method for multi-drones based on large language models. By introducing the Prompt technique, appropriate prompt information is provided for the multi-drone system.
Authors: Arne Binder, Leonhard Hennig, Christoph Alt
Abstract: The objective of Information Extraction (IE) is to derive structured representations from unstructured or semi-structured documents. However, developing IE models is complex due to the need of integrating several subtasks. Additionally, representation of data among varied tasks and transforming datasets into task-specific model inputs presents further challenges. To streamline this undertaking for researchers, we introduce PyTorch-IE, a deep-learning-based framework uniquely designed to enable swift, reproducible, and reusable implementations of IE models. PyTorch-IE offers a flexible data model capable of creating complex data structures by integrating interdependent layers of annotations derived from various data types, like plain text or semi-structured text, and even images. We propose task modules to decouple the concerns of data representation and model-specific representations, thereby fostering greater flexibility and reusability of code. PyTorch-IE also extends support for widely used libraries such as PyTorch-Lightning for training, HuggingFace datasets for dataset reading, and Hydra for experiment configuration. Supplementary libraries and GitHub templates for the easy setup of new projects are also provided. By ensuring functionality and versatility, PyTorch-IE provides vital support to the research community engaged in Information Extraction.
Authors: Shinnosuke Tanaka, James Barry, Vishnudev Kuruvanthodi, Movina Moses, Maxwell J. Giammona, Nathan Herr, Mohab Elkaref, Geeth De Mel
Abstract: This paper describes the KnowledgeHub tool, a scientific literature Information Extraction (IE) and Question Answering (QA) pipeline. This is achieved by supporting the ingestion of PDF documents that are converted to text and structured representations. An ontology can then be constructed where a user defines the types of entities and relationships they want to capture. A browser-based annotation tool enables annotating the contents of the PDF documents according to the ontology. Named Entity Recognition (NER) and Relation Classification (RC) models can be trained on the resulting annotations and can be used to annotate the unannotated portion of the documents. A knowledge graph is constructed from these entity and relation triples which can be queried to obtain insights from the data. Furthermore, we integrate a suite of Large Language Models (LLMs) that can be used for QA and summarisation that is grounded in the included documents via a retrieval component. KnowledgeHub is a unique tool that supports annotation, IE and QA, which gives the user full insight into the knowledge discovery pipeline.
Authors: Kamalkumar Rathinasamy, Jayarama Nettar, Amit Kumar, Vishal Manchanda, Arun Vijayakumar, Ayush Kataria, Venkateshprasanna Manjunath, Chidambaram GS, Jaskirat Singh Sodhi, Shoeb Shaikh, Wasim Akhtar Khan, Prashant Singh, Tanishq Dattatray Ige, Vipin Tiwari, Rajab Ali Mondal, Harshini K, S Reka, Chetana Amancharla, Faiz ur Rahman, Harikrishnan P A, Indraneel Saha, Bhavya Tiwary, Navin Shankar Patel, Pradeep T S, Balaji A J, Priyapravas, Mohammed Rafee Tarafdar
Abstract: Enterprises grapple with the significant challenge of managing proprietary unstructured data, hindering efficient information retrieval. This has led to the emergence of AI-driven information retrieval solutions, designed to adeptly extract relevant insights to address employee inquiries. These solutions often leverage pre-trained embedding models and generative models as foundational components. While pre-trained embeddings may exhibit proximity or disparity based on their original training objectives, they might not fully align with the unique characteristics of enterprise-specific data, leading to suboptimal alignment with the retrieval goals of enterprise environments. In this paper, we propose a methodology to fine-tune pre-trained embedding models specifically for enterprise environments. By adapting the embeddings to better suit the retrieval tasks prevalent in enterprises, we aim to enhance the performance of information retrieval solutions. We discuss the process of fine-tuning, its effect on retrieval accuracy, and the potential benefits for enterprise information management. Our findings demonstrate the efficacy of fine-tuned embedding models in improving the precision and relevance of search results in enterprise settings.
Authors: Jayaprakash Sundararaj
Abstract: Automatic summarization is the process of reducing a text document in order to generate a summary that retains the most important points of the original document. In this work, we study two problems - i) summarizing a text document as set of keywords/caption, for image recommedation, ii) generating opinion summary which good mix of relevancy and sentiment with the text document. Intially, we present our work on an recommending images for enhancing a substantial amount of existing plain text news articles. We use probabilistic models and word similarity heuristics to generate captions and extract Key-phrases which are re-ranked using a rank aggregation framework with relevance feedback mechanism. We show that such rank aggregation and relevant feedback which are typically used in Tagging Documents, Text Information Retrieval also helps in improving image retrieval. These queries are fed to the Yahoo Search Engine to obtain relevant images 1. Our proposed method is observed to perform better than all existing baselines. Additonally, We propose a set of submodular functions for opinion summarization. Opinion summarization has built in it the tasks of summarization and sentiment detection. However, it is not easy to detect sentiment and simultaneously extract summary. The two tasks conflict in the sense that the demand of compression may drop sentiment bearing sentences, and the demand of sentiment detection may bring in redundant sentences. However, using submodularity we show how to strike a balance between the two requirements. Our functions generate summaries such that there is good correlation between document sentiment and summary sentiment along with good ROUGE score. We also compare the performances of the proposed submodular functions.
Authors: Hajung Kim, Chanhwi Kim, Hoonick Lee, Kyochul Jang, Jiwoo Lee, Kyungjae Lee, Gangwoo Kim, Jaewoo Kang
Abstract: Transforming natural language questions into SQL queries is crucial for precise data retrieval from electronic health record (EHR) databases. A significant challenge in this process is detecting and rejecting unanswerable questions that request information beyond the database's scope or exceed the system's capabilities. In this paper, we introduce a novel text-to-SQL framework that robustly handles out-of-domain questions and verifies the generated queries with query execution.Our framework begins by standardizing the structure of questions into a templated format. We use a powerful large language model (LLM), fine-tuned GPT-3.5 with detailed prompts involving the table schemas of the EHR database system. Our experimental results demonstrate the effectiveness of our framework on the EHRSQL-2024 benchmark benchmark, a shared task in the ClinicalNLP workshop. Although a straightforward fine-tuning of GPT shows promising results on the development set, it struggled with the out-of-domain questions in the test set. With our framework, we improve our system's adaptability and achieve competitive performances in the official leaderboard of the EHRSQL-2024 challenge.
Authors: Megan Flynn, Alexander Wang, Dean Edward Alvarez, Christopher De Sa, Anil Damle
Abstract: We present STAT: a simple algorithm to prune transformer models without any fine-tuning. STAT eliminates both attention heads and neurons from the network, while preserving accuracy by calculating a correction to the weights of the next layer. Each layer block in the network is compressed using a series of principled matrix factorizations that preserve the network structure. Our entire algorithm takes minutes to compress BERT, and less than three hours to compress models with 7B parameters using a single GPU. Using only several hundred data examples, STAT preserves the output of the network and improves upon existing gradient-free pruning methods. It is even competitive with methods that include significant fine-tuning. We demonstrate our method on both encoder and decoder architectures, including BERT, DistilBERT, and Llama-2 using benchmarks such as GLUE, Squad, WikiText2.
Authors: Jiaqi Xue, Mengxin Zheng, Yebowen Hu, Fei Liu, Xun Chen, Qian Lou
Abstract: Large Language Models (LLMs) are constrained by outdated information and a tendency to generate incorrect data, commonly referred to as "hallucinations." Retrieval-Augmented Generation (RAG) addresses these limitations by combining the strengths of retrieval-based methods and generative models. This approach involves retrieving relevant information from a large, up-to-date dataset and using it to enhance the generation process, leading to more accurate and contextually appropriate responses. Despite its benefits, RAG introduces a new attack surface for LLMs, particularly because RAG databases are often sourced from public data, such as the web. In this paper, we propose \TrojRAG{} to identify the vulnerabilities and attacks on retrieval parts (RAG database) and their indirect attacks on generative parts (LLMs). Specifically, we identify that poisoning several customized content passages could achieve a retrieval backdoor, where the retrieval works well for clean queries but always returns customized poisoned adversarial queries. Triggers and poisoned passages can be highly customized to implement various attacks. For example, a trigger could be a semantic group like "The Republican Party, Donald Trump, etc." Adversarial passages can be tailored to different contents, not only linked to the triggers but also used to indirectly attack generative LLMs without modifying them. These attacks can include denial-of-service attacks on RAG and semantic steering attacks on LLM generations conditioned by the triggers. Our experiments demonstrate that by just poisoning 10 adversarial passages can induce 98.2\% success rate to retrieve the adversarial passages. Then, these passages can increase the reject ratio of RAG-based GPT-4 from 0.01\% to 74.6\% or increase the rate of negative responses from 0.22\% to 72\% for targeted queries.
Authors: Gregory Schwartzman
Abstract: We report that ChatGPT 4 and 4o are susceptible to a prompt injection attack that allows an attacker to query users' personal data. It is applicable without the use of any 3rd party tools and all users are currently affected. This vulnerability is exacerbated by the recent introduction of ChatGPT's memory feature, which allows an attacker to command ChatGPT to monitor the user for the desired personal data.
Authors: Yifan Zeng, Ojas Tendolkar, Raymond Baartmans, Qingyun Wu, Huazheng Wang, Lizhong Chen
Abstract: Ranking passages by prompting a large language model (LLM) can achieve promising performance in modern information retrieval (IR) systems. A common approach is to sort the ranking list by prompting LLMs for pairwise comparison. However, sorting-based methods require consistent comparisons to correctly sort the passages, which we show that LLMs often violate. We identify two kinds of intrinsic inconsistency in LLM-based pairwise comparisons: order inconsistency which leads to conflicting results when switching the passage order, and transitive inconsistency which leads to non-transitive triads among all preference pairs. In this paper, we propose LLM-RankFusion, an LLM-based ranking framework that mitigates these inconsistencies and produces a robust ranking list. LLM-RankFusion mitigates order inconsistency using in-context learning (ICL) to demonstrate order-agnostic comparisons and calibration to estimate the underlying preference probability between two passages. We then address transitive inconsistency by aggregating the ranking results from multiple rankers. In our experiments, we empirically show that LLM-RankFusion can significantly reduce inconsistent pairwise comparison results, and improve the ranking quality by making the final ranking list more robust.
Authors: Frank Weizhen Liu, Chenhui Hu
Abstract: As Large Language Models (LLMs) increasingly become key components in various AI applications, understanding their security vulnerabilities and the effectiveness of defense mechanisms is crucial. This survey examines the security challenges of LLMs, focusing on two main areas: Prompt Hacking and Adversarial Attacks, each with specific types of threats. Under Prompt Hacking, we explore Prompt Injection and Jailbreaking Attacks, discussing how they work, their potential impacts, and ways to mitigate them. Similarly, we analyze Adversarial Attacks, breaking them down into Data Poisoning Attacks and Backdoor Attacks. This structured examination helps us understand the relationships between these vulnerabilities and the defense strategies that can be implemented. The survey highlights these security challenges and discusses robust defensive frameworks to protect LLMs against these threats. By detailing these security issues, the survey contributes to the broader discussion on creating resilient AI systems that can resist sophisticated attacks.
Authors: Bowen Jiang, Yangxinyu Xie, Xiaomeng Wang, Weijie J. Su, Camillo J. Taylor, Tanwi Mallick
Abstract: Rationality is the quality of being guided by reason, characterized by logical thinking and decision-making that align with evidence and logical rules. This quality is essential for effective problem-solving, as it ensures that solutions are well-founded and systematically derived. Despite the advancements of large language models (LLMs) in generating human-like text with remarkable accuracy, they present biases inherited from the training data, inconsistency across different contexts, and difficulty understanding complex scenarios involving multiple layers of context. Therefore, recent research attempts to leverage the strength of multiple agents working collaboratively with various types of data and tools for enhanced consistency and reliability. To that end, this paper aims to understand whether multi-modal and multi-agent systems are advancing toward rationality by surveying the state-of-the-art works, identifying advancements over single-agent and single-modal systems in terms of rationality, and discussing open problems and future directions. We maintain an open repository at https://github.com/bowen-upenn/MMMA_Rationality.
Authors: Yubo Wang, Hao Xin, Lei Chen
Abstract: The semantic annotation of tabular data plays a crucial role in various downstream tasks. Previous research has proposed knowledge graph (KG)-based and deep learning-based methods, each with its inherent limitations. KG-based methods encounter difficulties annotating columns when there is no match for column cells in the KG. Moreover, KG-based methods can provide multiple predictions for one column, making it challenging to determine the semantic type with the most suitable granularity for the dataset. This type granularity issue limits their scalability. On the other hand, deep learning-based methods face challenges related to the valuable context missing issue. This occurs when the information within the table is insufficient for determining the correct column type. This paper presents KGLink, a method that combines WikiData KG information with a pre-trained deep learning language model for table column annotation, effectively addressing both type granularity and valuable context missing issues. Through comprehensive experiments on widely used tabular datasets encompassing numeric and string columns with varying type granularity, we showcase the effectiveness and efficiency of KGLink. By leveraging the strengths of KGLink, we successfully surmount challenges related to type granularity and valuable context issues, establishing it as a robust solution for the semantic annotation of tabular data.
Authors: Xiaoqian Liu, Guoqiang Hu, Yangfan Du, Erfeng He, YingFeng Luo, Chen Xu, Tong Xiao, Jingbo Zhu
Abstract: Simultaneous speech translation (SimulST) is a demanding task that involves generating translations in real-time while continuously processing speech input. This paper offers a comprehensive overview of the recent developments in SimulST research, focusing on four major challenges. Firstly, the complexities associated with processing lengthy and continuous speech streams pose significant hurdles. Secondly, satisfying real-time requirements presents inherent difficulties due to the need for immediate translation output. Thirdly, striking a balance between translation quality and latency constraints remains a critical challenge. Finally, the scarcity of annotated data adds another layer of complexity to the task. Through our exploration of these challenges and the proposed solutions, we aim to provide valuable insights into the current landscape of SimulST research and suggest promising directions for future exploration.
Authors: Christiane Rakotomalala, Jean-Marie Paillat, Fr\'ed\'eric Feder, Angel Avad\'i, Laurent Thuri\`es, Marie-Liesse Vermeire, Jean-Michel M\'edoc, Tom Wassenaar, Caroline Hottelart, Lilou Kieffer, Elisa Ndjie, Mathieu Picart, Jorel Tchamgoue, Alvin Tulle, Laurine Valade, Annie Boyer, Marie-Christine Duchamp, Mathieu Roche
Abstract: The text mining method presented in this paper was used for annotation of terms related to biological transformation and valorization of organic residues in agriculture in low and middle-income country. Specialized lexicon was obtained through different steps: corpus and extraction of terms, annotation of extracted terms, selection of relevant terms.
Authors: Haisu Guan, Huanxin Yang, Xinyu Wang, Shengwei Han, Yongge Liu, Lianwen Jin, Xiang Bai, Yuliang Liu
Abstract: Originating from China's Shang Dynasty approximately 3,000 years ago, the Oracle Bone Script (OBS) is a cornerstone in the annals of linguistic history, predating many established writing systems. Despite the discovery of thousands of inscriptions, a vast expanse of OBS remains undeciphered, casting a veil of mystery over this ancient language. The emergence of modern AI technologies presents a novel frontier for OBS decipherment, challenging traditional NLP methods that rely heavily on large textual corpora, a luxury not afforded by historical languages. This paper introduces a novel approach by adopting image generation techniques, specifically through the development of Oracle Bone Script Decipher (OBSD). Utilizing a conditional diffusion-based strategy, OBSD generates vital clues for decipherment, charting a new course for AI-assisted analysis of ancient languages. To validate its efficacy, extensive experiments were conducted on an oracle bone script dataset, with quantitative results demonstrating the effectiveness of OBSD. Code and decipherment results will be made available at https://github.com/guanhaisu/OBSD.
Authors: Sahar Abdelnabi, Aideen Fay, Giovanni Cherubin, Ahmed Salem, Mario Fritz, Andrew Paverd
Abstract: Large Language Models (LLMs) are routinely used in retrieval-augmented applications to orchestrate tasks and process inputs from users and other sources. These inputs, even in a single LLM interaction, can come from a variety of sources, of varying trustworthiness and provenance. This opens the door to prompt injection attacks, where the LLM receives and acts upon instructions from supposedly data-only sources, thus deviating from the user's original instructions. We define this as task drift, and we propose to catch it by scanning and analyzing the LLM's activations. We compare the LLM's activations before and after processing the external input in order to detect whether this input caused instruction drift. We develop two probing methods and find that simply using a linear classifier can detect drift with near perfect ROC AUC on an out-of-distribution test set. We show that this approach generalizes surprisingly well to unseen task domains, such as prompt injections, jailbreaks, and malicious instructions, without being trained on any of these attacks. Our setup does not require any modification of the LLM (e.g., fine-tuning) or any text generation, thus maximizing deployability and cost efficiency and avoiding reliance on unreliable model output. To foster future research on activation-based task inspection, decoding, and interpretability, we will release our large-scale TaskTracker toolkit, comprising a dataset of over 500K instances, representations from 4 SoTA language models, and inspection tools.
Authors: Nicholas Roberts, Samuel Guo, Zhiqi Gao, Satya Sai Srinath Namburi GNVV, Sonia Cromp, Chengjun Wu, Chengyu Duan, Frederic Sala
Abstract: While Transformers underpin modern large language models (LMs), there is a growing list of alternative architectures with new capabilities, promises, and tradeoffs. This makes choosing the right LM architecture challenging. Recently-proposed $\textit{hybrid architectures}$ seek a best-of-all-worlds approach that reaps the benefits of all architectures. Hybrid design is difficult for two reasons: it requires manual expert-driven search, and new hybrids must be trained from scratch. We propose $\textbf{Manticore}$, a framework that addresses these challenges. Manticore $\textit{automates the design of hybrid architectures}$ while reusing pretrained models to create $\textit{pretrained}$ hybrids. Our approach augments ideas from differentiable Neural Architecture Search (NAS) by incorporating simple projectors that translate features between pretrained blocks from different architectures. We then fine-tune hybrids that combine pretrained models from different architecture families -- such as the GPT series and Mamba -- end-to-end. With Manticore, we enable LM selection without training multiple models, the construction of pretrained hybrids from existing pretrained models, and the ability to $\textit{program}$ pretrained hybrids to have certain capabilities. Manticore hybrids outperform existing manually-designed hybrids, achieve strong performance on Long Range Arena (LRA) tasks, and can improve on pretrained transformers and state space models.
Authors: Erfan Loweimi, Andrea Carmantini, Peter Bell, Steve Renals, Zoran Cvetkovic
Abstract: In this paper, we analyse the error patterns of the raw waveform acoustic models in TIMIT's phone recognition task. Our analysis goes beyond the conventional phone error rate (PER) metric. We categorise the phones into three groups: {affricate, diphthong, fricative, nasal, plosive, semi-vowel, vowel, silence}, {consonant, vowel+, silence}, and {voiced, unvoiced, silence} and, compute the PER for each broad phonetic class in each category. We also construct a confusion matrix for each category using the substitution errors and compare the confusion patterns with those of the Filterbank and Wav2vec 2.0 systems. Our raw waveform acoustic models consists of parametric (Sinc2Net) or non-parametric CNNs and Bidirectional LSTMs, achieving down to 13.7%/15.2% PERs on TIMIT Dev/Test sets, outperforming reported PERs for raw waveform models in the literature. We also investigate the impact of transfer learning from WSJ on the phonetic error patterns and confusion matrices. It reduces the PER to 11.8%/13.7% on the Dev/Test sets.
Authors: Weijun Li, Qiongkai Xu, Mark Dras
Abstract: Recent studies have shown that distributed machine learning is vulnerable to gradient inversion attacks, where private training data can be reconstructed by analyzing the gradients of the models shared in training. Previous attacks established that such reconstructions are possible using gradients from all parameters in the entire models. However, we hypothesize that most of the involved modules, or even their sub-modules, are at risk of training data leakage, and we validate such vulnerabilities in various intermediate layers of language models. Our extensive experiments reveal that gradients from a single Transformer layer, or even a single linear component with 0.54% parameters, are susceptible to training data leakage. Additionally, we show that applying differential privacy on gradients during training offers limited protection against the novel vulnerability of data disclosure.
Authors: Ahmed M. Ahmed, Rafael Rafailov, Stepan Sharkov, Xuechen Li, Sanmi Koyejo
Abstract: Reinforcement Learning from Human Feedback (RLHF) has enabled significant advancements within language modeling for powerful, instruction-following models. However, the alignment of these models remains a pressing challenge as the policy tends to overfit the learned ``proxy" reward model past an inflection point of utility as measured by a ``gold" reward model that is more performant -- a phenomenon known as \textit{over-optimization}. Prior work has mitigated this issue by computing a pessimistic statistic over an ensemble of reward models, which is common in Offline Reinforcement Learning but incredibly costly for language models with high memory requirements, making such approaches infeasible for sufficiently large models. To this end, we propose using a shared encoder but separate linear heads. We find this leads to similar performance as the full ensemble while allowing tremendous savings in memory and time required for training for models of similar size. \end{abstract}
Authors: Tzu-Lin Kuo, Tzu-Wei Chiu, Tzung-Sheng Lin, Sheng-Yang Wu, Chao-Wei Huang, Yun-Nung Chen
Abstract: Generative Retrieval (GR) is an emerging paradigm in information retrieval that leverages generative models to directly map queries to relevant document identifiers (DocIDs) without the need for traditional query processing or document reranking. This survey provides a comprehensive overview of GR, highlighting key developments, indexing and retrieval strategies, and challenges. We discuss various document identifier strategies, including numerical and string-based identifiers, and explore different document representation methods. Our primary contribution lies in outlining future research directions that could profoundly impact the field: improving the quality of query generation, exploring learnable document identifiers, enhancing scalability, and integrating GR with multi-task learning frameworks. By examining state-of-the-art GR techniques and their applications, this survey aims to provide a foundational understanding of GR and inspire further innovations in this transformative approach to information retrieval. We also make the complementary materials such as paper collection publicly available at https://github.com/MiuLab/GenIR-Survey/
Authors: Viktor Shcherbakov, Fedor Krasnov
Abstract: Product search is uniquely different from search for documents, Internet resources or vacancies, therefore it requires the development of specialized search systems. The present work describes the H1 embdedding model, designed for an offline term indexing of product descriptions at e-commerce platforms. The model is compared to other state-of-the-art (SoTA) embedding models within a framework of hybrid product search system that incorporates the advantages of lexical methods for product retrieval and semantic embedding-based methods. We propose an approach to building semantically rich term vocabularies for search indexes. Compared to other production semantic models, H1 paired with the proposed approach stands out due to its ability to process multi-word product terms as one token. As an example, for search queries "new balance shoes", "gloria jeans kids wear" brand entity will be represented as one token - "new balance", "gloria jeans". This results in an increased precision of the system without affecting the recall. The hybrid search system with proposed model scores mAP@12 = 56.1% and R@1k = 86.6% on the WANDS public dataset, beating other SoTA analogues.
Authors: Diego Dorn, Alexandre Variengien, Charbel-Rapha\"el Segerie, Vincent Corruble
Abstract: Input-output safeguards are used to detect anomalies in the traces produced by Large Language Models (LLMs) systems. These detectors are at the core of diverse safety-critical applications such as real-time monitoring, offline evaluation of traces, and content moderation. However, there is no widely recognized methodology to evaluate them. To fill this gap, we introduce the Benchmarks for the Evaluation of LLM Safeguards (BELLS), a structured collection of tests, organized into three categories: (1) established failure tests, based on already-existing benchmarks for well-defined failure modes, aiming to compare the performance of current input-output safeguards; (2) emerging failure tests, to measure generalization to never-seen-before failure modes and encourage the development of more general safeguards; (3) next-gen architecture tests, for more complex scaffolding (such as LLM-agents and multi-agent systems), aiming to foster the development of safeguards that could adapt to future applications for which no safeguard currently exists. Furthermore, we implement and share the first next-gen architecture test, using the MACHIAVELLI environment, along with an interactive visualization of the dataset.
Authors: Yingwei Ma, Qingping Yang, Rongyu Cao, Binhua Li, Fei Huang, Yongbin Li
Abstract: Recently, Large Language Model (LLM) based agents have advanced the significant development of Automatic Software Engineering (ASE). Although verified effectiveness, the designs of the existing methods mainly focus on the local information of codes, e.g., issues, classes, and functions, leading to limitations in capturing the global context and interdependencies within the software system. From the practical experiences of the human SE developers, we argue that an excellent understanding of the whole repository will be the critical path to ASE. However, understanding the whole repository raises various challenges, e.g., the extremely long code input, the noisy code information, the complex dependency relationships, etc. To this end, we develop a novel ASE method named RepoUnderstander by guiding agents to comprehensively understand the whole repositories. Specifically, we first condense the critical information of the whole repository into the repository knowledge graph in a top-to-down mode to decrease the complexity of repository. Subsequently, we empower the agents the ability of understanding whole repository by proposing a Monte Carlo tree search based repository exploration strategy. In addition, to better utilize the repository-level knowledge, we guide the agents to summarize, analyze, and plan. Then, they can manipulate the tools to dynamically acquire information and generate the patches to solve the real-world GitHub issues. Extensive experiments demonstrate the superiority and effectiveness of the proposed RepoUnderstander. It achieved 18.5\% relative improvement on the SWE-bench Lite benchmark compared to SWE-agent.
Authors: Aleksandar Petrov, Tom A. Lamb, Alasdair Paren, Philip H. S. Torr, Adel Bibi
Abstract: Zero-shot and in-context learning enable solving tasks without model fine-tuning, making them essential for developing generative model solutions. Therefore, it is crucial to understand whether a pretrained model can be prompted to approximate any function, i.e., whether it is a universal in-context approximator. While it was recently shown that transformer models do possess this property, these results rely on their attention mechanism. Hence, these findings do not apply to fully recurrent architectures like RNNs, LSTMs, and the increasingly popular SSMs. We demonstrate that RNNs, LSTMs, GRUs, Linear RNNs, and linear gated architectures such as Mamba and Hawk/Griffin can also serve as universal in-context approximators. To streamline our argument, we introduce a programming language called LSRL that compiles to these fully recurrent architectures. LSRL may be of independent interest for further studies of fully recurrent models, such as constructing interpretability benchmarks. We also study the role of multiplicative gating and observe that architectures incorporating such gating (e.g., LSTMs, GRUs, Hawk/Griffin) can implement certain operations more stably, making them more viable candidates for practical in-context universal approximation.
Authors: Wonkee Lee, Seong-Hwan Heo, Jong-Hyeok Lee
Abstract: Semi-supervised learning that leverages synthetic data for training has been widely adopted for developing automatic post-editing (APE) models due to the lack of training data. With this aim, we focus on data-synthesis methods to create high-quality synthetic data. Given that APE takes as input a machine-translation result that might include errors, we present a data-synthesis method by which the resulting synthetic data mimic the translation errors found in actual data. We introduce a noising-based data-synthesis method by adapting the masked language model approach, generating a noisy text from a clean text by infilling masked tokens with erroneous tokens. Moreover, we propose selective corpus interleaving that combines two separate synthetic datasets by taking only the advantageous samples to enhance the quality of the synthetic data further. Experimental results show that using the synthetic data created by our approach results in significantly better APE performance than other synthetic data created by existing methods.
Authors: Kirk Baker
Abstract: Stemming is the process of reducing related words to a standard form by removing affixes from them. Existing algorithms vary with respect to their complexity, configurability, handling of unknown words, and ability to avoid under- and over-stemming. This paper presents a fast, simple, configurable, high-precision, high-recall stemming algorithm that combines the simplicity and performance of word-based lookup tables with the strong generalizability of rule-based methods to avert problems with out-of-vocabulary words.
Authors: Bang Yang, Fenglin Liu, Yuexian Zou, Xian Wu, Yaowei Wang, David A. Clifton
Abstract: Natural Language Generation (NLG) accepts input data in the form of images, videos, or text and generates corresponding natural language text as output. Existing NLG methods mainly adopt a supervised approach and rely heavily on coupled data-to-text pairs. However, for many targeted scenarios and for non-English languages, sufficient quantities of labeled data are often not available. To relax the dependency on labeled data of downstream tasks, we propose an intuitive and effective zero-shot learning framework, ZeroNLG, which can deal with multiple NLG tasks, including image-to-text (image captioning), video-to-text (video captioning), and text-to-text (neural machine translation), across English, Chinese, German, and French within a unified framework. ZeroNLG does not require any labeled downstream pairs for training. During training, ZeroNLG (i) projects different domains (across modalities and languages) to corresponding coordinates in a shared common latent space; (ii) bridges different domains by aligning their corresponding coordinates in this space; and (iii) builds an unsupervised multilingual auto-encoder to learn to generate text by reconstructing the input text given its coordinate in shared latent space. Consequently, during inference, based on the data-to-text pipeline, ZeroNLG can generate target sentences across different languages given the coordinate of input data in the common space. Within this unified framework, given visual (imaging or video) data as input, ZeroNLG can perform zero-shot visual captioning; given textual sentences as input, ZeroNLG can perform zero-shot machine translation. We present the results of extensive experiments on twelve NLG tasks, showing that, without using any labeled downstream pairs for training, ZeroNLG generates high-quality and believable outputs and significantly outperforms existing zero-shot methods.
Authors: Zeyuan Allen-Zhu, Yuanzhi Li
Abstract: Transformer-based language models are effective but complex, and understanding their inner workings is a significant challenge. Previous research has primarily explored how these models handle simple tasks like name copying or selection, and we extend this by investigating how these models grasp complex, recursive language structures defined by context-free grammars (CFGs). We introduce a family of synthetic CFGs that produce hierarchical rules, capable of generating lengthy sentences (e.g., hundreds of tokens) that are locally ambiguous and require dynamic programming to parse. Despite this complexity, we demonstrate that generative models like GPT can accurately learn this CFG language and generate sentences based on it. We explore the model's internals, revealing that its hidden states precisely capture the structure of CFGs, and its attention patterns resemble the information passing in a dynamic programming algorithm. This paper also presents several corollaries, including showing why positional embedding is inferior to relative attention or rotary embedding; demonstrating that encoder-based models (e.g., BERT, deBERTa) cannot learn very deeply nested CFGs as effectively as generative models (e.g., GPT); and highlighting the necessity of adding structural and syntactic errors to the pretraining data to make the model more robust to corrupted language prefixes.
Authors: Erica Cai, Brendan O'Connor
Abstract: Current social science efforts automatically populate event databases of "who did what to whom?" tuples, by applying event extraction (EE) to text such as news. The event databases are used to analyze sociopolitical dynamics between actor pairs (dyads) in, e.g., international relations. While most EE methods heavily rely on rules or supervised learning, \emph{zero-shot} event extraction could potentially allow researchers to flexibly specify arbitrary event classes for new research questions. Unfortunately, we find that current zero-shot EE methods, as well as a naive zero-shot approach of simple generative language model (LM) prompting, perform poorly for dyadic event extraction; most suffer from word sense ambiguity, modality sensitivity, and computational inefficiency. We address these challenges with a new fine-grained, multi-stage instruction-following generative LM pipeline, proposing a Monte Carlo approach to deal with, and even take advantage of, nondeterminism of generative outputs. Our pipeline includes explicit stages of linguistic analysis (synonym generation, contextual disambiguation, argument realization, event modality), \textit{improving control and interpretability} compared to purely neural methods. This method outperforms other zero-shot EE approaches, and outperforms naive applications of generative LMs by at least 17 F1 percent points. The pipeline's filtering mechanism greatly improves computational efficiency, allowing it to perform as few as 12% of queries that a previous zero-shot method uses. Finally, we demonstrate our pipeline's application to dyadic international relations analysis.
Authors: Jianing Wang, Qiushi Sun, Xiang Li, Ming Gao
Abstract: Recently, Chain-of-Thought (CoT) prompting has delivered success on complex reasoning tasks, which aims at designing a simple prompt like ``Let's think step by step'' or multiple in-context exemplars with well-designed rationales to elicit Large Language Models (LLMs) to generate intermediate reasoning steps. However, the generated rationales often come with mistakes, making unfactual and unfaithful reasoning chains. To mitigate this brittleness, we propose a novel Chain-of-Knowledge (CoK) prompting, where we aim at eliciting LLMs to generate explicit pieces of knowledge evidence in the form of structure triple. This is inspired by our human behaviors, i.e., we can draw a mind map or knowledge map as the reasoning evidence in the brain before answering a complex question. Benefiting from CoK, we additionally introduce a F^2-Verification method to estimate the reliability of the reasoning chains in terms of factuality and faithfulness. For the unreliable response, the wrong evidence can be indicated to prompt the LLM to rethink. Extensive experiments demonstrate that our method can further improve the performance of commonsense, factual, symbolic, and arithmetic reasoning tasks.
Authors: Victor Quach, Adam Fisch, Tal Schuster, Adam Yala, Jae Ho Sohn, Tommi S. Jaakkola, Regina Barzilay
Abstract: We propose a novel approach to conformal prediction for generative language models (LMs). Standard conformal prediction produces prediction sets -- in place of single predictions -- that have rigorous, statistical performance guarantees. LM responses are typically sampled from the model's predicted distribution over the large, combinatorial output space of natural language. Translating this process to conformal prediction, we calibrate a stopping rule for sampling different outputs from the LM that get added to a growing set of candidates until we are confident that the output set is sufficient. Since some samples may be low-quality, we also simultaneously calibrate and apply a rejection rule for removing candidates from the output set to reduce noise. Similar to conformal prediction, we prove that the sampled set returned by our procedure contains at least one acceptable answer with high probability, while still being empirically precise (i.e., small) on average. Furthermore, within this set of candidate responses, we show that we can also accurately identify subsets of individual components -- such as phrases or sentences -- that are each independently correct (e.g., that are not "hallucinations"), again with statistical guarantees. We demonstrate the promise of our approach on multiple tasks in open-domain question answering, text summarization, and radiology report generation using different LM variants.
Authors: Ran Zhang, Jihed Ouni, Steffen Eger
Abstract: While summarization has been extensively researched in natural language processing (NLP), cross-lingual cross-temporal summarization (CLCTS) is a largely unexplored area that has the potential to improve cross-cultural accessibility and understanding. This paper comprehensively addresses the CLCTS task, including dataset creation, modeling, and evaluation. We (1) build the first CLCTS corpus with 328 instances for hDe-En (extended version with 455 instances) and 289 for hEn-De (extended version with 501 instances), leveraging historical fiction texts and Wikipedia summaries in English and German; (2) examine the effectiveness of popular transformer end-to-end models with different intermediate finetuning tasks; (3) explore the potential of GPT-3.5 as a summarizer; (4) report evaluations from humans, GPT-4, and several recent automatic evaluation metrics. Our results indicate that intermediate task finetuned end-to-end models generate bad to moderate quality summaries while GPT-3.5, as a zero-shot summarizer, provides moderate to good quality outputs. GPT-3.5 also seems very adept at normalizing historical text. To assess data contamination in GPT-3.5, we design an adversarial attack scheme in which we find that GPT-3.5 performs slightly worse for unseen source documents compared to seen documents. Moreover, it sometimes hallucinates when the source sentences are inverted against its prior knowledge with a summarization accuracy of 0.67 for plot omission, 0.71 for entity swap, and 0.53 for plot negation. Overall, our regression results of model performances suggest that longer, older, and more complex source texts (all of which are more characteristic for historical language variants) are harder to summarize for all models, indicating the difficulty of the CLCTS task.
Authors: Yiming Wang, Zhuosheng Zhang, Pei Zhang, Baosong Yang, Rui Wang
Abstract: Neural-symbolic methods have demonstrated efficiency in enhancing the reasoning abilities of large language models (LLMs). However, existing methods mainly rely on syntactically mapping natural languages to complete formal languages like Python and SQL. Those methods require that reasoning tasks be convertible into programs, which cater to the computer execution mindset and deviate from human reasoning habits. To broaden symbolic methods' applicability and adaptability in the real world, we propose the Meta-Reasoning from a linguistic perspective. This method empowers LLMs to deconstruct reasoning-independent semantic information into generic symbolic representations, thereby efficiently capturing more generalized reasoning knowledge. We conduct extensive experiments on more than ten datasets encompassing conventional reasoning tasks like arithmetic, symbolic, and logical reasoning, and the more complex interactive reasoning tasks like theory-of-mind reasoning. Experimental results demonstrate that Meta-Reasoning significantly enhances in-context reasoning accuracy, learning efficiency, out-of-domain generalization, and output stability compared to the Chain-of-Thought technique. Code and data are publicly available at \url{https://github.com/Alsace08/Meta-Reasoning}.
Authors: Lukas Birkenmaier, Claudia Wagner, Clemens Lechner
Abstract: Guidance on how to validate computational text-based measures of social constructs is fragmented. While researchers generally acknowledge the importance of validating text-based measures, they often lack a shared vocabulary and a unified framework to do so. This paper introduces ValiText, a new validation framework designed to assist scholars in validly measuring social constructs in textual data. The framework is built on a conceptual foundation of validity in the social sciences, strengthened by an empirical review of validation practices in the social sciences and consultations with experts. Ultimately, ValiText prescribes researchers to demonstrate three types of validation evidence: substantive evidence (outlining the theoretical underpinning of the measure), structural evidence (examining the properties of the text model and its output) and external evidence (testing for how the measure relates to independent information). The framework is further supplemented by a checklist of validation steps, offering practical guidance in the form of documentation sheets that guide researchers in the validation process.
Authors: Zehui Wu, Ziwei Gong, Jaywon Koo, Julia Hirschberg
Abstract: This paper investigates the optimal selection and fusion of feature encoders across multiple modalities and combines these in one neural network to improve sentiment detection. We compare different fusion methods and examine the impact of multi-loss training within the multi-modality fusion network, identifying surprisingly important findings relating to subnet performance. We have also found that integrating context significantly enhances model performance. Our best model achieves state-of-the-art performance for three datasets (CMU-MOSI, CMU-MOSEI and CH-SIMS). These results suggest a roadmap toward an optimized feature selection and fusion approach for enhancing sentiment detection in neural networks.
Authors: Evan Crothers, Herna Viktor, Nathalie Japkowicz
Abstract: A common approach to quantifying neural text classifier interpretability is to calculate faithfulness metrics based on iteratively masking salient input tokens and measuring changes in the model prediction. We propose that this property is better described as "sensitivity to iterative masking", and highlight pitfalls in using this measure for comparing text classifier interpretability. We show that iterative masking produces large variation in faithfulness scores between otherwise comparable Transformer encoder text classifiers. We then demonstrate that iteratively masked samples produce embeddings outside the distribution seen during training, resulting in unpredictable behaviour. We further explore task-specific considerations that undermine principled comparison of interpretability using iterative masking, such as an underlying similarity to salience-based adversarial attacks. Our findings give insight into how these behaviours affect neural text classifiers, and provide guidance on how sensitivity to iterative masking should be interpreted.
Authors: Anna Rogers, Alexandra Sasha Luccioni
Abstract: Much of the recent discourse within the ML community has been centered around Large Language Models (LLMs), their functionality and potential -- yet not only do we not have a working definition of LLMs, but much of this discourse relies on claims and assumptions that are worth re-examining. We contribute a definition of LLMs, critically examine five common claims regarding their properties (including 'emergent properties'), and conclude with suggestions for future research directions and their framing.
Authors: Hoyoun Jung, Kyung-Joong Kim
Abstract: Compressed prompts aid instruction-tuned language models (LMs) in overcoming context window limitations and reducing computational costs. Existing methods, which primarily based on training embeddings, face various challenges associated with interpretability, the fixed number of embedding tokens, reusability across different LMs, and inapplicability when interacting with black-box APIs. This study proposes prompt compression with reinforcement learning (PCRL), which is a discrete prompt compression method that addresses these issues. The proposed PCRL method utilizes a computationally efficient policy network that edits prompts directly. The training approach employed in the proposed PCRLs can be applied flexibly to various types of LMs, including both decoder-only and encoder-decoder architecture and it can be trained without gradient access to the LMs or labeled data. The proposed PCRL achieves an average reduction of 24.6% in terms of the token count across various instruction prompts while maintaining sufficient performance. In addition, we demonstrate that the learned policy can be transferred to larger LMs, and through a comprehensive analysis, we explore the token importance within the prompts. Our code is accessible at https://github.com/nenomigami/PromptCompressor.
Authors: Bilgehan Sel, Ahmad Al-Tawaha, Vanshaj Khattar, Ruoxi Jia, Ming Jin
Abstract: Current literature, aiming to surpass the "Chain-of-Thought" approach, often resorts to external modi operandi involving halting, modifying, and then resuming the generation process to boost Large Language Models' (LLMs) reasoning capacities. Due to their myopic perspective, they escalate the number of query requests, leading to increased costs, memory, and computational overheads. Addressing this, we propose the Algorithm of Thoughts -- a novel strategy that propels LLMs through algorithmic reasoning pathways. By employing algorithmic examples fully in-context, this overarching view of the whole process exploits the innate recurrence dynamics of LLMs, expanding their idea exploration with merely one or a few queries. Our technique outperforms earlier single-query methods and even more recent multi-query strategies that employ an extensive tree search algorithms while using significantly fewer tokens. Intriguingly, our results suggest that instructing an LLM using an algorithm can lead to performance surpassing that of the algorithm itself, hinting at LLM's inherent ability to weave its intuition into optimized searches. We probe into the underpinnings of our method's efficacy and its nuances in application. The code and related content can be found in: https://algorithm-of-thoughts.github.io.
Authors: Huayang Li, Siheng Li, Deng Cai, Longyue Wang, Lemao Liu, Taro Watanabe, Yujiu Yang, Shuming Shi
Abstract: Large language models with instruction-following abilities have revolutionized the field of artificial intelligence. These models show exceptional generalizability to tackle various real-world tasks through their natural language interfaces. However, their performance heavily relies on high-quality exemplar data, which is often difficult to obtain. This challenge is further exacerbated when it comes to multimodal instruction following. We introduce TextBind, an almost annotation-free framework for empowering larger language models with the multi-turn interleaved multimodal instruction-following capabilities. Our approach requires only image-caption pairs and generates multi-turn multimodal instruction-response conversations from a language model. To accommodate interleaved image-text inputs and outputs, we devise MIM, a language model-centric architecture that seamlessly integrates image encoder and decoder models. We release our dataset, model, and demo to foster future research in the area of multimodal instruction following.
Authors: Lingfeng Shen, Aayush Mishra, Daniel Khashabi
Abstract: The emergence of In-Context Learning (ICL) in LLMs remains a remarkable phenomenon that is partially understood. To explain ICL, recent studies have created theoretical connections to Gradient Descent (GD). We ask, do such connections hold up in actual pre-trained language models? We highlight the limiting assumptions in prior works that make their setup considerably different from the practical setup in which language models are trained. For example, their experimental verification uses \emph{ICL objective} (training models explicitly for ICL), which differs from the emergent ICL in the wild. Furthermore, the theoretical hand-constructed weights used in these studies have properties that don't match those of real LLMs. We also look for evidence in real models. We observe that ICL and GD have different sensitivity to the order in which they observe demonstrations. Finally, we probe and compare the ICL vs. GD hypothesis in a natural setting. We conduct comprehensive empirical analyses on language models pre-trained on natural data (LLaMa-7B). Our comparisons of three performance metrics highlight the inconsistent behavior of ICL and GD as a function of various factors such as datasets, models, and the number of demonstrations. We observe that ICL and GD modify the output distribution of language models differently. These results indicate that \emph{the equivalence between ICL and GD remains an open hypothesis} and calls for further studies.
Authors: Xiaodong Yu, Hao Cheng, Xiaodong Liu, Dan Roth, Jianfeng Gao
Abstract: Despite remarkable advancements in mitigating hallucinations in large language models (LLMs) by retrieval augmentation, it remains challenging to measure the reliability of LLMs using static question-answering (QA) data. Specifically, given the potential of data contamination (e.g., leading to memorization), good static benchmark performance does not ensure that model can reliably use the provided evidence for responding, which is essential to avoid hallucination when the required knowledge is new or private. Inspired by adversarial machine learning, we investigate the feasibility of automatically perturbing existing static one for dynamic evaluation. Specifically, this paper presents ReEval, an LLM-based framework using prompt chaining to perturb the original evidence for generating new test cases for evaluating the LLMs' reliability in using new evidence for answering. We implement ReEval using ChatGPT and evaluate the resulting variants of two popular open-domain QA datasets on a collection of LLMs under various prompting settings. Our generated data is human-readable and useful to trigger hallucination in LLM. Accurate models on static data are observed to produce unsupported answers from the perturbed evidence, with pronounced accuracy drops across LLMs including GPT-4. We find that our adversarial examples are transferable across all considered LLMs. The examples generated by a small model can be used to evaluate a much larger model, making our approach cost-effective.
Authors: Yihao Wang, Ruiqi Song, Lingxiao Li, Yifan Tang, Ru Zhang, Jianyi Liu
Abstract: Linguistic steganalysis (LS) tasks aim to detect whether a text contains secret information. Existing LS methods focus on the deep-learning model design and they achieve excellent results in ideal data. However, they overlook the unique user characteristics, leading to weak performance in social networks. And a few stegos here that further complicate detection. We propose the UP4LS, a framework with the User Profile for enhancing LS in realistic scenarios. Three kinds of user attributes like writing habits are explored to build the profile. For each attribute, the specific feature extraction module is designed. The extracted features are mapped to high-dimensional user features via the deep-learning model of the method to be improved. The content feature is extracted by the language model. Then user and content features are integrated. Existing methods can improve LS results by adding the UP4LS framework without changing their deep-learning models. Experiments show that UP4LS can significantly enhance the performance of LS-task baselines in realistic scenarios, with the overall Acc increased by 25%, F1 increased by 51%, and SOTA results. The improvement is especially pronounced in fewer stegos. Additionally, UP4LS also sets the stage for the related-task SOTA methods to efficient LS.
Authors: Haoran Li, Dadi Guo, Donghao Li, Wei Fan, Qi Hu, Xin Liu, Chunkit Chan, Duanyi Yao, Yuan Yao, Yangqiu Song
Abstract: The rapid development of language models (LMs) brings unprecedented accessibility and usage for both models and users. On the one hand, powerful LMs achieve state-of-the-art performance over numerous downstream NLP tasks. On the other hand, more and more attention is paid to unrestricted model accesses that may bring malicious privacy risks of data leakage. To address these issues, many recent works propose privacy-preserving language models (PPLMs) with differential privacy (DP). Unfortunately, different DP implementations make it challenging for a fair comparison among existing PPLMs. In this paper, we present PrivLM-Bench, a multi-perspective privacy evaluation benchmark to empirically and intuitively quantify the privacy leakage of LMs. Instead of only reporting DP parameters, PrivLM-Bench sheds light on the neglected inference data privacy during actual usage. PrivLM-Bench first clearly defines multi-faceted privacy objectives. Then, PrivLM-Bench constructs a unified pipeline to perform private fine-tuning. Lastly, PrivLM-Bench performs existing privacy attacks on LMs with pre-defined privacy objectives as the empirical evaluation results. The empirical attack results are used to fairly and intuitively evaluate the privacy leakage of various PPLMs. We conduct extensive experiments on three datasets of GLUE for mainstream LMs.
Authors: Letitia Parcalabescu, Anette Frank
Abstract: Large language models (LLMs) can explain their predictions through post-hoc or Chain-of-Thought (CoT) explanations. But an LLM could make up reasonably sounding explanations that are unfaithful to its underlying reasoning. Recent work has designed tests that aim to judge the faithfulness of post-hoc or CoT explanations. In this work we argue that these faithfulness tests do not measure faithfulness to the models' inner workings -- but rather their self-consistency at output level. Our contributions are three-fold: i) We clarify the status of faithfulness tests in view of model explainability, characterising them as self-consistency tests instead. This assessment we underline by ii) constructing a Comparative Consistency Bank for self-consistency tests that for the first time compares existing tests on a common suite of 11 open LLMs and 5 tasks -- including iii) our new self-consistency measure CC-SHAP. CC-SHAP is a fine-grained measure (not a test) of LLM self-consistency. It compares how a model's input contributes to the predicted answer and to generating the explanation. Our fine-grained CC-SHAP metric allows us iii) to compare LLM behaviour when making predictions and to analyse the effect of other consistency tests at a deeper level, which takes us one step further towards measuring faithfulness by bringing us closer to the internals of the model than strictly surface output-oriented tests. Our code is available at \url{https://github.com/Heidelberg-NLP/CC-SHAP}
Authors: Pengyu Cheng, Yifan Yang, Jian Li, Yong Dai, Tianhao Hu, Peixin Cao, Nan Du, Xiaolong Li
Abstract: Human preference alignment is essential to improve the interaction quality of large language models (LLMs). Existing alignment methods depend on manually annotated preference data to guide the LLM optimization directions. However, continuously updating LLMs for alignment raises a distribution gap between model-generated samples and human-annotated responses, hindering training effectiveness. To mitigate this issue, previous methods require additional preference annotation on newly generated samples to adapt to the shifted distribution, which consumes a large amount of annotation resources. Targeting more efficient human preference optimization, we propose an Adversarial Preference Optimization (APO) framework, in which the LLM and the reward model update alternatively via a min-max game. Through adversarial training, the reward model can adapt to the shifted generation distribution of the LLM without any additional annotation. With comprehensive experiments, we find the proposed adversarial training framework further enhances existing alignment baselines in terms of LLM helpfulness and harmlessness. The code is at https://github.com/Linear95/APO.
Authors: Fei Yuan, Shuai Yuan, Zhiyong Wu, Lei Li
Abstract: Large Language Models (LLMs), often show strong performance on English tasks, while exhibiting limitations on other languages. What is an LLM's multilingual capability when it is trained only on certain languages? The underlying mechanism remains unclear. This study endeavors to examine the multilingual capability of LLMs from the vocabulary sharing perspective by conducting an exhaustive analysis across 101 languages. Through the investigation of the performance gap before and after embedding fine-tuning, we discovered four distinct quadrants. By delving into each quadrant we provide actionable and efficient guidelines for tuning these languages. Extensive experiments reveal that existing LLMs possess multilingual capabilities that surpass our expectations, and we can significantly improve the multilingual performance of LLMs based on these attributes of each quadrant~\footnote{\url{https://github.com/CONE-MT/Vocabulary-Sharing-Facilitates-Multilingualism}.}.
URLs: https://github.com/CONE-MT/Vocabulary-Sharing-Facilitates-Multilingualism
Authors: Wenhong Zhu, Hongkun Hao, Zhiwei He, Yunze Song, Yumeng Zhang, Hanxu Hu, Yiran Wei, Rui Wang, Hongyuan Lu
Abstract: We are currently in an era of fierce competition among various large language models (LLMs) continuously pushing the boundaries of benchmark performance. However, genuinely assessing the capabilities of these LLMs has become a challenging and critical issue due to potential data contamination, and it wastes dozens of time and effort for researchers and engineers to download and try those contaminated models. To save our precious time, we propose a novel and useful method, Clean-Eval, which mitigates the issue of data contamination and evaluates the LLMs in a cleaner manner. Clean-Eval employs an LLM to paraphrase and back-translate the contaminated data into a candidate set, generating expressions with the same meaning but in different surface forms. A semantic detector is then used to filter the generated low-quality samples to narrow down this candidate set. The best candidate is finally selected from this set based on the BLEURT score. According to human assessment, this best candidate is semantically similar to the original contamination data but expressed differently. All candidates can form a new benchmark to evaluate the model. Our experiments illustrate that Clean-Eval substantially restores the actual evaluation results on contaminated LLMs under both few-shot learning and fine-tuning scenarios.
Authors: Haoan Jin, Siyuan Chen, Dilawaier Dilixiati, Yewei Jiang, Mengyue Wu, Kenny Q. Zhu
Abstract: Evaluating Large Language Models (LLMs) in the mental health domain poses distinct challenged from other domains, given the subtle and highly subjective nature of symptoms that exhibit significant variability among individuals. This paper presents PsyEval, the first comprehensive suite of mental health-related tasks for evaluating LLMs. PsyEval encompasses five sub-tasks that evaluate three critical dimensions of mental health. This comprehensive framework is designed to thoroughly assess the unique challenges and intricacies of mental health-related tasks, making PsyEval a highly specialized and valuable tool for evaluating LLM performance in this domain. We evaluate twelve advanced LLMs using PsyEval. Experiment results not only demonstrate significant room for improvement in current LLMs concerning mental health but also unveil potential directions for future model optimization.
Authors: I\~nigo Alonso, Eneko Agirre, Mirella Lapata
Abstract: Table-to-text generation involves generating appropriate textual descriptions given structured tabular data. It has attracted increasing attention in recent years thanks to the popularity of neural network models and the availability of large-scale datasets. A common feature across existing methods is their treatment of the input as a string, i.e., by employing linearization techniques that do not always preserve information in the table, are verbose, and lack space efficiency. We propose to rethink data-to-text generation as a visual recognition task, removing the need for rendering the input in a string format. We present PixT3, a multimodal table-to-text model that overcomes the challenges of linearization and input size limitations encountered by existing models. PixT3 is trained with a new self-supervised learning objective to reinforce table structure awareness and is applicable to open-ended and controlled generation settings. Experiments on the ToTTo and Logic2Text benchmarks show that PixT3 is competitive and, in some settings, superior to generators that operate solely on text.
Authors: Yunshui Li, Binyuan Hui, Xiaobo Xia, Jiaxi Yang, Min Yang, Lei Zhang, Shuzheng Si, Ling-Hao Chen, Junhao Liu, Tongliang Liu, Fei Huang, Yongbin Li
Abstract: Contemporary practices in instruction tuning often hinge on enlarging data scaling without a clear strategy for ensuring data quality, inadvertently introducing noise that may compromise model performance. To address this challenge, we introduce \textsc{Nuggets}, a novel and efficient methodology that leverages one-shot learning to discern and select high-quality instruction data from extensive datasets. \textsc{Nuggets} assesses the potential of individual instruction examples to act as effective one-shot learning instances, thereby identifying those that can significantly improve performance across diverse tasks. \textsc{Nuggets} utilizes a scoring system based on the impact of candidate examples on the perplexity of a diverse anchor set, facilitating the selection of the most advantageous data for instruction tuning. Through comprehensive evaluations on two benchmarks, including MT-Bench and Alpaca-Eval, we show that instruction tuning with the top 1\% of examples curated by \textsc{Nuggets} substantially outperforms conventional methods employing the entire dataset.
Authors: David Ponce, Thierry Etchegoyhen, Jes\'us Calleja P\'erez, Harritxu Gete
Abstract: The Split and Rephrase (SPRP) task, which consists in splitting complex sentences into a sequence of shorter grammatical sentences, while preserving the original meaning, can facilitate the processing of complex texts for humans and machines alike. It is also a valuable testbed to evaluate natural language processing models, as it requires modelling complex grammatical aspects. In this work, we evaluate large language models on the task, showing that they can provide large improvements over the state of the art on the main metrics, although still lagging in terms of splitting compliance. Results from two human evaluations further support the conclusions drawn from automated metric results. We provide a comprehensive study that includes prompting variants, domain shift, fine-tuned pretrained language models of varying parameter size and training data volumes, contrasted with both zero-shot and few-shot approaches on instruction-tuned language models. Although the latter were markedly outperformed by fine-tuned models, they may constitute a reasonable off-the-shelf alternative. Our results provide a fine-grained analysis of the potential and limitations of large language models for SPRP, with significant improvements achievable using relatively small amounts of training data and model parameters overall, and remaining limitations for all models on the task.
Authors: Jiaxi Cui, Liuzhenghao Lv, Jing Wen, Rongsheng Wang, Jing Tang, YongHong Tian, Li Yuan
Abstract: We present a novel approach for integrating Myers-Briggs Type Indicator (MBTI) personality traits into large language models (LLMs), addressing the challenges of personality consistency in personalized AI. Our method, "Machine Mindset," involves a two-phase fine-tuning and Direct Preference Optimization (DPO) to embed MBTI traits into LLMs. This approach ensures that models internalize these traits, offering a stable and consistent personality profile. We demonstrate the effectiveness of our models across various domains, showing alignment between model performance and their respective MBTI traits. The paper highlights significant contributions in the development of personality datasets and a new training methodology for personality integration in LLMs, enhancing the potential for personalized AI applications. We also open-sourced our model and part of the data at \url{https://github.com/PKU-YuanGroup/Machine-Mindset}.
Authors: Weiwen Xu, Deng Cai, Zhisong Zhang, Wai Lam, Shuming Shi
Abstract: As humans, we consistently interact with our peers and receive feedback in the form of natural language. This language feedback allows us to maintain appropriate behavior, and rectify potential errors. The question arises naturally: can we use language feedback to align large language models (LLMs)? In contrast to previous research that aligns LLMs with scalar rewards, we present the first systematic exploration of alignment through the lens of language feedback (i.e., judgment). We start with an in-depth investigation of potential methods that can be adapted for aligning LLMs with judgments, revealing that these methods cannot fully capitalize on judgments. To facilitate more effective utilization of judgments, we propose a novel framework, Contrastive Unlikelihood Training (CUT), that allows for fine-grained inappropriate content detection and correction based on judgments. Our results show that, with merely 1317 off-the-shelf judgment data, CUT (LLaMA2-13b) can beat the 175B DaVinci003 and surpass the best baseline by 50.84 points on AlpacaEval. CUT (LLaMA2-chat-13b) can also align LLMs in an iterative fashion using up-to-date model-specific judgments, improving performance from 81.09 to 91.68 points on AlpacaEval. Further analysis suggests that judgments hold greater potential than rewards in LLM alignment.
Authors: Shichao Sun, Junlong Li, Weizhe Yuan, Ruifeng Yuan, Wenjie Li, Pengfei Liu
Abstract: Critique, as a natural language description for assessing the quality of model-generated content, has played a vital role in the training, evaluation, and refinement of LLMs. However, a systematic method to evaluate the quality of critique is lacking. In this paper, we pioneer the critique of critique, termed MetaCritique, which builds specific quantification criteria. To achieve a reliable evaluation outcome, we propose Atomic Information Units (AIUs), which describe the critique in a more fine-grained manner. MetaCritique aggregates each AIU's judgment for the overall score. Moreover, MetaCritique delivers a natural language rationale for the intricate reasoning within each judgment. Lastly, we construct a meta-evaluation dataset covering 4 tasks across 16 public datasets involving human-written and LLM-generated critiques. Experiments demonstrate that MetaCritique can achieve near-human performance. Our study can facilitate future research in LLM critiques based on our following observations and released resources: (1) superior critiques judged by MetaCritique can lead to better refinements, indicating that it can potentially enhance the alignment of existing LLMs; (2) the leaderboard of critique models reveals that open-source critique models commonly suffer from factuality issues; (3) relevant code and data are publicly available at https://github.com/GAIR-NLP/MetaCritique to support deeper exploration; (4) an API at PyPI with the usage documentation in Appendix C allows users to assess the critique conveniently.
Authors: Harvey Lederman, Kyle Mahowald
Abstract: Are LLMs cultural technologies like photocopiers or printing presses, which transmit information but cannot create new content? A challenge for this idea, which we call bibliotechnism, is that LLMs generate novel text. We begin with a defense of bibliotechnism, showing how even novel text may inherit its meaning from original human-generated text. We then argue that bibliotechnism faces an independent challenge from examples in which LLMs generate novel reference, using new names to refer to new entities. Such examples could be explained if LLMs were not cultural technologies but had beliefs, desires, and intentions. According to interpretationism in the philosophy of mind, a system has such attitudes if and only if its behavior is well explained by the hypothesis that it does. Interpretationists may hold that LLMs have attitudes, and thus have a simple solution to the novel reference problem. We emphasize, however, that interpretationism is compatible with very simple creatures having attitudes and differs sharply from views that presuppose these attitudes require consciousness, sentience, or intelligence (topics about which we make no claims).
Authors: Siheng Xiong, Ali Payani, Ramana Kompella, Faramarz Fekri
Abstract: While large language models (LLMs) have demonstrated remarkable reasoning capabilities, they are not without their flaws and inaccuracies. Recent studies have introduced various methods to mitigate these limitations. Temporal reasoning (TR), in particular, presents a significant challenge for LLMs due to its reliance on diverse temporal concepts and intricate temporal logic. In this paper, we propose TG-LLM, a novel framework towards language-based TR. Instead of reasoning over the original context, we adopt a latent representation, temporal graph (TG) that enhances the learning of TR. A synthetic dataset (TGQA), which is fully controllable and requires minimal supervision, is constructed for fine-tuning LLMs on this text-to-TG translation task. We confirmed in experiments that the capability of TG translation learned on our dataset can be transferred to other TR tasks and benchmarks. On top of that, we teach LLM to perform deliberate reasoning over the TGs via Chain-of-Thought (CoT) bootstrapping and graph data augmentation. We observed that those strategies, which maintain a balance between usefulness and diversity, bring more reliable CoTs and final results than the vanilla CoT distillation.
Authors: Moritz Plenz, Anette Frank
Abstract: While Language Models (LMs) are the workhorses of NLP, their interplay with structured knowledge graphs (KGs) is still actively researched. Current methods for encoding such graphs typically either (i) linearize them for embedding with LMs -- which underutilize structural information, or (ii) use Graph Neural Networks (GNNs) to preserve the graph structure -- but GNNs cannot represent text features as well as pretrained LMs. In our work we introduce a novel LM type, the Graph Language Model (GLM), that integrates the strengths of both approaches and mitigates their weaknesses. The GLM parameters are initialized from a pretrained LM to enhance understanding of individual graph concepts and triplets. Simultaneously, we design the GLM's architecture to incorporate graph biases, thereby promoting effective knowledge distribution within the graph. This enables GLMs to process graphs, texts, and interleaved inputs of both. Empirical evaluations on relation classification tasks show that GLM embeddings surpass both LM- and GNN-based baselines in supervised and zero-shot setting, demonstrating their versatility.
Authors: Haoran Xu, Amr Sharaf, Yunmo Chen, Weiting Tan, Lingfeng Shen, Benjamin Van Durme, Kenton Murray, Young Jin Kim
Abstract: Moderate-sized large language models (LLMs) -- those with 7B or 13B parameters -- exhibit promising machine translation (MT) performance. However, even the top-performing 13B LLM-based translation models, like ALMA, does not match the performance of state-of-the-art conventional encoder-decoder translation models or larger-scale LLMs such as GPT-4. In this study, we bridge this performance gap. We first assess the shortcomings of supervised fine-tuning for LLMs in the MT task, emphasizing the quality issues present in the reference data, despite being human-generated. Then, in contrast to SFT which mimics reference translations, we introduce Contrastive Preference Optimization (CPO), a novel approach that trains models to avoid generating adequate but not perfect translations. Applying CPO to ALMA models with only 22K parallel sentences and 12M parameters yields significant improvements. The resulting model, called ALMA-R, can match or exceed the performance of the WMT competition winners and GPT-4 on WMT'21, WMT'22 and WMT'23 test datasets.
Authors: Dongkeun Yoon, Joel Jang, Sungdong Kim, Seungone Kim, Sheikh Shafayat, Minjoon Seo
Abstract: We introduce LangBridge, a zero-shot approach to adapt language models for multilingual reasoning tasks without multilingual supervision. LangBridge operates by bridging two models, each specialized in different aspects: (1) one specialized in understanding multiple languages (e.g., mT5 encoder) and (2) one specialized in reasoning (e.g., MetaMath). LangBridge connects the two models by introducing minimal trainable parameters between them. Despite utilizing only English data for training, LangBridge considerably enhances the performance of language models on low-resource languages across mathematical reasoning, code completion, logical reasoning, and commonsense reasoning. Our analysis suggests that the efficacy of LangBridge stems from the language-agnostic characteristics of multilingual representations. We publicly release our code and models.
Authors: Xuanming Zhang, Zixun Chen, Zhou Yu
Abstract: Lexical Substitution discovers appropriate substitutes for a given target word in a context sentence. However, the task fails to consider substitutes that are of equal or higher proficiency than the target, an aspect that could be beneficial for language learners looking to improve their writing. To bridge this gap, we propose a new task, language proficiency-oriented lexical substitution. We also introduce ProLex, a novel benchmark designed to assess systems' ability to generate not only appropriate substitutes but also substitutes that demonstrate better language proficiency. Besides the benchmark, we propose models that can automatically perform the new task. We show that our best model, a Llama2-13B model fine-tuned with task-specific synthetic data, outperforms ChatGPT by an average of 3.2% in F-score and achieves comparable results with GPT-4 on ProLex.
Authors: Shijue Huang, Wanjun Zhong, Jianqiao Lu, Qi Zhu, Jiahui Gao, Weiwen Liu, Yutai Hou, Xingshan Zeng, Yasheng Wang, Lifeng Shang, Xin Jiang, Ruifeng Xu, Qun Liu
Abstract: The recent trend of using Large Language Models (LLMs) as tool agents in real-world applications underscores the necessity for comprehensive evaluations of their capabilities, particularly in complex scenarios involving planning, creating, and using tools. However, existing benchmarks typically focus on simple synthesized queries that do not reflect real-world complexity, thereby offering limited perspectives in evaluating tool utilization. To address this issue, we present UltraTool, a novel benchmark designed to improve and evaluate LLMs' ability in tool utilization within real-world scenarios. UltraTool focuses on the entire process of using tools - from planning and creating to applying them in complex tasks. It emphasizes real-world complexities, demanding accurate, multi-step planning for effective problem-solving. A key feature of UltraTool is its independent evaluation of planning with natural language, which happens before tool usage and simplifies the task solving by mapping out the intermediate steps. Thus, unlike previous work, it eliminates the restriction of pre-defined toolset. Through extensive experiments on various LLMs, we offer novel insights into the evaluation of capabilities of LLMs in tool utilization, thereby contributing a fresh perspective to this rapidly evolving field. The benchmark is publicly available at https://github.com/JoeYing1019/UltraTool.
Authors: Yuan Chiang, Elvis Hsieh, Chia-Hong Chou, Janosh Riebesell
Abstract: Reducing hallucination of Large Language Models (LLMs) is imperative for use in the sciences, where reliability and reproducibility are crucial. However, LLMs inherently lack long-term memory, making it a nontrivial, ad hoc, and inevitably biased task to fine-tune them on domain-specific literature and data. Here we introduce LLaMP, a multimodal retrieval-augmented generation (RAG) framework of hierarchical reasoning-and-acting (ReAct) agents that can dynamically and recursively interact with computational and experimental data on Materials Project (MP) and run atomistic simulations via high-throughput workflow interface. Without fine-tuning, LLaMP demonstrates strong tool usage ability to comprehend and integrate various modalities of materials science concepts, fetch relevant data stores on the fly, process higher-order data (such as crystal structure and elastic tensor), and streamline complex tasks in computational materials and chemistry. We propose a simple metric combining uncertainty and confidence estimates to evaluate the self-consistency of responses by LLaMP and vanilla LLMs. Our benchmark shows that LLaMP effectively mitigates the intrinsic bias in LLMs, counteracting the errors on bulk moduli, electronic bandgaps, and formation energies that seem to derive from mixed data sources. We also demonstrate LLaMP's capability to edit crystal structures and run annealing molecular dynamics simulations using pre-trained machine-learning force fields. The framework offers an intuitive and nearly hallucination-free approach to exploring and scaling materials informatics, and establishes a pathway for knowledge distillation and fine-tuning other language models. Code and live demo are available at https://github.com/chiang-yuan/llamp
Authors: Jiaming Ji, Boyuan Chen, Hantao Lou, Donghai Hong, Borong Zhang, Xuehai Pan, Juntao Dai, Tianyi Qiu, Yaodong Yang
Abstract: With the rapid development of large language models (LLMs) and ever-evolving practical requirements, finding an efficient and effective alignment method has never been more critical. However, the tension between the complexity of current alignment methods and the need for rapid iteration in deployment scenarios necessitates the development of a model-agnostic alignment approach that can operate under these constraints. In this paper, we introduce Aligner, a novel and simple alignment paradigm that learns the correctional residuals between preferred and dispreferred answers using a small model. Designed as a model-agnostic, plug-and-play module, Aligner can be directly applied to various open-source and API-based models with only one-off training, making it suitable for rapid iteration. Notably, Aligner can be applied to any powerful, large-scale upstream models. Moreover, it can even iteratively bootstrap the upstream models using corrected responses as synthetic human preference data, breaking through the model's performance ceiling. Our experiments demonstrate performance improvements by deploying the same Aligner model across 11 different LLMs, evaluated on the 3H dimensions (helpfulness, harmlessness, and honesty). Specifically, Aligner-7B has achieved an average improvement of 68.9\% in helpfulness and 23.8\% in harmlessness across the tested LLMs while also effectively reducing hallucination. In the Alpaca-Eval leaderboard, stacking Aligner-2B on GPT-4 Turbo improved its LC Win Rate from 55.0\% to 58.3\%, surpassing GPT-4 Omni's 57.5\% Win Rate (community report).
Authors: Fei Yuan, Chang Ma, Shuai Yuan, Qiushi Sun, Lei Li
Abstract: The lottery ticket hypothesis posits the existence of ``winning tickets'' within a randomly initialized neural network. Do winning tickets exist for LLMs in fine-tuning scenarios? How can we find such winning tickets? In this paper, we propose KS-Lottery, a method to identify a small subset of LLM parameters highly effective in multilingual fine-tuning. Our key idea is to use Kolmogorov-Smirnov Test to analyze the distribution shift of parameters before and after fine-tuning. We further theoretically prove that KS-Lottery can find the certified winning tickets in the embedding layer, fine-tuning on the found parameters is guaranteed to perform as well as full fine-tuning. Comparing KS-Lottery with other parameter-efficient tuning algorithms on translation tasks, the experimental results show that KS-Lottery finds a much smaller set of parameters for fine-tuning while achieving the comparable performance as full fine-tuning LLM. Surprisingly, we find that fine-tuning 18 tokens' embedding of LLaMA suffices to reach the fine-tuning translation performance~\footnote{https://github.com/CONE-MT/KS-Lottery.}.
Authors: Jannis Vamvas, Rico Sennrich
Abstract: Minimum Bayes Risk (MBR) decoding is a text generation technique that has been shown to improve the quality of machine translations, but is expensive, even if a sampling-based approximation is used. Besides requiring a large number of sampled sequences, it requires the pairwise calculation of a utility metric, which has quadratic complexity. In this paper, we propose to approximate pairwise metric scores with scores calculated against aggregated reference representations. This changes the complexity of utility estimation from $O(n^2)$ to $O(n)$, while empirically preserving most of the quality gains of MBR decoding. We release our source code at https://github.com/ZurichNLP/mbr
Authors: Yiyou Sun, Junjie Hu, Wei Cheng, Haifeng Chen
Abstract: This paper introduces the retrieval-augmented large language model with Definite Finite Automaton (DFA-RAG), a novel framework designed to enhance the capabilities of conversational agents using large language models (LLMs). Traditional LLMs face challenges in generating regulated and compliant responses in special scenarios with predetermined response guidelines, like emotional support and customer service. Our framework addresses these challenges by embedding a Definite Finite Automaton (DFA), learned from training dialogues, within the LLM. This structured approach acts as a semantic router which enables the LLM to adhere to a deterministic response pathway. The routing is achieved by the retrieval-augmentation generation (RAG) strategy, which carefully selects dialogue examples aligned with the current conversational context. The advantages of DFA-RAG include an interpretable structure through human-readable DFA, context-aware retrieval for responses in conversations, and plug-and-play compatibility with existing LLMs. Extensive benchmarks validate DFA-RAG's effectiveness, indicating its potential as a valuable contribution to the conversational agent.
Authors: Haihui Yang, Xiaojun Quan
Abstract: Chinese grammatical error correction (CGEC) faces serious overcorrection challenges when employing autoregressive generative models such as sequence-to-sequence (Seq2Seq) models and decoder-only large language models (LLMs). While previous methods aim to address overcorrection in Seq2Seq models, they are difficult to adapt to decoder-only LLMs. In this paper, we propose an alignment-enhanced corrector for the overcorrection problem that applies to both Seq2Seq models and decoder-only LLMs. Our method first trains a correction model to generate an initial correction of the source sentence. Then, we combine the source sentence with the initial correction and feed it through an alignment model for another round of correction, aiming to enforce the alignment model to focus on potential overcorrection. Moreover, to enhance the model's ability to identify nuances, we further explore the reverse alignment of the source sentence and the initial correction. Finally, we transfer the alignment knowledge from two alignment models to the correction model, instructing it on how to avoid overcorrection. Experimental results on three CGEC datasets demonstrate the effectiveness of our approach in alleviating overcorrection and improving overall performance. Our code has been made publicly available.
Authors: Jianhui Pang, Fanghua Ye, Derek Fai Wong, Xin He, Wanshun Chen, Longyue Wang
Abstract: Large language models (LLMs) predominantly employ decoder-only transformer architectures, necessitating the retention of keys/values information for historical tokens to provide contextual information and avoid redundant computation. However, the substantial size and parameter volume of these LLMs require massive GPU memory. This memory demand increases with the length of the input text, leading to an urgent need for more efficient methods of information storage and processing. This study introduces Anchor-based LLMs (AnLLMs), which utilize an innovative anchor-based self-attention network (AnSAN) and also an anchor-based inference strategy. This approach enables LLMs to compress sequence information into an anchor token, reducing the keys/values cache and enhancing inference efficiency. Experiments on question-answering benchmarks reveal that AnLLMs maintain similar accuracy levels while achieving up to 99% keys/values cache reduction and up to 3.5 times faster inference. Despite a minor compromise in accuracy, the substantial enhancements of AnLLMs employing the AnSAN technique in resource utilization and computational efficiency underscore their potential for practical LLM applications.
Authors: Tobias Schimanski, Jingwei Ni, Mathias Kraus, Elliott Ash, Markus Leippold
Abstract: Advances towards more faithful and traceable answers of Large Language Models (LLMs) are crucial for various research and practical endeavors. One avenue in reaching this goal is basing the answers on reliable sources. However, this Evidence-Based QA has proven to work insufficiently with LLMs in terms of citing the correct sources (source quality) and truthfully representing the information within sources (answer attributability). In this work, we systematically investigate how to robustly fine-tune LLMs for better source quality and answer attributability. Specifically, we introduce a data generation pipeline with automated data quality filters, which can synthesize diversified high-quality training and testing data at scale. We further introduce four test sets to benchmark the robustness of fine-tuned specialist models. Extensive evaluation shows that fine-tuning on synthetic data improves performance on both in- and out-of-distribution. Furthermore, we show that data quality, which can be drastically improved by proposed quality filters, matters more than quantity in improving Evidence-Based QA.
Authors: Xiangming Gu, Xiaosen Zheng, Tianyu Pang, Chao Du, Qian Liu, Ye Wang, Jing Jiang, Min Lin
Abstract: A multimodal large language model (MLLM) agent can receive instructions, capture images, retrieve histories from memory, and decide which tools to use. Nonetheless, red-teaming efforts have revealed that adversarial images/prompts can jailbreak an MLLM and cause unaligned behaviors. In this work, we report an even more severe safety issue in multi-agent environments, referred to as infectious jailbreak. It entails the adversary simply jailbreaking a single agent, and without any further intervention from the adversary, (almost) all agents will become infected exponentially fast and exhibit harmful behaviors. To validate the feasibility of infectious jailbreak, we simulate multi-agent environments containing up to one million LLaVA-1.5 agents, and employ randomized pair-wise chat as a proof-of-concept instantiation for multi-agent interaction. Our results show that feeding an (infectious) adversarial image into the memory of any randomly chosen agent is sufficient to achieve infectious jailbreak. Finally, we derive a simple principle for determining whether a defense mechanism can provably restrain the spread of infectious jailbreak, but how to design a practical defense that meets this principle remains an open question to investigate. Our project page is available at https://sail-sg.github.io/Agent-Smith/.
Authors: Jiwon Song, Kyungseok Oh, Taesu Kim, Hyungjun Kim, Yulhwa Kim, Jae-Joon Kim
Abstract: Large language models (LLMs) have proven to be highly effective across various natural language processing tasks. However, their large number of parameters poses significant challenges for practical deployment. Pruning, a technique aimed at reducing the size and complexity of LLMs, offers a potential solution by removing redundant components from the network. Despite the promise of pruning, existing methods often struggle to achieve substantial end-to-end LLM inference speedup. In this paper, we introduce SLEB, a novel approach designed to streamline LLMs by eliminating redundant transformer blocks. We choose the transformer block as the fundamental unit for pruning, because LLMs exhibit block-level redundancy with high similarity between the outputs of neighboring blocks. This choice allows us to effectively enhance the processing speed of LLMs. Our experimental results demonstrate that SLEB outperforms previous LLM pruning methods in accelerating LLM inference while also maintaining superior perplexity and accuracy, making SLEB as a promising technique for enhancing the efficiency of LLMs. The code is available at: https://github.com/jiwonsong-dev/SLEB.
Authors: Kenza Amara, Rita Sevastjanova, Mennatallah El-Assady
Abstract: To harness the power of large language models in safety-critical domains, we need to ensure the explainability of their predictions. However, despite the significant attention to model interpretability, there remains an unexplored domain in explaining sequence-to-sequence tasks using methods tailored for textual data. This paper introduces SyntaxShap, a local, model-agnostic explainability method for text generation that takes into consideration the syntax in the text data. The presented work extends Shapley values to account for parsing-based syntactic dependencies. Taking a game theoric approach, SyntaxShap only considers coalitions constraint by the dependency tree. We adopt a model-based evaluation to compare SyntaxShap and its weighted form to state-of-the-art explainability methods adapted to text generation tasks, using diverse metrics including faithfulness, coherency, and semantic alignment of the explanations to the model. We show that our syntax-aware method produces explanations that help build more faithful and coherent explanations for predictions by autoregressive models. Confronted with the misalignment of human and AI model reasoning, this paper also highlights the need for cautious evaluation strategies in explainable AI.
Authors: Shih-Yang Liu, Chien-Yi Wang, Hongxu Yin, Pavlo Molchanov, Yu-Chiang Frank Wang, Kwang-Ting Cheng, Min-Hung Chen
Abstract: Among the widely used parameter-efficient fine-tuning (PEFT) methods, LoRA and its variants have gained considerable popularity because of avoiding additional inference costs. However, there still often exists an accuracy gap between these methods and full fine-tuning (FT). In this work, we first introduce a novel weight decomposition analysis to investigate the inherent differences between FT and LoRA. Aiming to resemble the learning capacity of FT from the findings, we propose Weight-Decomposed Low-Rank Adaptation (DoRA). DoRA decomposes the pre-trained weight into two components, magnitude and direction, for fine-tuning, specifically employing LoRA for directional updates to efficiently minimize the number of trainable parameters. By employing \ours, we enhance both the learning capacity and training stability of LoRA while avoiding any additional inference overhead. \ours~consistently outperforms LoRA on fine-tuning LLaMA, LLaVA, and VL-BART on various downstream tasks, such as commonsense reasoning, visual instruction tuning, and image/video-text understanding. Code is available at https://github.com/NVlabs/DoRA.
Authors: Jingwei Ni, Minjing Shi, Dominik Stammbach, Mrinmaya Sachan, Elliott Ash, Markus Leippold
Abstract: With the rise of generative AI, automated fact-checking methods to combat misinformation are becoming more and more important. However, factual claim detection, the first step in a fact-checking pipeline, suffers from two key issues that limit its scalability and generalizability: (1) inconsistency in definitions of the task and what a claim is, and (2) the high cost of manual annotation. To address (1), we review the definitions in related work and propose a unifying definition of factual claims that focuses on verifiability. To address (2), we introduce AFaCTA (Automatic Factual Claim deTection Annotator), a novel framework that assists in the annotation of factual claims with the help of large language models (LLMs). AFaCTA calibrates its annotation confidence with consistency along three predefined reasoning paths. Extensive evaluation and experiments in the domain of political speech reveal that AFaCTA can efficiently assist experts in annotating factual claims and training high-quality classifiers, and can work with or without expert supervision. Our analyses also result in PoliClaim, a comprehensive claim detection dataset spanning diverse political topics.
Authors: Xuan Ren, Biao Wu, Lingqiao Liu
Abstract: This paper explores an intriguing observation: fine-tuning a large language model (LLM) with responses generated by a LLM often yields better results than using responses generated by humans. We conduct an in-depth investigation to understand why this occurs. Contrary to the common belief that these instances is simply due to the more detailed nature of LLM-generated content, our study identifies another contributing factor: an LLM is inherently more "familiar" with LLM generated responses. This familiarity is evidenced by lower perplexity before fine-tuning. We design a series of experiments to understand the impact of the "familiarity" and our conclusion reveals that this "familiarity" significantly impacts learning performance. Training with LLM-generated responses not only enhances performance but also helps maintain the model's capabilities in other tasks after fine-tuning on a specific task.
Authors: Yixin Yang, Zheng Li, Qingxiu Dong, Heming Xia, Zhifang Sui
Abstract: Understanding the deep semantics of images is essential in the era dominated by social media. However, current research works primarily on the superficial description of images, revealing a notable deficiency in the systematic investigation of the inherent deep semantics. In this work, we introduce DEEPEVAL, a comprehensive benchmark to assess Large Multimodal Models' (LMMs) capacities of visual deep semantics. DEEPEVAL includes human-annotated dataset and three progressive subtasks: fine-grained description selection, in-depth title matching, and deep semantics understanding. Utilizing DEEPEVAL, we evaluate 9 open-source LMMs and GPT-4V(ision). Our evaluation demonstrates a substantial gap between the deep semantic comprehension capabilities of existing LMMs and humans. For example, GPT-4V is 30% behind humans in understanding deep semantics, even though it achieves human-comparable performance in image description. Further analysis reveals that LMM performance on DEEPEVAL varies according to the specific facets of deep semantics explored, indicating the fundamental challenges remaining in developing LMMs.
Authors: Zhihao Wen, Jie Zhang, Yuan Fang
Abstract: Fine-tuning all parameters of large language models (LLMs) necessitates substantial computational power and extended time. Latest advancements in parameter-efficient fine-tuning (PEFT) techniques, such as Adapter tuning and LoRA, allow for adjustments to only a minor fraction of the parameters of these LLMs. Concurrently, it has been noted that the issue of over-smoothing diminishes the effectiveness of these Transformer-based LLMs, resulting in suboptimal performances in downstream tasks. In this paper, we present SIBO, which is a SImple BOoster to enhance PEFT, by injecting an initial residual. SIBO is straightforward and readily extensible to a range of state-of-the-art PEFT techniques to alleviate over-smoothing and enhance performance. Extensive experiments on 22 benchmark datasets demonstrate that SIBO significantly enhances the performance of various strong baselines, achieving up to 15.7% and 23.5% improvement over existing PEFT methods on the arithmetic and commonsense reasoning tasks, respectively.
Authors: Tianjie Ju, Yijin Chen, Xinwei Yuan, Zhuosheng Zhang, Wei Du, Yubin Zheng, Gongshen Liu
Abstract: Recent work has showcased the powerful capability of large language models (LLMs) in recalling knowledge and reasoning. However, the reliability of LLMs in combining these two capabilities into reasoning through multi-hop facts has not been widely explored. This paper systematically investigates the possibilities for LLMs to utilize shortcuts based on direct connections between the initial and terminal entities of multi-hop knowledge. We first explore the existence of factual shortcuts through Knowledge Neurons, revealing that: (i) the strength of factual shortcuts is highly correlated with the frequency of co-occurrence of initial and terminal entities in the pre-training corpora; (ii) few-shot prompting leverage more shortcuts in answering multi-hop questions compared to chain-of-thought prompting. Then, we analyze the risks posed by factual shortcuts from the perspective of multi-hop knowledge editing. Analysis shows that approximately 20% of the failures are attributed to shortcuts, and the initial and terminal entities in these failure instances usually have higher co-occurrences in the pre-training corpus. Finally, we propose erasing shortcut neurons to mitigate the associated risks and find that this approach significantly reduces failures in multiple-hop knowledge editing caused by shortcuts.
Authors: Xinbei Ma, Zhuosheng Zhang, Hai Zhao
Abstract: Multimodal large language models (MLLMs) have shown remarkable potential as human-like autonomous language agents to interact with real-world environments, especially for graphical user interface (GUI) automation. However, those GUI agents require comprehensive cognition ability including exhaustive perception and reliable action response. We propose a Comprehensive Cognitive LLM Agent, CoCo-Agent, with two novel approaches, comprehensive environment perception (CEP) and conditional action prediction (CAP), to systematically improve the GUI automation performance. First, CEP facilitates the GUI perception through different aspects and granularity, including screenshots and complementary detailed layouts for the visual channel and historical actions for the textual channel. Second, CAP decomposes the action prediction into sub-problems: action type prediction and action target conditioned on the action type. With our technical design, our agent achieves new state-of-the-art performance on AITW and META-GUI benchmarks, showing promising abilities in realistic scenarios. Code is available at https://github.com/xbmxb/CoCo-Agent.
Authors: Zongru Wu, Zhuosheng Zhang, Pengzhou Cheng, Gongshen Liu
Abstract: Despite the notable success of language models (LMs) in various natural language processing (NLP) tasks, the reliability of LMs is susceptible to backdoor attacks. Prior research attempts to mitigate backdoor learning while training the LMs on the poisoned dataset, yet struggles against complex backdoor attacks in real-world scenarios. In this paper, we investigate the learning mechanisms of backdoor LMs in the frequency space by Fourier analysis. Our findings indicate that the backdoor mapping presented on the poisoned datasets exhibits a more discernible inclination towards lower frequency compared to clean mapping, resulting in the faster convergence of backdoor mapping. To alleviate this dilemma, we propose Multi-Scale Low-Rank Adaptation (MuScleLoRA), which deploys multiple radial scalings in the frequency space with low-rank adaptation to the target model and further aligns the gradients when updating parameters. Through downscaling in the frequency space, MuScleLoRA encourages the model to prioritize the learning of relatively high-frequency clean mapping, consequently mitigating backdoor learning. Experimental results demonstrate that MuScleLoRA outperforms baselines significantly. Notably, MuScleLoRA reduces the average success rate of diverse backdoor attacks to below 15\% across multiple datasets and generalizes to various backbone LMs, including BERT, RoBERTa, GPT2-XL, and Llama2. The codes are publicly available at https://github.com/ZrW00/MuScleLoRA.
Authors: Brielen Madureira, Patrick Kahardipraja, David Schlangen
Abstract: Incremental models that process sentences one token at a time will sometimes encounter points where more than one interpretation is possible. Causal models are forced to output one interpretation and continue, whereas models that can revise may edit their previous output as the ambiguity is resolved. In this work, we look at how restart-incremental Transformers build and update internal states, in an effort to shed light on what processes cause revisions not viable in autoregressive models. We propose an interpretable way to analyse the incremental states, showing that their sequential structure encodes information on the garden path effect and its resolution. Our method brings insights on various bidirectional encoders for contextualised meaning representation and dependency parsing, contributing to show their advantage over causal models when it comes to revisions.
Authors: Pengcheng Qiu, Chaoyi Wu, Xiaoman Zhang, Weixiong Lin, Haicheng Wang, Ya Zhang, Yanfeng Wang, Weidi Xie
Abstract: The development of open-source, multilingual medical language models can benefit a wide, linguistically diverse audience from different regions. To promote this domain, we present contributions from the following: First, we construct a multilingual medical corpus, containing approximately 25.5B tokens encompassing 6 main languages, termed as MMedC, enabling auto-regressive domain adaptation for general LLMs; Second, to monitor the development of multilingual medical LLMs, we propose a multilingual medical multi-choice question-answering benchmark with rationale, termed as MMedBench; Third, we have assessed a number of open-source large language models (LLMs) on our benchmark, along with those further auto-regressive trained on MMedC. Our final model, MMed-Llama 3, with only 8B parameters, achieves superior performance compared to all other open-source models on both MMedBench and English benchmarks, even rivaling GPT-4. In conclusion, in this work, we present a large-scale corpus, a benchmark and a series of models to support the development of multilingual medical LLMs.
Authors: Yunfan Shao, Linyang Li, Zhaoye Fei, Hang Yan, Dahua Lin, Xipeng Qiu
Abstract: Data plays a fundamental role in the training of Large Language Models (LLMs). While attention has been paid to the collection and composition of datasets, determining the data sampling strategy in training remains an open question. Most LLMs are trained with a simple strategy, random sampling. However, this sampling strategy ignores the unbalanced nature of training data distribution, which can be sub-optimal. In this paper, we propose ClusterClip Sampling to balance the text distribution of training data for better model training. Specifically, ClusterClip Sampling utilizes data clustering to reflect the data distribution of the training set and balances the common samples and rare samples during training based on the cluster results. A repetition clip operation is introduced to mitigate the overfitting issue led by samples from certain clusters. Extensive experiments validate the effectiveness of ClusterClip Sampling, which outperforms random sampling and other cluster-based sampling variants under various training datasets and large language models.
Authors: Zhipeng Xu, Zhenghao Liu, Yukun Yan, Zhiyuan Liu, Ge Yu, Chenyan Xiong
Abstract: The web contains large-scale, diverse, and abundant information to satisfy the information-seeking needs of humans. Through meticulous data collection, preprocessing, and curation, webpages can be used as a fundamental data resource for language model pretraining. However, when confronted with the progressively revolutionized and intricate nature of webpages, rule-based/feature-based web scrapers are becoming increasingly inadequate. This paper presents a simple, fast, and effective Neural web Scraper (NeuScraper) to help extract primary and clean text contents from webpages. Experimental results show that NeuScraper surpasses the baseline scrapers by achieving more than a 20% improvement, demonstrating its potential in extracting higher-quality data to facilitate the language model pretraining. All of the code is available at https://github.com/OpenMatch/NeuScraper.
Authors: Zicheng Lin, Zhibin Gou, Tian Liang, Ruilin Luo, Haowei Liu, Yujiu Yang
Abstract: The ability of Large Language Models (LLMs) to critique and refine their reasoning is crucial for their application in evaluation, feedback provision, and self-improvement. This paper introduces CriticBench, a comprehensive benchmark designed to assess LLMs' abilities to critique and rectify their reasoning across a variety of tasks. CriticBench encompasses five reasoning domains: mathematical, commonsense, symbolic, coding, and algorithmic. It compiles 15 datasets and incorporates responses from three LLM families. Utilizing CriticBench, we evaluate and dissect the performance of 17 LLMs in generation, critique, and correction reasoning, i.e., GQC reasoning. Our findings reveal: (1) a linear relationship in GQC capabilities, with critique-focused training markedly enhancing performance; (2) a task-dependent variation in correction effectiveness, with logic-oriented tasks being more amenable to correction; (3) GQC knowledge inconsistencies that decrease as model size increases; and (4) an intriguing inter-model critiquing dynamic, where stronger models are better at critiquing weaker ones, while weaker models can surprisingly surpass stronger ones in their self-critique. We hope these insights into the nuanced critique-correct reasoning of LLMs will foster further research in LLM critique and self-improvement.
Authors: Philipp Mondorf, Barbara Plank
Abstract: Deductive reasoning plays a pivotal role in the formulation of sound and cohesive arguments. It allows individuals to draw conclusions that logically follow, given the truth value of the information provided. Recent progress in the domain of large language models (LLMs) has showcased their capability in executing deductive reasoning tasks. Nonetheless, a significant portion of research primarily assesses the accuracy of LLMs in solving such tasks, often overlooking a deeper analysis of their reasoning behavior. In this study, we draw upon principles from cognitive psychology to examine inferential strategies employed by LLMs, through a detailed evaluation of their responses to propositional logic problems. Our findings indicate that LLMs display reasoning patterns akin to those observed in humans, including strategies like $\textit{supposition following}$ or $\textit{chain construction}$. Moreover, our research demonstrates that the architecture and scale of the model significantly affect its preferred method of reasoning, with more advanced models tending to adopt strategies more frequently than less sophisticated ones. Importantly, we assert that a model's accuracy, that is the correctness of its final conclusion, does not necessarily reflect the validity of its reasoning process. This distinction underscores the necessity for more nuanced evaluation procedures in the field.
Authors: Zhuohao Yu, Chang Gao, Wenjin Yao, Yidong Wang, Wei Ye, Jindong Wang, Xing Xie, Yue Zhang, Shikun Zhang
Abstract: Automatic evaluation methods for large language models (LLMs) are hindered by data contamination, leading to inflated assessments of their effectiveness. Existing strategies, which aim to detect contaminated texts, focus on quantifying contamination status instead of accurately gauging model performance. In this paper, we introduce KIEval, a Knowledge-grounded Interactive Evaluation framework, which incorporates an LLM-powered "interactor" role for the first time to accomplish a dynamic contamination-resilient evaluation. Starting with a question in a conventional LLM benchmark involving domain-specific knowledge, KIEval utilizes dynamically generated, multi-round, and knowledge-focused dialogues to determine whether a model's response is merely a recall of benchmark answers or demonstrates a deep comprehension to apply knowledge in more complex conversations. Extensive experiments on seven leading LLMs across five datasets validate KIEval's effectiveness and generalization. We also reveal that data contamination brings no contribution or even negative effect to models' real-world applicability and understanding, and existing contamination detection methods for LLMs can only identify contamination in pre-training but not during supervised fine-tuning.
Authors: Zhou Yang, Zhaochun Ren, Yufeng Wang, Chao Chen, Haizhou Sun, Xiaofei Zhu, Xiangwen Liao
Abstract: Empathetic response generation aims to comprehend the cognitive and emotional states in dialogue utterances and generate proper responses. Psychological theories posit that comprehending emotional and cognitive states necessitates iteratively capturing and understanding associated words across dialogue utterances. However, existing approaches regard dialogue utterances as either a long sequence or independent utterances for comprehension, which are prone to overlook the associated words between them. To address this issue, we propose an Iterative Associative Memory Model (IAMM) for empathetic response generation. Specifically, we employ a novel second-order interaction attention mechanism to iteratively capture vital associated words between dialogue utterances and situations, dialogue history, and a memory module (for storing associated words), thereby accurately and nuancedly comprehending the utterances. We conduct experiments on the Empathetic-Dialogue dataset. Both automatic and human evaluations validate the efficacy of the model. Variant experiments on LLMs also demonstrate that attending to associated words improves empathetic comprehension and expression.
Authors: Wenhong Zhu, Hongkun Hao, Zhiwei He, Yiming Ai, Rui Wang
Abstract: Current language models decode text token by token according to probabilistic distribution, and determining the appropriate candidates for the next token is crucial to ensure generation quality. This study introduces adaptive decoding, a mechanism that dynamically empowers language models to ascertain a sensible candidate set during generation. Specifically, we introduce an entropy-based metric called confidence and conceptualize determining the optimal candidate set as a confidence-increasing process. The rationality of including a token in the candidate set is assessed by leveraging the increment of confidence. Experimental results reveal that our method balances diversity and coherence well. The human evaluation shows that our method can generate human-preferred text. Additionally, our method can potentially improve the reasoning ability of language models.
Authors: Ansh Arora, Xuanli He, Maximilian Mozes, Srinibas Swain, Mark Dras, Qiongkai Xu
Abstract: The democratization of pre-trained language models through open-source initiatives has rapidly advanced innovation and expanded access to cutting-edge technologies. However, this openness also brings significant security risks, including backdoor attacks, where hidden malicious behaviors are triggered by specific inputs, compromising natural language processing (NLP) system integrity and reliability. This paper suggests that merging a backdoored model with other homogeneous models can significantly remediate backdoor vulnerabilities even if such models are not entirely secure. In our experiments, we verify our hypothesis on various models (BERT-Base, RoBERTa-Large, Llama2-7B, and Mistral-7B) and datasets (SST-2, OLID, AG News, and QNLI). Compared to multiple advanced defensive approaches, our method offers an effective and efficient inference-stage defense against backdoor attacks on classification and instruction-tuned tasks without additional resources or specific knowledge. Our approach consistently outperforms recent advanced baselines, leading to an average of about 75% reduction in the attack success rate. Since model merging has been an established approach for improving model performance, the extra advantage it provides regarding defense can be seen as a cost-free bonus.
Authors: Taichi Aida, Danushka Bollegala
Abstract: Detecting temporal semantic changes of words is an important task for various NLP applications that must make time-sensitive predictions. Lexical Semantic Change Detection (SCD) task involves predicting whether a given target word, $w$, changes its meaning between two different text corpora, $C_1$ and $C_2$. For this purpose, we propose a supervised two-staged SCD method that uses existing Word-in-Context (WiC) datasets. In the first stage, for a target word $w$, we learn two sense-aware encoders that represent the meaning of $w$ in a given sentence selected from a corpus. Next, in the second stage, we learn a sense-aware distance metric that compares the semantic representations of a target word across all of its occurrences in $C_1$ and $C_2$. Experimental results on multiple benchmark datasets for SCD show that our proposed method achieves strong performance in multiple languages. Additionally, our method achieves significant improvements on WiC benchmarks compared to a sense-aware encoder with conventional distance functions. Source code is available at https://github.com/LivNLP/svp-sdml .
Authors: Miao Li, Ming-Bin Chen, Bo Tang, Shengbin Hou, Pengyu Wang, Haiying Deng, Zhiyu Li, Feiyu Xiong, Keming Mao, Peng Cheng, Yi Luo
Abstract: We present NewsBench, a novel evaluation framework to systematically assess the capabilities of Large Language Models (LLMs) for editorial capabilities in Chinese journalism. Our constructed benchmark dataset is focused on four facets of writing proficiency and six facets of safety adherence, and it comprises manually and carefully designed 1,267 test samples in the types of multiple choice questions and short answer questions for five editorial tasks in 24 news domains. To measure performances, we propose different GPT-4 based automatic evaluation protocols to assess LLM generations for short answer questions in terms of writing proficiency and safety adherence, and both are validated by the high correlations with human evaluations. Based on the systematic evaluation framework, we conduct a comprehensive analysis of ten popular LLMs which can handle Chinese. The experimental results highlight GPT-4 and ERNIE Bot as top performers, yet reveal a relative deficiency in journalistic safety adherence in creative writing tasks. Our findings also underscore the need for enhanced ethical guidance in machine-generated journalistic content, marking a step forward in aligning LLMs with journalistic standards and safety considerations.
Authors: Yiqian Yang, Yiqun Duan, Qiang Zhang, Hyejeong Jo, Jinni Zhou, Won Hee Lee, Renjing Xu, Hui Xiong
Abstract: Decoding language from brain dynamics is an important open direction in the realm of brain-computer interface (BCI), especially considering the rapid growth of large language models. Compared to invasive-based signals which require electrode implantation surgery, non-invasive neural signals (e.g. EEG, MEG) have attracted increasing attention considering their safety and generality. However, the exploration is not adequate in three aspects: 1) previous methods mainly focus on EEG but none of the previous works address this problem on MEG with better signal quality; 2) prior works have predominantly used $``teacher-forcing"$ during generative decoding, which is impractical; 3) prior works are mostly $``BART-based"$ not fully auto-regressive, which performs better in other sequence tasks. In this paper, we explore the brain-to-text translation of MEG signals in a speech-decoding formation. Here we are the first to investigate a cross-attention-based ``whisper" model for generating text directly from MEG signals without teacher forcing. Our model achieves impressive BLEU-1 scores of 60.30 and 52.89 without pretraining $\&$ teacher-forcing on two major datasets ($\textit{GWilliams}$ and $\textit{Schoffelen}$). This paper conducts a comprehensive review to understand how speech decoding formation performs on the neural decoding tasks, including pretraining initialization, training $\&$ evaluation set splitting, augmentation, and scaling law. Code is available at https://github.com/NeuSpeech/NeuSpeech1$.
Authors: Zhishuai Li, Xiang Wang, Jingjing Zhao, Sun Yang, Guoqing Du, Xiaoru Hu, Bin Zhang, Yuxiao Ye, Ziyue Li, Rui Zhao, Hangyu Mao
Abstract: Recent advancements in Text-to-SQL (Text2SQL) emphasize stimulating the large language models (LLM) on in-context learning, achieving significant results. Nevertheless, they face challenges when dealing with verbose database information and complex user intentions. This paper presents a two-stage framework to enhance the performance of current LLM-based natural language to SQL systems. We first introduce a novel prompt representation, called reference-enhanced representation, which includes schema information and randomly sampled cell values from tables to instruct LLMs in generating SQL queries. Then, in the first stage, question-SQL pairs are retrieved as few-shot demonstrations, prompting the LLM to generate a preliminary SQL (PreSQL). After that, the mentioned entities in PreSQL are parsed to conduct schema linking, which can significantly compact the useful information. In the second stage, with the linked schema, we simplify the prompt's schema information and instruct the LLM to produce the final SQL. Finally, as the post-refinement module, we propose using cross-consistency across different LLMs rather than self-consistency within a particular LLM. Our methods achieve new SOTA results on the Spider benchmark, with an execution accuracy of 87.6%.
Authors: Zhe Chen, Heyang Liu, Wenyi Yu, Guangzhi Sun, Hongcheng Liu, Ji Wu, Chao Zhang, Yu Wang, Yanfeng Wang
Abstract: Publishing open-source academic video recordings is an emergent and prevalent approach to sharing knowledge online. Such videos carry rich multimodal information including speech, the facial and body movements of the speakers, as well as the texts and pictures in the slides and possibly even the papers. Although multiple academic video datasets have been constructed and released, few of them support both multimodal content recognition and understanding tasks, which is partially due to the lack of high-quality human annotations. In this paper, we propose a novel multimodal, multigenre, and multipurpose audio-visual academic lecture dataset (M$^3$AV), which has almost 367 hours of videos from five sources covering computer science, mathematics, and medical and biology topics. With high-quality human annotations of the slide text and spoken words, in particular high-valued name entities, the dataset can be used for multiple audio-visual recognition and understanding tasks. Evaluations performed on contextual speech recognition, speech synthesis, and slide and script generation tasks demonstrate that the diversity of M$^3$AV makes it a challenging dataset.
Authors: Junyuan Hong, Jinhao Duan, Chenhui Zhang, Zhangheng Li, Chulin Xie, Kelsey Lieberman, James Diffenderfer, Brian Bartoldson, Ajay Jaiswal, Kaidi Xu, Bhavya Kailkhura, Dan Hendrycks, Dawn Song, Zhangyang Wang, Bo Li
Abstract: Compressing high-capability Large Language Models (LLMs) has emerged as a favored strategy for resource-efficient inferences. While state-of-the-art (SoTA) compression methods boast impressive advancements in preserving benign task performance, the potential risks of compression in terms of safety and trustworthiness have been largely neglected. This study conducts the first, thorough evaluation of three (3) leading LLMs using five (5) SoTA compression techniques across eight (8) trustworthiness dimensions. Our experiments highlight the intricate interplay between compression and trustworthiness, revealing some interesting patterns. We find that quantization is currently a more effective approach than pruning in achieving efficiency and trustworthiness simultaneously. For instance, a 4-bit quantized model retains the trustworthiness of its original counterpart, but model pruning significantly degrades trustworthiness, even at 50% sparsity. Moreover, employing quantization within a moderate bit range could unexpectedly improve certain trustworthiness dimensions such as ethics and fairness. Conversely, extreme quantization to very low bit levels (3 bits) tends to reduce trustworthiness significantly. This increased risk cannot be uncovered by looking at benign performance alone, in turn, mandating comprehensive trustworthiness evaluation in practice. These findings culminate in practical recommendations for simultaneously achieving high utility, efficiency, and trustworthiness in LLMs. Code and models are available at https://decoding-comp-trust.github.io.
Authors: Tianqi Zhong, Zhaoyi Li, Quan Wang, Linqi Song, Ying Wei, Defu Lian, Zhendong Mao
Abstract: Compositional generalization, representing the model's ability to generate text with new attribute combinations obtained by recombining single attributes from the training data, is a crucial property for multi-aspect controllable text generation (MCTG) methods. Nonetheless, a comprehensive compositional generalization evaluation benchmark of MCTG is still lacking. We propose CompMCTG, a benchmark encompassing diverse multi-aspect labeled datasets and a crafted three-dimensional evaluation protocol, to holistically evaluate the compositional generalization of MCTG approaches. We observe that existing MCTG works generally confront a noticeable performance drop in compositional testing. To mitigate this issue, we introduce Meta-MCTG, a training framework incorporating meta-learning, where we enable models to learn how to generalize by simulating compositional generalization scenarios in the training phase. We demonstrate the effectiveness of Meta-MCTG through achieving obvious improvement (by at most 3.64%) for compositional testing performance in 94.4% cases.
Authors: Shengding Hu, Yuge Tu, Xu Han, Chaoqun He, Ganqu Cui, Xiang Long, Zhi Zheng, Yewei Fang, Yuxiang Huang, Weilin Zhao, Xinrong Zhang, Zheng Leng Thai, Kaihuo Zhang, Chongyi Wang, Yuan Yao, Chenyang Zhao, Jie Zhou, Jie Cai, Zhongwu Zhai, Ning Ding, Chao Jia, Guoyang Zeng, Dahai Li, Zhiyuan Liu, Maosong Sun
Abstract: The burgeoning interest in developing Large Language Models (LLMs) with up to trillion parameters has been met with concerns regarding resource efficiency and practical expense, particularly given the immense cost of experimentation. This scenario underscores the importance of exploring the potential of Small Language Models (SLMs) as a resource-efficient alternative. In this context, we introduce MiniCPM, specifically the 1.2B and 2.4B non-embedding parameter variants, not only excel in their respective categories but also demonstrate capabilities on par with 7B-13B LLMs. While focusing on SLMs, our approach exhibits scalability in both model and data dimensions for future LLM research. Regarding model scaling, we employ extensive model wind tunnel experiments for stable and optimal scaling. For data scaling, we introduce a Warmup-Stable-Decay (WSD) learning rate scheduler (LRS), conducive to continuous training and domain adaptation. We present an in-depth analysis of the intriguing training dynamics that occurred in the WSD LRS. With WSD LRS, we are now able to efficiently study data-model scaling law without extensive retraining experiments on both axes of model and data, from which we derive the much higher compute optimal data-model ratio than Chinchilla Optimal. Additionally, we introduce MiniCPM family, including MiniCPM-DPO, MiniCPM-MoE and MiniCPM-128K, whose excellent performance further cementing MiniCPM's foundation in diverse SLM applications. MiniCPM models are available publicly at https://github.com/OpenBMB/MiniCPM .
Authors: Yewei Song, Cedric Lothritz, Daniel Tang, Tegawend\'e F. Bissyand\'e, Jacques Klein
Abstract: This paper revisits recent code similarity evaluation metrics, particularly focusing on the application of Abstract Syntax Tree (AST) editing distance in diverse programming languages. In particular, we explore the usefulness of these metrics and compare them to traditional sequence similarity metrics. Our experiments showcase the effectiveness of AST editing distance in capturing intricate code structures, revealing a high correlation with established metrics. Furthermore, we explore the strengths and weaknesses of AST editing distance and prompt-based GPT similarity scores in comparison to BLEU score, execution match, and Jaccard Similarity. We propose, optimize, and publish an adaptable metric that demonstrates effectiveness across all tested languages, representing an enhanced version of Tree Similarity of Edit Distance (TSED).
Authors: Hengyuan Zhang, Yanru Wu, Dawei Li, Sak Yang, Rui Zhao, Yong Jiang, Fei Tan
Abstract: Aligned Large Language Models (LLMs) showcase remarkable versatility, capable of handling diverse real-world tasks. Meanwhile, aligned LLMs are also expected to exhibit speciality, excelling in specific applications. However, fine-tuning with extra data, a common practice to gain speciality, often leads to catastrophic forgetting (CF) of previously acquired versatility, hindering the model's performance across diverse tasks. In response to this challenge, we propose CoFiTune, a coarse to fine framework in an attempt to strike the balance between speciality and versatility. At the coarse-grained level, an empirical tree-search algorithm is utilized to pinpoint and update specific modules that are crucial for speciality, while keeping other parameters frozen; at the fine-grained level, a soft-masking mechanism regulates the update to the LLMs, mitigating the CF issue without harming speciality. In an overall evaluation of both speciality and versatility, CoFiTune consistently outperforms baseline methods across diverse tasks and model scales. Compared to the full-parameter SFT, CoFiTune leads to about 14% versatility improvement and marginal speciality loss on a 13B model. Lastly, based on further analysis, we provide a speculative insight into the information forwarding process in LLMs, which helps explain the effectiveness of the proposed method. The code is available at https://github.com/rattlesnakey/CoFiTune.
Authors: Yongcheng Zeng, Guoqing Liu, Weiyu Ma, Ning Yang, Haifeng Zhang, Jun Wang
Abstract: Fine-tuning pre-trained Large Language Models (LLMs) is essential to align them with human values and intentions. This process often utilizes methods like pairwise comparisons and KL divergence against a reference LLM, focusing on the evaluation of full answers generated by the models. However, the generation of these responses occurs in a token level, following a sequential, auto-regressive fashion. In this paper, we introduce Token-level Direct Preference Optimization (TDPO), a novel approach to align LLMs with human preferences by optimizing policy at the token level. Unlike previous methods, which face challenges in divergence efficiency, TDPO incorporates forward KL divergence constraints for each token, improving alignment and diversity. Utilizing the Bradley-Terry model for a token-based reward system, TDPO enhances the regulation of KL divergence, while preserving simplicity without the need for explicit reward modeling. Experimental results across various text tasks demonstrate TDPO's superior performance in balancing alignment with generation diversity. Notably, fine-tuning with TDPO strikes a better balance than DPO in the controlled sentiment generation and single-turn dialogue datasets, and significantly improves the quality of generated responses compared to both DPO and PPO-based RLHF methods. Our code is open-sourced at https://github.com/Vance0124/Token-level-Direct-Preference-Optimization.
URLs: https://github.com/Vance0124/Token-level-Direct-Preference-Optimization.
Authors: Zhengwei Tao, Ting-En Lin, Xiancai Chen, Hangyu Li, Yuchuan Wu, Yongbin Li, Zhi Jin, Fei Huang, Dacheng Tao, Jingren Zhou
Abstract: Large language models (LLMs) have significantly advanced in various fields and intelligent agent applications. However, current LLMs that learn from human or external model supervision are costly and may face performance ceilings as task complexity and diversity increase. To address this issue, self-evolution approaches that enable LLM to autonomously acquire, refine, and learn from experiences generated by the model itself are rapidly growing. This new training paradigm inspired by the human experiential learning process offers the potential to scale LLMs towards superintelligence. In this work, we present a comprehensive survey of self-evolution approaches in LLMs. We first propose a conceptual framework for self-evolution and outline the evolving process as iterative cycles composed of four phases: experience acquisition, experience refinement, updating, and evaluation. Second, we categorize the evolution objectives of LLMs and LLM-based agents; then, we summarize the literature and provide taxonomy and insights for each module. Lastly, we pinpoint existing challenges and propose future directions to improve self-evolution frameworks, equipping researchers with critical insights to fast-track the development of self-evolving LLMs. Our corresponding GitHub repository is available at https://github.com/AlibabaResearch/DAMO-ConvAI/tree/main/Awesome-Self-Evolution-of-LLM
URLs: https://github.com/AlibabaResearch/DAMO-ConvAI/tree/main/Awesome-Self-Evolution-of-LLM
Authors: Yousef Khan, Ahmed Abdeen Hamed
Abstract: Addressing the global challenge of breast cancer, this research explores the fusion of generative AI, focusing on ChatGPT 3.5 turbo model, and the intricacies of breast cancer risk assessment. The research aims to evaluate ChatGPT's reasoning capabilities, emphasizing its potential to process rules and provide explanations for screening recommendations. The study seeks to bridge the technology gap between intelligent machines and clinicians by demonstrating ChatGPT's unique proficiency in natural language reasoning. The methodology employs a supervised prompt-engineering approach to enforce detailed explanations for ChatGPT's recommendations. Synthetic use cases, generated algorithmically, serve as the testing ground for the encoded rules, evaluating the model's processing prowess. Findings highlight ChatGPT's promising capacity in processing rules comparable to Expert System Shells, with a focus on natural language reasoning. The research introduces the concept of reinforcement explainability, showcasing its potential in elucidating outcomes and facilitating user-friendly interfaces for breast cancer risk assessment.
Authors: Kabir Ahuja, Vidhisha Balachandran, Madhur Panwar, Tianxing He, Noah A. Smith, Navin Goyal, Yulia Tsvetkov
Abstract: Transformers trained on natural language data have been shown to learn its hierarchical structure and generalize to sentences with unseen syntactic structures without explicitly encoding any structural bias. In this work, we investigate sources of inductive bias in transformer models and their training that could cause such generalization behavior to emerge. We extensively experiment with transformer models trained on multiple synthetic datasets and with different training objectives and show that while other objectives e.g. sequence-to-sequence modeling, prefix language modeling, often failed to lead to hierarchical generalization, models trained with the language modeling objective consistently learned to generalize hierarchically. We then conduct pruning experiments to study how transformers trained with the language modeling objective encode hierarchical structure. When pruned, we find joint existence of subnetworks within the model with different generalization behaviors (subnetworks corresponding to hierarchical structure and linear order). Finally, we take a Bayesian perspective to further uncover transformers' preference for hierarchical generalization: We establish a correlation between whether transformers generalize hierarchically on a dataset and whether the simplest explanation of that dataset is provided by a hierarchical grammar compared to regular grammars exhibiting linear generalization.
Authors: Inderjeet Nair, Lu Wang
Abstract: We study the task of conducting structured reasoning as generating a reasoning graph from natural language input using large language models (LLMs). Previous approaches have explored various prompting schemes, yet they suffer from error propagation due to the autoregressive nature and single-pass-based decoding, which lack error correction capability. Additionally, relying solely on a single sample may result in the omission of true nodes and edges. To counter this, we draw inspiration from self-consistency (SC), which involves sampling a diverse set of reasoning chains and taking the majority vote as the final answer. To tackle the substantial challenge of applying SC on generated graphs, we propose MIDGARD (MInimum Description length Guided Aggregation of Reasoning in Directed acyclic graph) that leverages Minimum Description Length (MDL)-based formulation to identify consistent properties among the different graph samples generated by an LLM. This formulation helps reject properties that appear in only a few samples, which are likely to be erroneous, while enabling the inclusion of missing elements without compromising precision. Our method demonstrates superior performance than comparisons across various structured reasoning tasks, including argument structure extraction, explanation graph generation, inferring dependency relations among actions for everyday tasks, and semantic graph generation from natural texts.
Authors: Jianhao Chen, Haoyuan Ouyang, Junyang Ren, Wentao Ding, Wei Hu, Yuzhong Qu
Abstract: Facts extraction is pivotal for constructing knowledge graphs. Recently, the increasing demand for temporal facts in downstream tasks has led to the emergence of the task of temporal fact extraction. In this paper, we specifically address the extraction of temporal facts from natural language text. Previous studies fail to handle the challenge of establishing time-to-fact correspondences in complex sentences. To overcome this hurdle, we propose a timeline-based sentence decomposition strategy using large language models (LLMs) with in-context learning, ensuring a fine-grained understanding of the timeline associated with various facts. In addition, we evaluate the performance of LLMs for direct temporal fact extraction and get unsatisfactory results. To this end, we introduce TSDRE, a method that incorporates the decomposition capabilities of LLMs into the traditional fine-tuning of smaller pre-trained language models (PLMs). To support the evaluation, we construct ComplexTRED, a complex temporal fact extraction dataset. Our experiments show that TSDRE achieves state-of-the-art results on both HyperRED-Temporal and ComplexTRED datasets.
Authors: Jiahao Li, Quan Wang, Licheng Zhang, Guoqing Jin, Zhendong Mao
Abstract: In-context learning (ICL), which promotes inference with several demonstrations, has become a widespread paradigm to stimulate LLM capabilities for downstream tasks. Due to context length constraints, it cannot be further improved in spite of more training data, and general features directly from LLMs in ICL are not adaptive to the specific downstream task. In this paper, we propose a feature-adaptive and data-scalable in-context learning framework (FADS-ICL), which can leverage task-adaptive features to promote inference on the downstream task, with the supervision of beyond-context samples. Specifically, it first extracts general features of beyond-context samples via the LLM with ICL input form one by one, and introduces a task-specific modulator to perform feature refinement and prediction after fitting a specific downstream task. We conduct extensive experiments on FADS-ICL under varying data settings (4$\sim$128 shots) and LLM scale (0.8$\sim$70B) settings. Experimental results show that FADS-ICL consistently outperforms previous state-of-the-art methods by a significant margin under all settings, verifying the effectiveness and superiority of FADS-ICL. For example, under the 1.5B and 32 shots setting, FADS-ICL can achieve \textbf{+14.3} average accuracy from feature adaptation over vanilla ICL on 10 datasets, with \textbf{+6.2} average accuracy over the previous state-of-the-art method, and the performance can further improve with increasing training data. Code and data are publicly available at \url{https://github.com/jiahaozhenbang/FADS-ICL}.
Authors: Amr Keleg, Walid Magdy, Sharon Goldwater
Abstract: On annotating multi-dialect Arabic datasets, it is common to randomly assign the samples across a pool of native Arabic speakers. Recent analyses recommended routing dialectal samples to native speakers of their respective dialects to build higher-quality datasets. However, automatically identifying the dialect of samples is hard. Moreover, the pool of annotators who are native speakers of specific Arabic dialects might be scarce. Arabic Level of Dialectness (ALDi) was recently introduced as a quantitative variable that measures how sentences diverge from Standard Arabic. On randomly assigning samples to annotators, we hypothesize that samples of higher ALDi scores are harder to label especially if they are written in dialects that the annotators do not speak. We test this by analyzing the relation between ALDi scores and the annotators' agreement, on 15 public datasets having raw individual sample annotations for various sentence-classification tasks. We find strong evidence supporting our hypothesis for 11 of them. Consequently, we recommend prioritizing routing samples of high ALDi scores to native speakers of each sample's dialect, for which the dialect could be automatically identified at higher accuracies.
Authors: Chen Huang, Yiping Jin, Ilija Ilievski, Wenqiang Lei, Jiancheng Lv
Abstract: Human annotation is a time-consuming task that requires a significant amount of effort. To address this issue, interactive data annotation utilizes an annotation model to provide suggestions for humans to approve or correct. However, annotation models trained with limited labeled data are prone to generating incorrect suggestions, leading to extra human correction effort. To tackle this challenge, we propose Araida, an analogical reasoning-based approach that enhances automatic annotation accuracy in the interactive data annotation setting and reduces the need for human corrections. Araida involves an error-aware integration strategy that dynamically coordinates an annotation model and a k-nearest neighbors (KNN) model, giving more importance to KNN's predictions when predictions from the annotation model are deemed inaccurate. Empirical studies demonstrate that Araida is adaptable to different annotation tasks and models. On average, it reduces human correction labor by 11.02% compared to vanilla interactive data annotation methods.
Authors: Yue Chen, Chen Huang, Yang Deng, Wenqiang Lei, Dingnan Jin, Jia Liu, Tat-Seng Chua
Abstract: Equipping a conversational search engine with strategies regarding when to ask clarification questions is becoming increasingly important across various domains. Attributing to the context understanding capability of LLMs and their access to domain-specific sources of knowledge, LLM-based clarification strategies feature rapid transfer to various domains in a post-hoc manner. However, they still struggle to deliver promising performance on unseen domains, struggling to achieve effective domain transferability. We take the first step to investigate this issue and existing methods tend to produce one-size-fits-all strategies across diverse domains, limiting their search effectiveness. In response, we introduce a novel method, called Style, to achieve effective domain transferability. Our experimental results indicate that Style bears strong domain transferability, resulting in an average search performance improvement of ~10% on four unseen domains.
Authors: Tong Zhang, Peixin Qin, Yang Deng, Chen Huang, Wenqiang Lei, Junhong Liu, Dingnan Jin, Hongru Liang, Tat-Seng Chua
Abstract: Large language models (LLMs) are increasingly used to meet user information needs, but their effectiveness in dealing with user queries that contain various types of ambiguity remains unknown, ultimately risking user trust and satisfaction. To this end, we introduce CLAMBER, a benchmark for evaluating LLMs using a well-organized taxonomy. Building upon the taxonomy, we construct ~12K high-quality data to assess the strengths, weaknesses, and potential risks of various off-the-shelf LLMs. Our findings indicate the limited practical utility of current LLMs in identifying and clarifying ambiguous user queries, even enhanced by chain-of-thought (CoT) and few-shot prompting. These techniques may result in overconfidence in LLMs and yield only marginal enhancements in identifying ambiguity. Furthermore, current LLMs fall short in generating high-quality clarifying questions due to a lack of conflict resolution and inaccurate utilization of inherent knowledge. In this paper, CLAMBER presents a guidance and promotes further research on proactive and trustworthy LLMs. Our dataset is available at https://github.com/zt991211/CLAMBER
Authors: Bilgehan Sel, Priya Shanmugasundaram, Mohammad Kachuee, Kun Zhou, Ruoxi Jia, Ming Jin
Abstract: Large Language Models (LLMs) have shown remarkable capabilities in tasks such as summarization, arithmetic reasoning, and question answering. However, they encounter significant challenges in the domain of moral reasoning and ethical decision-making, especially in complex scenarios with multiple stakeholders. This paper introduces the Skin-in-the-Game (SKIG) framework, aimed at enhancing moral reasoning in LLMs by exploring decisions' consequences from multiple stakeholder perspectives. Central to SKIG's mechanism is simulating accountability for actions, which, alongside empathy exercises and risk assessment, is pivotal to its effectiveness. We validate SKIG's performance across various moral reasoning benchmarks with proprietary and opensource LLMs, and investigate its crucial components through extensive ablation analyses.
Authors: Chan-Jan Hsu, Yi-Chang Chen, Feng-Ting Liao, Pei-Chen Ho, Yu-Hsiang Wang, Po-Chun Hsu, Da-shan Shiu
Abstract: We introduce "Generative Fusion Decoding" (GFD), a novel shallow fusion framework, utilized to integrate Large Language Models (LLMs) into multi-modal text recognition systems such as automatic speech recognition (ASR) and optical character recognition (OCR). We derive the formulas necessary to enable GFD to operate across mismatched token spaces of different models by mapping text token space to byte token space, enabling seamless fusion during the decoding process. The framework is plug-and-play, compatible with various auto-regressive models, and does not require re-training for feature alignment, thus overcoming limitations of previous fusion techniques. We highlight three main advantages of GFD: First, by simplifying the complexity of aligning different model sample spaces, GFD allows LLMs to correct errors in tandem with the recognition model, reducing computation latencies. Second, the in-context learning ability of LLMs is fully capitalized by GFD, increasing robustness in long-form speech recognition and instruction aware speech recognition. Third, GFD enables fusing recognition models deficient in Chinese text recognition with LLMs extensively trained on Chinese. Our evaluation demonstrates that GFD significantly improves performance in ASR and OCR tasks, with ASR reaching state-of-the-art in the NTUML2021 benchmark. GFD provides a significant step forward in model integration, offering a unified solution that could be widely applicable to leveraging existing pre-trained models through step by step fusion.
Authors: Abhishek Kumar, Sarfaroz Yunusov, Ali Emami
Abstract: Research on Large Language Models (LLMs) has often neglected subtle biases that, although less apparent, can significantly influence the models' outputs toward particular social narratives. This study addresses two such biases within LLMs: representative bias, which denotes a tendency of LLMs to generate outputs that mirror the experiences of certain identity groups, and affinity bias, reflecting the models' evaluative preferences for specific narratives or viewpoints. We introduce two novel metrics to measure these biases: the Representative Bias Score (RBS) and the Affinity Bias Score (ABS), and present the Creativity-Oriented Generation Suite (CoGS), a collection of open-ended tasks such as short story writing and poetry composition, designed with customized rubrics to detect these subtle biases. Our analysis uncovers marked representative biases in prominent LLMs, with a preference for identities associated with being white, straight, and men. Furthermore, our investigation of affinity bias reveals distinctive evaluative patterns within each model, akin to `bias fingerprints'. This trend is also seen in human evaluators, highlighting a complex interplay between human and machine bias perceptions.
Authors: Brendan Park, Madeline Janecek, Naser Ezzati-Jivan, Yifeng Li, Ali Emami
Abstract: Large Language Models (LLMs) have demonstrated remarkable success in tasks like the Winograd Schema Challenge (WSC), showcasing advanced textual common-sense reasoning. However, applying this reasoning to multimodal domains, where understanding text and images together is essential, remains a substantial challenge. To address this, we introduce WinoVis, a novel dataset specifically designed to probe text-to-image models on pronoun disambiguation within multimodal contexts. Utilizing GPT-4 for prompt generation and Diffusion Attentive Attribution Maps (DAAM) for heatmap analysis, we propose a novel evaluation framework that isolates the models' ability in pronoun disambiguation from other visual processing challenges. Evaluation of successive model versions reveals that, despite incremental advancements, Stable Diffusion 2.0 achieves a precision of 56.7% on WinoVis, only marginally surpassing random guessing. Further error analysis identifies important areas for future research aimed at advancing text-to-image models in their ability to interpret and interact with the complex visual world.
Authors: Abhishek Kumar, Robert Morabito, Sanzhar Umbet, Jad Kabbara, Ali Emami
Abstract: As the use of Large Language Models (LLMs) becomes more widespread, understanding their self-evaluation of confidence in generated responses becomes increasingly important as it is integral to the reliability of the output of these models. We introduce the concept of Confidence-Probability Alignment, that connects an LLM's internal confidence, quantified by token probabilities, to the confidence conveyed in the model's response when explicitly asked about its certainty. Using various datasets and prompting techniques that encourage model introspection, we probe the alignment between models' internal and expressed confidence. These techniques encompass using structured evaluation scales to rate confidence, including answer options when prompting, and eliciting the model's confidence level for outputs it does not recognize as its own. Notably, among the models analyzed, OpenAI's GPT-4 showed the strongest confidence-probability alignment, with an average Spearman's $\hat{\rho}$ of 0.42, across a wide range of tasks. Our work contributes to the ongoing efforts to facilitate risk assessment in the application of LLMs and to further our understanding of model trustworthiness.
Authors: Arle Lommel, Serge Gladkoff, Alan Melby, Sue Ellen Wright, Ingemar Strandvik, Katerina Gasova, Angelika Vaasa, Andy Benzo, Romina Marazzato Sparano, Monica Foresi, Johani Innis, Lifeng Han, Goran Nenadic
Abstract: The year 2024 marks the 10th anniversary of the Multidimensional Quality Metrics (MQM) framework for analytic translation quality evaluation. The MQM error typology has been widely used by practitioners in the translation and localization industry and has served as the basis for many derivative projects. The annual Conference on Machine Translation (WMT) shared tasks on both human and automatic translation quality evaluations used the MQM error typology. The metric stands on two pillars: error typology and the scoring model. The scoring model calculates the quality score from annotation data, detailing how to convert error type and severity counts into numeric scores to determine if the content meets specifications. Previously, only the raw scoring model had been published. This April, the MQM Council published the Linear Calibrated Scoring Model, officially presented herein, along with the Non-Linear Scoring Model, which had not been published before. This paper details the latest MQM developments and presents a universal approach to translation quality measurement across three sample size ranges. It also explains why Statistical Quality Control should be used for very small sample sizes, starting from a single sentence.
Authors: Gregory Camilli
Abstract: Natural language processing (NLP) is rapidly developing for applications in educational assessment. In this paper, I describe an NLP-based procedure that can be used to support subject matter experts in establishing a crosswalk between item specifications and content standards. This paper extends recent work by proposing and demonstrating the use of multivariate similarity based on embedding vectors for sentences or texts. In particular, a hybrid regression procedure is demonstrated for establishing the match of each content standard to multiple item specifications. The procedure is used to evaluate the match of the Common Core State Standards (CCSS) for mathematics at grade 4 to the corresponding item specifications for the 2026 National Assessment of Educational Progress (NAEP).
Authors: Yash Sarrof, Yana Veitsman, Michael Hahn
Abstract: Recently, recurrent models based on linear state space models (SSMs) have shown promising performance in language modeling (LM), competititve with transformers. However, there is little understanding of the in-principle abilities of such models, which could provide useful guidance to the search for better LM architectures. We present a comprehensive theoretical study of the capacity of such SSMs as it compares to that of transformers and traditional RNNs. We find that SSMs and transformers have overlapping but distinct strengths. In star-free state tracking, SSMs implement straightforward and exact solutions to problems that transformers struggle to represent exactly. They can also model bounded hierarchical structure with optimal memory even without simulating a stack. On the other hand, we identify a design choice in current SSMs that limits their expressive power. We discuss implications for SSM and LM research, and verify results empirically on a recent SSM, Mamba.
Authors: Xiao Peng, Tao Liu, Ying Wang
Abstract: Large language models (LLMs) like ChatGPT, Gemini, or LLaMA have been trending recently, demonstrating considerable advancement and generalizability power in countless domains. However, LLMs create an even bigger black box exacerbating opacity, with interpretability limited to few approaches. The uncertainty and opacity embedded in LLMs' nature restrict their application in high-stakes domains like financial fraud, phishing, etc. Current approaches mainly rely on traditional textual classification with posterior interpretable algorithms, suffering from attackers who may create versatile adversarial samples to break the system's defense, forcing users to make trade-offs between efficiency and robustness. To address this issue, we propose a novel cascading framework called Genshin (General Shield for Natural Language Processing with Large Language Models), utilizing LLMs as defensive one-time plug-ins. Unlike most applications of LLMs that try to transform text into something new or structural, Genshin uses LLMs to recover text to its original state. Genshin aims to combine the generalizability of the LLM, the discrimination of the median model, and the interpretability of the simple model. Our experiments on the task of sentimental analysis and spam detection have shown fatal flaws of the current median models and exhilarating results on LLMs' recovery ability, demonstrating that Genshin is both effective and efficient. In our ablation study, we unearth several intriguing observations. Utilizing the LLM defender, a tool derived from the 4th paradigm, we have reproduced BERT's 15% optimal mask rate results in the 3rd paradigm of NLP. Additionally, when employing the LLM as a potential adversarial tool, attackers are capable of executing effective attacks that are nearly semantically lossless.
Authors: Peter Devine
Abstract: Training Large Language Models (LLMs) with Reinforcement Learning from AI Feedback (RLAIF) aligns model outputs more closely with human preferences. This involves an evaluator model ranking multiple candidate responses to user prompts. However, the rankings from popular evaluator models such as GPT-4 can be inconsistent. We propose the Repeat Ranking method - where we evaluate the same responses multiple times and train only on those responses which are consistently ranked. Using 2,714 prompts in 62 languages, we generated responses from 7 top multilingual LLMs and had GPT-4 rank them five times each. Evaluating on MT-Bench chat benchmarks in six languages, our method outperformed the standard practice of training on all available prompts. Our work highlights the quality versus quantity trade-off in RLAIF dataset generation and offers a stackable strategy for enhancing dataset and thus model quality.
Authors: Renzhi Wang, Piji Li
Abstract: Model editing aims to efficiently alter the behavior of Large Language Models (LLMs) within a desired scope, while ensuring no adverse impact on other inputs. Recent years have witnessed various model editing methods been proposed. However, these methods either exhibit poor overall performance or struggle to strike a balance between generalization and locality. We propose MEMoE, a model editing adapter utilizing a Mixture of Experts (MoE) architecture with a knowledge anchor routing strategy. MEMoE updates knowledge using a bypass MoE structure, keeping the original parameters unchanged to preserve the general ability of LLMs. And, the knowledge anchor routing ensures that inputs requiring similar knowledge are routed to the same expert, thereby enhancing the generalization of the updated knowledge. Experimental results show the superiority of our approach over both batch editing and sequential batch editing tasks, exhibiting exceptional overall performance alongside outstanding balance between generalization and locality. Our code will be available.
Authors: Dingkang Yang, Jinjie Wei, Dongling Xiao, Shunli Wang, Tong Wu, Gang Li, Mingcheng Li, Shuaibing Wang, Jiawei Chen, Yue Jiang, Qingyao Xu, Ke Li, Peng Zhai, Lihua Zhang
Abstract: Developing intelligent pediatric consultation systems offers promising prospects for improving diagnostic efficiency, especially in China, where healthcare resources are scarce. Despite recent advances in Large Language Models (LLMs) for Chinese medicine, their performance is sub-optimal in pediatric applications due to inadequate instruction data and vulnerable training procedures. To address the above issues, this paper builds PedCorpus, a high-quality dataset of over 300,000 multi-task instructions from pediatric textbooks, guidelines, and knowledge graph resources to fulfil diverse diagnostic demands. Upon well-designed PedCorpus, we propose PediatricsGPT, the first Chinese pediatric LLM assistant built on a systematic and robust training pipeline. In the continuous pre-training phase, we introduce a hybrid instruction pre-training mechanism to mitigate the internal-injected knowledge inconsistency of LLMs for medical domain adaptation. Immediately, the full-parameter Supervised Fine-Tuning (SFT) is utilized to incorporate the general medical knowledge schema into the models. After that, we devise a direct following preference optimization to enhance the generation of pediatrician-like humanistic responses. In the parameter-efficient secondary SFT phase, a mixture of universal-specific experts strategy is presented to resolve the competency conflict between medical generalist and pediatric expertise mastery. Extensive results based on the metrics, GPT-4, and doctor evaluations on distinct doctor downstream tasks show that PediatricsGPT consistently outperforms previous Chinese medical LLMs. Our model and dataset will be open-source for community development.
Authors: Ge Zhang, Scott Qu, Jiaheng Liu, Chenchen Zhang, Chenghua Lin, Chou Leuang Yu, Danny Pan, Esther Cheng, Jie Liu, Qunshu Lin, Raven Yuan, Tuney Zheng, Wei Pang, Xinrun Du, Yiming Liang, Yinghao Ma, Yizhi Li, Ziyang Ma, Bill Lin, Emmanouil Benetos, Huan Yang, Junting Zhou, Kaijing Ma, Minghao Liu, Morry Niu, Noah Wang, Quehry Que, Ruibo Liu, Sine Liu, Shawn Guo, Soren Gao, Wangchunshu Zhou, Xinyue Zhang, Yizhi Zhou, Yubo Wang, Yuelin Bai, Yuhan Zhang, Yuxiang Zhang, Zenith Wang, Zhenzhu Yang, Zijian Zhao, Jiajun Zhang, Wanli Ouyang, Wenhao Huang, Wenhu Chen
Abstract: Large Language Models (LLMs) have made great strides in recent years to achieve unprecedented performance across different tasks. However, due to commercial interest, the most competitive models like GPT, Gemini, and Claude have been gated behind proprietary interfaces without disclosing the training details. Recently, many institutions have open-sourced several strong LLMs like LLaMA-3, comparable to existing closed-source LLMs. However, only the model's weights are provided with most details (e.g., intermediate checkpoints, pre-training corpus, and training code, etc.) being undisclosed. To improve the transparency of LLMs, the research community has formed to open-source truly open LLMs (e.g., Pythia, Amber, OLMo), where more details (e.g., pre-training corpus and training code) are being provided. These models have greatly advanced the scientific study of these large models including their strengths, weaknesses, biases and risks. However, we observe that the existing truly open LLMs on reasoning, knowledge, and coding tasks are still inferior to existing state-of-the-art LLMs with similar model sizes. To this end, we open-source MAP-Neo, a highly capable and transparent bilingual language model with 7B parameters trained from scratch on 4.5T high-quality tokens. Our MAP-Neo is the first fully open-sourced bilingual LLM with comparable performance compared to existing state-of-the-art LLMs. Moreover, we open-source all details to reproduce our MAP-Neo, where the cleaned pre-training corpus, data cleaning pipeline, checkpoints, and well-optimized training/evaluation framework are provided. Finally, we hope our MAP-Neo will enhance and strengthen the open research community and inspire more innovations and creativities to facilitate the further improvements of LLMs.
Authors: Mithilesh Vaidya, Binaya Kumar Sahoo, Preeti Rao
Abstract: Reading fluency assessment is a critical component of literacy programmes, serving to guide and monitor early education interventions. Given the resource intensive nature of the exercise when conducted by teachers, the development of automatic tools that can operate on audio recordings of oral reading is attractive as an objective and highly scalable solution. Multiple complex aspects such as accuracy, rate and expressiveness underlie human judgements of reading fluency. In this work, we investigate end-to-end modeling on a training dataset of children's audio recordings of story texts labeled by human experts. The pre-trained wav2vec2.0 model is adopted due its potential to alleviate the challenges from the limited amount of labeled data. We report the performance of a number of system variations on the relevant measures, and also probe the learned embeddings for lexical and acoustic-prosodic features known to be important to the perception of reading fluency.
Authors: Lavanya Prahallad, Radhika Mamidi
Abstract: Gender bias in machine translation (MT) sys- tems poses a significant challenge to achieving accurate and inclusive translations. This paper examines gender bias in machine translation systems for languages such as Telugu and Kan- nada from the Dravidian family, analyzing how gender inflections affect translation accuracy and neutrality using Google Translate and Chat- GPT. It finds that while plural forms can reduce bias, individual-centric sentences often main- tain the bias due to historical stereotypes. The study evaluates the Chain of Thought process- ing, noting significant bias mitigation from 80% to 4% in Telugu and from 40% to 0% in Kan- nada. It also compares Telugu and Kannada translations, emphasizing the need for language specific strategies to address these challenges and suggesting directions for future research to enhance fairness in both data preparation and prompts during inference.
Authors: Jeiyoon Park, Chanjun Park, Heuiseok Lim
Abstract: With the recent introduction of Assistants API, it is expected that document-based language models will be actively used in various domains, especially Role-playing. However, a key challenge lies in utilizing protagonist's persona: Assistants API often fails to achieve with its search because the information extraction part is different each time and it often omits important information such as protagonist's backstory or relationships. It is hard to maintain a consistent persona simply by using the persona document as input to the Assistants API. To address the challenge of achieving stable persona consistency, we propose CharacterGPT, a novel persona reconstruction framework to alleviate the shortcomings of the Assistants API. Our method involves Character Persona Training (CPT), an effective persona rebuilding process that updates the character persona by extracting the character's traits from given summary of the novel for each character as if the story in a novel progresses. In our experiments, we ask each character to take the Big Five Inventory personality test in various settings and analyze the results. To assess whether it can think outside the box, we let each character generate short novels. Extensive experiments and human evaluation demonstrate that CharacterGPT presents new possibilities for role-playing agent research.
Authors: Jiahui Xu, Feng Jiang, Anningzhe Gao, Haizhou Li
Abstract: The advancement of large language models (LLMs) has propelled the development of dialogue systems. Unlike the popular ChatGPT-like assistant model, which only satisfies the user's preferences, task-oriented dialogue systems have also faced new requirements and challenges in the broader business field. They are expected to provide correct responses at each dialogue turn, at the same time, achieve the overall goal defined by the task. By understanding rhetorical structures and topic structures via topic segmentation and discourse parsing, a dialogue system may do a better planning to achieve both objectives. However, while both structures belong to discourse structure in linguistics, rhetorical structure and topic structure are mostly modeled separately or with one assisting the other in the prior work. The interaction between these two structures has not been considered for joint modeling and mutual learning. Furthermore, unsupervised learning techniques to achieve the above are not well explored. To fill this gap, we propose an unsupervised mutual learning framework of two structures leveraging the global and local connections between them. We extend the topic modeling between non-adjacent discourse units to ensure global structural relevance with rhetorical structures. We also incorporate rhetorical structures into the topic structure through a graph neural network model to ensure local coherence consistency. Finally, we utilize the similarity between the two fused structures for mutual learning. The experimental results demonstrate that our methods outperform all strong baselines on two dialogue rhetorical datasets (STAC and Molweni), as well as dialogue topic datasets (Doc2Dial and TIAGE). We provide our code at https://github.com/Jeff-Sue/URT.
Authors: Wei Zhong, Manasa Bharadwaj
Abstract: Speculative decoding (SD) has attracted a significant amount of research attention due to the substantial speedup it can achieve for LLM inference. However, despite the high speedups they offer, speculative decoding methods often achieve optimal performance on high-end devices or with a substantial GPU memory overhead. Given limited memory and the necessity of quantization, a high-performing model on a high-end GPU can slow down by up to 7 times. To this end, we propose Skippy Simultaneous Speculative Decoding (or S3D), a cost-effective self-speculative SD method based on simultaneous multi-token decoding and mid-layer skipping. When compared against recent effective open-source SD systems, our method has achieved one of the top performance-memory ratios while requiring minimal architecture changes and training data. Leveraging our memory efficiency, we created a smaller yet more effective SD model based on Phi-3. It is 1.4 to 2 times faster than the quantized EAGLE model and operates in half-precision while using less VRAM.
Authors: Nikhil Kumar Rajput, Bhavya Ahuja Grover, Vipin Kumar Rathi, Riya Bansal
Abstract: The COVID-19 epidemic has had a great impact on social media conversation, especially on sites like Twitter, which has emerged as a hub for public reaction and information sharing. This paper deals by analyzing a vast dataset of Twitter messages related to this disease, starting from January 2020. Two approaches were used: a statistical analysis of word frequencies and a sentiment analysis to gauge user attitudes. Word frequencies are modeled using unigrams, bigrams, and trigrams, with power law distribution as the fitting model. The validity of the model is confirmed through metrics like Sum of Squared Errors (SSE), R-squared ($R^2$), and Root Mean Squared Error (RMSE). High $R^2$ and low SSE/RMSE values indicate a good fit for the model. Sentiment analysis is conducted to understand the general emotional tone of Twitter users messages. The results reveal that a majority of tweets exhibit neutral sentiment polarity, with only 2.57\% expressing negative polarity.
Authors: Ahan Gupta, Hao Guo, Yueming Yuan, Yanqi Zhou, Charith Mendis
Abstract: Many efficient $\textit{approximate}$ self-attention techniques have become prevalent since the inception of the transformer architecture. Two popular classes of these techniques are low-rank and kernel methods. Each of these methods has its strengths. We observe these strengths synergistically complement each other and exploit them to fuse low-rank and kernel methods, producing a new class of transformers: FLuRKA ($\textbf{F}$ast $\textbf{L}$ow-$\textbf{R}$ank & $\textbf{K}$ernel$ \textbf{A}$ttention). FLuRKA are highly $\textit{training-efficient}$ with faster model speeds $\textit{and}$ similar model qualities compared to constituent low-rank and kernel methods. We theoretically and empirically evaluate the speed and quality of FLuRKA. Our model speed analysis posits a variety of parameter configurations where FLuRKA exhibit speedups over low-rank and kernel approximations and our model quality analysis bounds the error of FLuRKA with respect to full-attention. Empirically, we instantiate three FLuRKA variants which experience speedups of up to 3.3x and 1.7x over low-rank and kernel methods respectively. This translates to speedups of up to 20x over models with flash-attention. Across a diverse set of tasks spanning language modeling, language understanding, long sequence modeling, machine translation, and image classification, FLuRKA achieve comparable accuracy with underlying low-rank and kernel approximations, occasionally surpassing both.
Authors: Yupei Liu, Yuqi Jia, Runpeng Geng, Jinyuan Jia, Neil Zhenqiang Gong
Abstract: A prompt injection attack aims to inject malicious instruction/data into the input of an LLM-Integrated Application such that it produces results as an attacker desires. Existing works are limited to case studies. As a result, the literature lacks a systematic understanding of prompt injection attacks and their defenses. We aim to bridge the gap in this work. In particular, we propose a framework to formalize prompt injection attacks. Existing attacks are special cases in our framework. Moreover, based on our framework, we design a new attack by combining existing ones. Using our framework, we conduct a systematic evaluation on 5 prompt injection attacks and 10 defenses with 10 LLMs and 7 tasks. Our work provides a common benchmark for quantitatively evaluating future prompt injection attacks and defenses. To facilitate research on this topic, we make our platform public at https://github.com/liu00222/Open-Prompt-Injection.
Authors: Deqing Fu, Tian-Qi Chen, Robin Jia, Vatsal Sharan
Abstract: Transformers excel at in-context learning (ICL) -- learning from demonstrations without parameter updates -- but how they do so remains a mystery. Recent work suggests that Transformers may internally run Gradient Descent (GD), a first-order optimization method, to perform ICL. In this paper, we instead demonstrate that Transformers learn to approximate higher-order optimization methods for ICL. For in-context linear regression, Transformers share a similar convergence rate as Iterative Newton's Method; both are exponentially faster than GD. Empirically, predictions from successive Transformer layers closely match different iterations of Newton's Method linearly, with each middle layer roughly computing 3 iterations; thus, Transformers and Newton's method converge at roughly the same rate. In contrast, Gradient Descent converges exponentially more slowly. We also show that Transformers can learn in-context on ill-conditioned data, a setting where Gradient Descent struggles but Iterative Newton succeeds. Finally, to corroborate our empirical findings, we prove that Transformers can implement $k$ iterations of Newton's method with $k + \mathcal{O}(1)$ layers.
Authors: Wenhao Xu, Rongtao Xu, Changwei Wang, Shibiao Xu, Li Guo, Man Zhang, Xiaopeng Zhang
Abstract: Recently, CLIP has found practical utility in the domain of pixel-level zero-shot segmentation tasks. The present landscape features two-stage methodologies beset by issues such as intricate pipelines and elevated computational costs. While current one-stage approaches alleviate these concerns and incorporate Visual Prompt Training (VPT) to uphold CLIP's generalization capacity, they still fall short in fully harnessing CLIP's potential for pixel-level unseen class demarcation and precise pixel predictions. To further stimulate CLIP's zero-shot dense prediction capability, we propose SPT-SEG, a one-stage approach that improves CLIP's adaptability from image to pixel. Specifically, we initially introduce Spectral Prompt Tuning (SPT), incorporating spectral prompts into the CLIP visual encoder's shallow layers to capture structural intricacies of images, thereby enhancing comprehension of unseen classes. Subsequently, we introduce the Spectral Guided Decoder (SGD), utilizing both high and low-frequency information to steer the network's spatial focus towards more prominent classification features, enabling precise pixel-level prediction outcomes. Through extensive experiments on two public datasets, we demonstrate the superiority of our method over state-of-the-art approaches, performing well across all classes and particularly excelling in handling unseen classes. Code is available at:https://github.com/clearxu/SPT.
Authors: Max Ku, Dongfu Jiang, Cong Wei, Xiang Yue, Wenhu Chen
Abstract: In the rapidly advancing field of conditional image generation research, challenges such as limited explainability lie in effectively evaluating the performance and capabilities of various models. This paper introduces VIEScore, a Visual Instruction-guided Explainable metric for evaluating any conditional image generation tasks. VIEScore leverages general knowledge from Multimodal Large Language Models (MLLMs) as the backbone and does not require training or fine-tuning. We evaluate VIEScore on seven prominent tasks in conditional image tasks and found: (1) VIEScore (GPT4-o) achieves a high Spearman correlation of 0.4 with human evaluations, while the human-to-human correlation is 0.45. (2) VIEScore (with open-source MLLM) is significantly weaker than GPT-4o and GPT-4v in evaluating synthetic images. (3) VIEScore achieves a correlation on par with human ratings in the generation tasks but struggles in editing tasks. With these results, we believe VIEScore shows its great potential to replace human judges in evaluating image synthesis tasks.
Authors: Haochen Li, Xin Zhou, Zhiqi Shen
Abstract: In code search, the Generation-Augmented Retrieval (GAR) framework, which generates exemplar code snippets to augment queries, has emerged as a promising strategy to address the principal challenge of modality misalignment between code snippets and natural language queries, particularly with the demonstrated code generation capabilities of Large Language Models (LLMs). Nevertheless, our preliminary investigations indicate that the improvements conferred by such an LLM-augmented framework are somewhat constrained. This limitation could potentially be ascribed to the fact that the generated codes, albeit functionally accurate, frequently display a pronounced stylistic deviation from the ground truth code in the codebase. In this paper, we extend the foundational GAR framework and propose a simple yet effective method that additionally Rewrites the Code (ReCo) within the codebase for style normalization. Experimental results demonstrate that ReCo significantly boosts retrieval accuracy across sparse (up to 35.7%), zero-shot dense (up to 27.6%), and fine-tuned dense (up to 23.6%) retrieval settings in diverse search scenarios. To further elucidate the advantages of ReCo and stimulate research in code style normalization, we introduce Code Style Similarity, the first metric tailored to quantify stylistic similarities in code. Notably, our empirical findings reveal the inadequacy of existing metrics in capturing stylistic nuances. The source code and data are available at \url{https://github.com/Alex-HaochenLi/ReCo}.
Authors: Yuhao Wang, Yusheng Liao, Heyang Liu, Hongcheng Liu, Yu Wang, Yanfeng Wang
Abstract: Recent advancements in Multimodal Large Language Models (MLLMs) have demonstrated exceptional capabilities in visual perception and understanding. However, these models also suffer from hallucinations, which limit their reliability as AI systems. We believe that these hallucinations are partially due to the models' struggle with understanding what they can and cannot perceive from images, a capability we refer to as self-awareness in perception. Despite its importance, this aspect of MLLMs has been overlooked in prior studies. In this paper, we aim to define and evaluate the self-awareness of MLLMs in perception. To do this, we first introduce the knowledge quadrant in perception, which helps define what MLLMs know and do not know about images. Using this framework, we propose a novel benchmark, the Self-Awareness in Perception for MLLMs (MM-SAP), specifically designed to assess this capability. We apply MM-SAP to a variety of popular MLLMs, offering a comprehensive analysis of their self-awareness and providing detailed insights. The experiment results reveal that current MLLMs possess limited self-awareness capabilities, pointing to a crucial area for future advancement in the development of trustworthy MLLMs. Code and data are available at https://github.com/YHWmz/MM-SAP.
Authors: Zhumin Chu, Qingyao Ai, Yiteng Tu, Haitao Li, Yiqun Liu
Abstract: The impressive performance of large language models (LLMs) has attracted considerable attention from the academic and industrial communities. Besides how to construct and train LLMs, how to effectively evaluate and compare the capacity of LLMs has also been well recognized as an important yet difficult problem. Existing paradigms rely on either human annotators or model-based evaluators to evaluate the performance of LLMs on different tasks. However, these paradigms often suffer from high cost, low generalizability, and inherited biases in practice, which make them incapable of supporting the sustainable development of LLMs in long term. In order to address these issues, inspired by the peer review systems widely used in academic publication process, we propose a novel framework that can automatically evaluate LLMs through a peer-review process. Specifically, for the evaluation of a specific task, we first construct a small qualification exam to select "reviewers" from a couple of powerful LLMs. Then, to actually evaluate the "submissions" written by different candidate LLMs, i.e., the evaluatees, we use the reviewer LLMs to rate or compare the submissions. The final ranking of evaluatee LLMs is generated based on the results provided by all reviewers. We conducted extensive experiments on text summarization tasks with eleven LLMs including GPT-4. The results demonstrate the existence of biasness when evaluating using a single LLM. Also, our PRE model outperforms all the baselines, illustrating the effectiveness of the peer review mechanism.
Authors: Vassilis Papadopoulos, J\'er\'emie Wenger, Cl\'ement Hongler
Abstract: We study the probabilistic modeling performed by Autoregressive Large Language Models (LLMs) through the angle of time directionality, addressing a question first raised in (Shannon, 1951). For large enough models, we empirically find a time asymmetry in their ability to learn natural language: a difference in the average log-perplexity when trying to predict the next token versus when trying to predict the previous one. This difference is at the same time subtle and very consistent across various modalities (language, model size, training time, ...). Theoretically, this is surprising: from an information-theoretic point of view, there should be no such difference. We provide a theoretical framework to explain how such an asymmetry can appear from sparsity and computational complexity considerations, and outline a number of perspectives opened by our results.
Authors: Chujie Zheng, Fan Yin, Hao Zhou, Fandong Meng, Jie Zhou, Kai-Wei Chang, Minlie Huang, Nanyun Peng
Abstract: Prepending model inputs with safety prompts is a common practice for safeguarding large language models (LLMs) against queries with harmful intents. However, the underlying working mechanisms of safety prompts have not been unraveled yet, restricting the possibility of automatically optimizing them to improve LLM safety. In this work, we investigate how LLMs' behavior (i.e., complying with or refusing user queries) is affected by safety prompts from the perspective of model representation. We find that in the representation space, the input queries are typically moved by safety prompts in a "higher-refusal" direction, in which models become more prone to refusing to provide assistance, even when the queries are harmless. On the other hand, LLMs are naturally capable of distinguishing harmful and harmless queries without safety prompts. Inspired by these findings, we propose a method for safety prompt optimization, namely DRO (Directed Representation Optimization). Treating a safety prompt as continuous, trainable embeddings, DRO learns to move the queries' representations along or opposite the refusal direction, depending on their harmfulness. Experiments with eight LLMs on out-of-domain and jailbreak benchmarks demonstrate that DRO remarkably improves the safeguarding performance of human-crafted safety prompts, without compromising the models' general performance.
Authors: David Rozado
Abstract: I report here a comprehensive analysis about the political preferences embedded in Large Language Models (LLMs). Namely, I administer 11 political orientation tests, designed to identify the political preferences of the test taker, to 24 state-of-the-art conversational LLMs, both closed and open source. When probed with questions/statements with political connotations, most conversational LLMs tend to generate responses that are diagnosed by most political test instruments as manifesting preferences for left-of-center viewpoints. This does not appear to be the case for five additional base (i.e. foundation) models upon which LLMs optimized for conversation with humans are built. However, the weak performance of the base models at coherently answering the tests' questions makes this subset of results inconclusive. Finally, I demonstrate that LLMs can be steered towards specific locations in the political spectrum through Supervised Fine-Tuning (SFT) with only modest amounts of politically aligned data, suggesting SFT's potential to embed political orientation in LLMs. With LLMs beginning to partially displace traditional information sources like search engines and Wikipedia, the societal implications of political biases embedded in LLMs are substantial.
Authors: Junhua Zeng, Chao Li, Zhun Sun, Qibin Zhao, Guoxu Zhou
Abstract: Tensor networks are efficient for extremely high-dimensional representation, but their model selection, known as tensor network structure search (TN-SS), is a challenging problem. Although several works have targeted TN-SS, most existing algorithms are manually crafted heuristics with poor performance, suffering from the curse of dimensionality and local convergence. In this work, we jump out of the box, studying how to harness large language models (LLMs) to automatically discover new TN-SS algorithms, replacing the involvement of human experts. By observing how human experts innovate in research, we model their common workflow and propose an automatic algorithm discovery framework called tnGPS. The proposed framework is an elaborate prompting pipeline that instruct LLMs to generate new TN-SS algorithms through iterative refinement and enhancement. The experimental results demonstrate that the algorithms discovered by tnGPS exhibit superior performance in benchmarks compared to the current state-of-the-art methods.
Authors: Philip Quirke, Clement Neo, Fazl Barez
Abstract: Language Models (LMs) are increasingly used for a wide range of prediction tasks, but their training can often neglect rare edge cases, reducing their reliability. Here, we define a stringent standard of trustworthiness whereby the task algorithm and circuit implementation must be verified, accounting for edge cases, with no known failure modes. We show that a transformer model can be trained to meet this standard if built using mathematically and logically specified frameworks. In this paper, we fully verify a model for n-digit integer addition. To exhibit the reusability of verified modules, we insert the trained integer addition model into an untrained model and train the combined model to perform both addition and subtraction. We find extensive reuse of the addition circuits for both tasks, easing verification of the more complex subtractor model. We discuss how inserting verified task modules into LMs can leverage model reuse to improve verifiability and trustworthiness of language models built using them. The reuse of verified circuits reduces the effort to verify more complex composite models which we believe to be a significant step towards safety of language models.
Authors: Fangru Lin, Emanuele La Malfa, Valentin Hofmann, Elle Michelle Yang, Anthony Cohn, Janet B. Pierrehumbert
Abstract: Planning is a fundamental property of human intelligence. Reasoning about asynchronous plans is challenging since it requires sequential and parallel planning to optimize time costs. Can large language models (LLMs) succeed at this task? Here, we present the first large-scale study investigating this question. We find that a representative set of closed and open-source LLMs, including GPT-4 and LLaMA-2, behave poorly when not supplied with illustrations about the task-solving process in our benchmark AsyncHow. We propose a novel technique called Plan Like a Graph (PLaG) that combines graphs with natural language prompts and achieves state-of-the-art results. We show that although PLaG can boost model performance, LLMs still suffer from drastic degradation when task complexity increases, highlighting the limits of utilizing LLMs for simulating digital devices. We see our study as an exciting step towards using LLMs as efficient autonomous agents. Our code and data are available at https://github.com/fangru-lin/graph-llm-asynchow-plan.
URLs: https://github.com/fangru-lin/graph-llm-asynchow-plan.
Authors: Yang Jin, Zhicheng Sun, Kun Xu, Kun Xu, Liwei Chen, Hao Jiang, Quzhe Huang, Chengru Song, Yuliang Liu, Di Zhang, Yang Song, Kun Gai, Yadong Mu
Abstract: In light of recent advances in multimodal Large Language Models (LLMs), there is increasing attention to scaling them from image-text data to more informative real-world videos. Compared to static images, video poses unique challenges for effective large-scale pre-training due to the modeling of its spatiotemporal dynamics. In this paper, we address such limitations in video-language pre-training with an efficient video decomposition that represents each video as keyframes and temporal motions. These are then adapted to an LLM using well-designed tokenizers that discretize visual and temporal information as a few tokens, thus enabling unified generative pre-training of videos, images, and text. At inference, the generated tokens from the LLM are carefully recovered to the original continuous pixel space to create various video content. Our proposed framework is both capable of comprehending and generating image and video content, as demonstrated by its competitive performance across 13 multimodal benchmarks in image and video understanding and generation. Our code and models are available at https://video-lavit.github.io.
Authors: Jinghong Li, Huy Phan, Wen Gu, Koichi Ota, Shinobu Hasegawa
Abstract: Research surveys have always posed a challenge for beginner researchers who lack of research training. These researchers struggle to understand the directions within their research topic, and the discovery of new research findings within a short time. One way to provide intuitive assistance to beginner researchers is by offering relevant knowledge graphs(KG) and recommending related academic papers. However, existing navigation knowledge graphs primarily rely on keywords in the research field and often fail to present the logical hierarchy among multiple related papers clearly. Moreover, most recommendation systems for academic papers simply rely on high text similarity, which can leave researchers confused as to why a particular article is being recommended. They may lack of grasp important information about the insight connection between "Issue resolved" and "Issue finding" that they hope to obtain. To address these issues, this study aims to support research insight surveys for beginner researchers by establishing a hierarchical tree-structured knowledge graph that reflects the inheritance insight of research topics and the relevance insight among the academic papers.
Authors: Hainiu Xu, Runcong Zhao, Lixing Zhu, Jinhua Du, Yulan He
Abstract: Neural Theory-of-Mind (N-ToM), machine's ability to understand and keep track of the mental states of others, is pivotal in developing socially intelligent agents. However, prevalent N-ToM benchmarks have several shortcomings, including the presence of ambiguous and artificial narratives, absence of personality traits and preferences, a lack of questions addressing characters' psychological mental states, and limited diversity in the questions posed. In response to these issues, we construct OpenToM, a new benchmark for assessing N-ToM with (1) longer and clearer narrative stories, (2) characters with explicit personality traits, (3) actions that are triggered by character intentions, and (4) questions designed to challenge LLMs' capabilities of modeling characters' mental states of both the physical and psychological world. Using OpenToM, we reveal that state-of-the-art LLMs thrive at modeling certain aspects of mental states in the physical world but fall short when tracking characters' mental states in the psychological world.
Authors: Ashish Shenoy, Yichao Lu, Srihari Jayakumar, Debojeet Chatterjee, Mohsen Moslehpour, Pierce Chuang, Abhay Harpale, Vikas Bhardwaj, Di Xu, Shicong Zhao, Longfang Zhao, Ankit Ramchandani, Xin Luna Dong, Anuj Kumar
Abstract: We introduce Lumos, the first end-to-end multimodal question-answering system with text understanding capabilities. At the core of Lumos is a Scene Text Recognition (STR) component that extracts text from first person point-of-view images, the output of which is used to augment input to a Multimodal Large Language Model (MM-LLM). While building Lumos, we encountered numerous challenges related to STR quality, overall latency, and model inference. In this paper, we delve into those challenges, and discuss the system architecture, design choices, and modeling techniques employed to overcome these obstacles. We also provide a comprehensive evaluation for each component, showcasing high quality and efficiency.
Authors: Shashwat Singh, Shauli Ravfogel, Jonathan Herzig, Roee Aharoni, Ryan Cotterell, Ponnurangam Kumaraguru
Abstract: Language models often exhibit undesirable behavior, e.g., generating toxic or gender-biased text. In the case of neural language models, an encoding of the undesirable behavior is often present in the model's representations. Thus, one natural (and common) approach to prevent the model from exhibiting undesirable behavior is to steer the model's representations in a manner that reduces the probability of it generating undesirable text. This paper investigates the formal and empirical properties of steering functions, i.e., transformation of the neural language model's representations that alter its behavior. First, we derive two optimal, in the least-squares sense, affine steering functions under different constraints. Our theory provides justification for existing approaches and offers a novel, improved steering approach. Second, we offer a series of experiments that demonstrate the empirical effectiveness of the methods in mitigating bias and reducing toxic generation.
Authors: Marios Papachristou, Yuan Yuan
Abstract: Social networks shape opinions, behaviors, and information dissemination in human societies. As large language models (LLMs) increasingly integrate into social and professional environments, understanding their behavior within the context of social interactions and networks becomes essential. Our study analyzes LLMs' network formation behavior to examine whether the dynamics of multiple LLMs are similar to or different from human social dynamics. We observe that LLMs exhibit key social network principles, including preferential attachment, triadic closure, homophily, community structure, and the small-world phenomenon, when asked about their preferences in network formation. We also investigate LLMs' decision-making based on real-world networks, revealing that triadic closure and homophily have a stronger influence than preferential attachment and that LLMs perform well in network formation predictions. Overall, our study opens up new possibilities for using LLMs in network science research and helps develop socially aware LLMs by shedding light on their social interaction behaviors and exploring their impacts on social dynamics.
Authors: Jihyung Kil, Farideh Tavazoee, Dongyeop Kang, Joo-Kyung Kim
Abstract: Visual Question Answering (VQA) often involves diverse reasoning scenarios across Vision and Language (V&L). Most prior VQA studies, however, have merely focused on assessing the model's overall accuracy without evaluating it on different reasoning cases. Furthermore, some recent works observe that conventional Chain-of-Thought (CoT) prompting fails to generate effective reasoning for VQA, especially for complex scenarios requiring multi-hop reasoning. In this paper, we propose II-MMR, a novel idea to identify and improve multi-modal multi-hop reasoning in VQA. In specific, II-MMR takes a VQA question with an image and finds a reasoning path to reach its answer using two novel language promptings: (i) answer prediction-guided CoT prompt, or (ii) knowledge triplet-guided prompt. II-MMR then analyzes this path to identify different reasoning cases in current VQA benchmarks by estimating how many hops and what types (i.e., visual or beyond-visual) of reasoning are required to answer the question. On popular benchmarks including GQA and A-OKVQA, II-MMR observes that most of their VQA questions are easy to answer, simply demanding "single-hop" reasoning, whereas only a few questions require "multi-hop" reasoning. Moreover, while the recent V&L model struggles with such complex multi-hop reasoning questions even using the traditional CoT method, II-MMR shows its effectiveness across all reasoning cases in both zero-shot and fine-tuning settings.
Authors: Govind Gangadhar, Karl Stratos
Abstract: Standard fine-tuning is considered not as effective as specialized methods for model editing due to its comparatively poor performance. However, it is simple, agnostic to the architectural details of the model being edited, and able to leverage advances in standard training techniques with no additional work (e.g., black-box PEFT for computational efficiency), making it an appealing choice for a model editor. In this work, we show that standard fine-tuning alone can yield competitive model editing performance with two minor modifications. First, we optimize the conditional likelihood rather than the full likelihood. Second, in addition to the typical practice of training on randomly paraphrased edit prompts to encourage generalization, we also train on random or similar unedited facts to encourage locality. Our experiments on the ZsRE and CounterFact datasets demonstrate that these simple modifications allow standard fine-tuning to match or outperform highly specialized editors in terms of edit score.
Authors: Muling Wu, Wenhao Liu, Xiaohua Wang, Tianlong Li, Changze Lv, Zixuan Ling, Jianhao Zhu, Cenyuan Zhang, Xiaoqing Zheng, Xuanjing Huang
Abstract: Parameter Efficient Fine-Tuning (PEFT) techniques have drawn significant attention due to their ability to yield competitive results while updating only a small portion of the adjustable parameters. However, existing PEFT methods pose challenges in hyperparameter selection, such as choosing the rank for LoRA or Adapter, or specifying the length of soft prompts. To address these challenges, we propose a novel fine-tuning approach for neural models, named Representation EDiting (RED), which modifies the representations generated at some layers through the application of scaling and biasing operations. While existing PEFT methods still demonstrate over-parameterization that could potentially undermine the generalization ability acquired from pre-training, RED can substantially reduce the number of trainable parameters by a factor of 25, 700 compared to full parameter fine-tuning and by a factor of 32 relative to LoRA. Remarkably, RED achieves results comparable or superior to both full parameter fine-tuning and other PEFT methods. Extensive experiments across various model architectures and scales, including RoBERTa, GPT-2, T5, and LLaMA-2, have demonstrated the effectiveness and efficiency of RED1, thereby positioning it as a promising PEFT strategy for large-scale neural models.
Authors: Lei Li, Yuqi Wang, Runxin Xu, Peiyi Wang, Xiachong Feng, Lingpeng Kong, Qi Liu
Abstract: Large vision-language models (LVLMs) excel across diverse tasks involving concrete images from natural scenes. However, their ability to interpret abstract figures, such as geometry shapes and scientific plots, remains limited due to a scarcity of training datasets in scientific domains. To fill this gap, we introduce Multimodal ArXiv, consisting of ArXivCap and ArXivQA, for enhancing LVLMs scientific comprehension. ArXivCap is a figure-caption dataset comprising 6.4M images and 3.9M captions, sourced from 572K ArXiv papers spanning various scientific domains. Drawing from ArXivCap, we introduce ArXivQA, a question-answering dataset generated by prompting GPT-4V based on scientific figures. ArXivQA greatly enhances open-sourced LVLMs' mathematical reasoning capabilities, achieving a 10.4\% absolute accuracy gain on a multimodal mathematical reasoning benchmark. Furthermore, employing ArXivCap, we devise four vision-to-text tasks for benchmarking LVLMs. Evaluation results with state-of-the-art LVLMs underscore their struggle with the nuanced semantics of academic figures, while domain-specific training yields substantial performance gains. Our error analysis uncovers misinterpretations of visual context, recognition errors, and the production of overly simplified captions by current LVLMs, shedding light on future improvements.
Authors: Max Lamparth, Anthony Corso, Jacob Ganz, Oriana Skylar Mastro, Jacquelyn Schneider, Harold Trinkunas
Abstract: To some, the advent of artificial intelligence (AI) promises better decision-making and increased military effectiveness while reducing the influence of human error and emotions. However, there is still debate about how AI systems, especially large language models (LLMs), behave compared to humans in high-stakes military decision-making scenarios with the potential for increased risks towards escalation and unnecessary conflicts. To test this potential and scrutinize the use of LLMs for such purposes, we use a new wargame experiment with 107 national security experts designed to look at crisis escalation in a fictional US-China scenario and compare human players to LLM-simulated responses in separate simulations. Wargames have a long history in the development of military strategy and the response of nations to threats or attacks. Here, we show a considerable high-level agreement in the LLM and human responses and significant quantitative and qualitative differences in individual actions and strategic tendencies. These differences depend on intrinsic biases in LLMs regarding the appropriate level of violence following strategic instructions, the choice of LLM, and whether the LLMs are tasked to decide for a team of players directly or first to simulate dialog between players. When simulating the dialog, the discussions lack quality and maintain a farcical harmony. The LLM simulations cannot account for human player characteristics, showing no significant difference even for extreme traits, such as "pacifist" or "aggressive sociopath". Our results motivate policymakers to be cautious before granting autonomy or following AI-based strategy recommendations.
Authors: Egor Zverev, Sahar Abdelnabi, Soroush Tabesh, Mario Fritz, Christoph H. Lampert
Abstract: Instruction-tuned Large Language Models (LLMs) show impressive results in numerous practical applications, but they lack essential safety features that are common in other areas of computer science, particularly an explicit separation of instructions and data. This makes them vulnerable to manipulations such as indirect prompt injections and generally unsuitable for safety-critical tasks. Surprisingly, there is currently no established definition or benchmark to quantify this phenomenon. In this work, we close this gap by introducing a formal measure for instruction-data separation and an empirical variant that is calculable from a model's outputs. We also present a new dataset, SEP, that allows estimating the measure for real-world models. Our results on various LLMs show that the problem of instruction-data separation is real: all models fail to achieve high separation, and canonical mitigation techniques, such as prompt engineering and fine-tuning, either fail to substantially improve separation or reduce model utility. The source code and SEP dataset are openly accessible at https://github.com/egozverev/Shold-It-Be-Executed-Or-Processed.
URLs: https://github.com/egozverev/Shold-It-Be-Executed-Or-Processed.
Authors: Abdelrhman Werby, Chenguang Huang, Martin B\"uchner, Abhinav Valada, Wolfram Burgard
Abstract: Recent open-vocabulary robot mapping methods enrich dense geometric maps with pre-trained visual-language features. While these maps allow for the prediction of point-wise saliency maps when queried for a certain language concept, large-scale environments and abstract queries beyond the object level still pose a considerable hurdle, ultimately limiting language-grounded robotic navigation. In this work, we present HOV-SG, a hierarchical open-vocabulary 3D scene graph mapping approach for language-grounded robot navigation. Leveraging open-vocabulary vision foundation models, we first obtain state-of-the-art open-vocabulary segment-level maps in 3D and subsequently construct a 3D scene graph hierarchy consisting of floor, room, and object concepts, each enriched with open-vocabulary features. Our approach is able to represent multi-story buildings and allows robotic traversal of those using a cross-floor Voronoi graph. HOV-SG is evaluated on three distinct datasets and surpasses previous baselines in open-vocabulary semantic accuracy on the object, room, and floor level while producing a 75% reduction in representation size compared to dense open-vocabulary maps. In order to prove the efficacy and generalization capabilities of HOV-SG, we showcase successful long-horizon language-conditioned robot navigation within real-world multi-storage environments. We provide code and trial video data at http://hovsg.github.io/.
URLs: http://hovsg.github.io/.
Authors: Ziyao Zeng, Daniel Wang, Fengyu Yang, Hyoungseob Park, Yangchao Wu, Stefano Soatto, Byung-Woo Hong, Dong Lao, Alex Wong
Abstract: Three-dimensional (3D) reconstruction from a single image is an ill-posed problem with inherent ambiguities, i.e. scale. Predicting a 3D scene from text description(s) is similarly ill-posed, i.e. spatial arrangements of objects described. We investigate the question of whether two inherently ambiguous modalities can be used in conjunction to produce metric-scaled reconstructions. To test this, we focus on monocular depth estimation, the problem of predicting a dense depth map from a single image, but with an additional text caption describing the scene. To this end, we begin by encoding the text caption as a mean and standard deviation; using a variational framework, we learn the distribution of the plausible metric reconstructions of 3D scenes corresponding to the text captions as a prior. To "select" a specific reconstruction or depth map, we encode the given image through a conditional sampler that samples from the latent space of the variational text encoder, which is then decoded to the output depth map. Our approach is trained alternatingly between the text and image branches: in one optimization step, we predict the mean and standard deviation from the text description and sample from a standard Gaussian, and in the other, we sample using a (image) conditional sampler. Once trained, we directly predict depth from the encoded text using the conditional sampler. We demonstrate our approach on indoor (NYUv2) and outdoor (KITTI) scenarios, where we show that language can consistently improve performance in both.
Authors: Dongzhi Jiang, Guanglu Song, Xiaoshi Wu, Renrui Zhang, Dazhong Shen, Zhuofan Zong, Yu Liu, Hongsheng Li
Abstract: Diffusion models have demonstrated great success in the field of text-to-image generation. However, alleviating the misalignment between the text prompts and images is still challenging. The root reason behind the misalignment has not been extensively investigated. We observe that the misalignment is caused by inadequate token attention activation. We further attribute this phenomenon to the diffusion model's insufficient condition utilization, which is caused by its training paradigm. To address the issue, we propose CoMat, an end-to-end diffusion model fine-tuning strategy with an image-to-text concept matching mechanism. We leverage an image captioning model to measure image-to-text alignment and guide the diffusion model to revisit ignored tokens. A novel attribute concentration module is also proposed to address the attribute binding problem. Without any image or human preference data, we use only 20K text prompts to fine-tune SDXL to obtain CoMat-SDXL. Extensive experiments show that CoMat-SDXL significantly outperforms the baseline model SDXL in two text-to-image alignment benchmarks and achieves start-of-the-art performance.
Authors: Cai Chen, Runzhong Zhang, Jianjun Gao, Kejun Wu, Kim-Hui Yap, Yi Wang
Abstract: Temporal sentence grounding involves the retrieval of a video moment with a natural language query. Many existing works directly incorporate the given video and temporally localized query for temporal grounding, overlooking the inherent domain gap between different modalities. In this paper, we utilize pseudo-query features containing extensive temporally global textual knowledge sourced from the same video-query pair, to enhance the bridging of domain gaps and attain a heightened level of similarity between multi-modal features. Specifically, we propose a Pseudo-query Intermediary Network (PIN) to achieve an improved alignment of visual and comprehensive pseudo-query features within the feature space through contrastive learning. Subsequently, we utilize learnable prompts to encapsulate the knowledge of pseudo-queries, propagating them into the textual encoder and multi-modal fusion module, further enhancing the feature alignment between visual and language for better temporal grounding. Extensive experiments conducted on the Charades-STA and ActivityNet-Captions datasets demonstrate the effectiveness of our method.
Authors: Jinghan Jia, Yihua Zhang, Yimeng Zhang, Jiancheng Liu, Bharat Runwal, James Diffenderfer, Bhavya Kailkhura, Sijia Liu
Abstract: Large Language Models (LLMs) have highlighted the necessity of effective unlearning mechanisms to comply with data regulations and ethical AI practices. LLM unlearning aims at removing undesired data influences and associated model capabilities without compromising utility out of the scope of unlearning. While interest in studying LLM unlearning is growing,the impact of the optimizer choice for LLM unlearning remains under-explored. In this work, we shed light on the significance of optimizer selection in LLM unlearning for the first time, establishing a clear connection between {second-order optimization} and influence unlearning (a classical approach using influence functions to update the model for data influence removal). This insight propels us to develop a second-order unlearning framework, termed SOUL, built upon the second-order clipped stochastic optimization (Sophia)-based LLM training method. SOUL extends the static, one-shot model update using influence unlearning to a dynamic, iterative unlearning process. Our extensive experiments show that SOUL consistently outperforms conventional first-order methods across various unlearning tasks, models, and metrics, suggesting the promise of second-order optimization in providing a scalable and easily implementable solution for LLM unlearning.
Authors: Parshin Shojaee, Kazem Meidani, Shashank Gupta, Amir Barati Farimani, Chandan K Reddy
Abstract: Mathematical equations have been unreasonably effective in describing complex natural phenomena across various scientific disciplines. However, discovering such insightful equations from data presents significant challenges due to the necessity of navigating extremely high-dimensional combinatorial and nonlinear hypothesis spaces. Traditional methods of equation discovery, commonly known as symbolic regression, largely focus on extracting equations from data alone, often neglecting the rich domain-specific prior knowledge that scientists typically depend on. To bridge this gap, we introduce LLM-SR, a novel approach that leverages the extensive scientific knowledge and robust code generation capabilities of Large Language Models (LLMs) to discover scientific equations from data in an efficient manner. Specifically, LLM-SR treats equations as programs with mathematical operators and combines LLMs' scientific priors with evolutionary search over equation programs. The LLM iteratively proposes new equation skeleton hypotheses, drawing from its physical understanding, which are then optimized against data to estimate skeleton parameters. We demonstrate LLM-SR's effectiveness across three diverse scientific domains, where it discovers physically accurate equations that provide significantly better fits to in-domain and out-of-domain data compared to the well-established symbolic regression baselines. Incorporating scientific prior knowledge also enables LLM-SR to search the equation space more efficiently than baselines. Code is available at: https://github.com/deep-symbolic-mathematics/LLM-SR
Authors: Fatemeh Nazary, Yashar Deldjoo, Tommaso Di Noia, Eugenio di Sciascio
Abstract: The integration of Large Language Models (LLMs) into healthcare diagnostics offers a promising avenue for clinical decision-making. This study outlines the development of a novel method for zero-shot/few-shot in-context learning (ICL) by integrating medical domain knowledge using a multi-layered structured prompt. We also explore the efficacy of two communication styles between the user and LLMs: the Numerical Conversational (NC) style, which processes data incrementally, and the Natural Language Single-Turn (NL-ST) style, which employs long narrative prompts. Our study systematically evaluates the diagnostic accuracy and risk factors, including gender bias and false negative rates, using a dataset of 920 patient records in various few-shot scenarios. Results indicate that traditional clinical machine learning (ML) models generally outperform LLMs in zero-shot and few-shot settings. However, the performance gap narrows significantly when employing few-shot examples alongside effective explainable AI (XAI) methods as sources of domain knowledge. Moreover, with sufficient time and an increased number of examples, the conversational style (NC) nearly matches the performance of ML models. Most notably, LLMs demonstrate comparable or superior cost-sensitive accuracy relative to ML models. This research confirms that, with appropriate domain knowledge and tailored communication strategies, LLMs can significantly enhance diagnostic processes. The findings highlight the importance of optimizing the number of training examples and communication styles to improve accuracy and reduce biases in LLM applications.
Authors: Jian Hu, Xibin Wu, Weixun Wang, Xianyu, Dehao Zhang, Yu Cao
Abstract: As large language models (LLMs) continue to grow by scaling laws, reinforcement learning from human feedback (RLHF) has gained significant attention due to its outstanding performance. However, unlike pretraining or fine-tuning a single model, scaling reinforcement learning from human feedback (RLHF) for training large language models poses coordination challenges across four models. We present OpenRLHF, an open-source framework enabling efficient RLHF scaling. Unlike existing RLHF frameworks that co-locate four models on the same GPUs, OpenRLHF re-designs scheduling for the models beyond 70B parameters using Ray, vLLM, and DeepSpeed, leveraging improved resource utilization and diverse training approaches. Integrating seamlessly with Hugging Face, OpenRLHF provides an out-of-the-box solution with optimized algorithms and launch scripts, which ensures user-friendliness. OpenRLHF implements RLHF, DPO, rejection sampling, and other alignment techniques. Empowering state-of-the-art LLM development, OpenRLHF's code is available at https://github.com/OpenLLMAI/OpenRLHF.
Authors: Xinting Huang, Madhur Panwar, Navin Goyal, Michael Hahn
Abstract: The inner workings of neural networks can be better understood if we can fully decipher the information encoded in neural activations. In this paper, we argue that this information is embodied by the subset of inputs that give rise to similar activations. Computing such subsets is nontrivial as the input space is exponentially large. We propose InversionView, which allows us to practically inspect this subset by sampling from a trained decoder model conditioned on activations. This helps uncover the information content of activation vectors, and facilitates understanding of the algorithms implemented by transformer models. We present three case studies where we investigate models ranging from small transformers to GPT-2. In these studies, we demonstrate the characteristics of our method, show the distinctive advantages it offers, and provide causally verified circuits.
Authors: Hao Mark Chen, Wayne Luk, Ka Fai Cedric Yiu, Rui Li, Konstantin Mishchenko, Stylianos I. Venieris, Hongxiang Fan
Abstract: The auto-regressive decoding of Large Language Models (LLMs) results in significant overheads in their hardware performance. While recent research has investigated various speculative decoding techniques for multi-token generation, these efforts have primarily focused on improving processing speed such as throughput. Crucially, they often neglect other metrics essential for real-life deployments, such as memory consumption and training cost. To overcome these limitations, we propose a novel parallel prompt decoding that requires only $0.0002$% trainable parameters, enabling efficient training on a single A100-40GB GPU in just 16 hours. Inspired by the human natural language generation process, $PPD$ approximates outputs generated at future timesteps in parallel by using multiple prompt tokens. This approach partially recovers the missing conditional dependency information necessary for multi-token generation, resulting in up to a 28% higher acceptance rate for long-range predictions. Furthermore, we present a hardware-aware dynamic sparse tree technique that adaptively optimizes this decoding scheme to fully leverage the computational capacities on different GPUs. Through extensive experiments across LLMs ranging from MobileLlama to Vicuna-13B on a wide range of benchmarks, our approach demonstrates up to 2.49$\times$ speedup and maintains a minimal runtime memory overhead of just $0.0004$%. More importantly, our parallel prompt decoding can serve as an orthogonal optimization for synergistic integration with existing speculative decoding, showing up to $1.22\times$ further speed improvement. Our code is available at https://github.com/hmarkc/parallel-prompt-decoding.
Authors: Renyi Qu, Mark Yatskar
Abstract: (Renyi Qu's Master's Thesis) Recent advancements in interpretable models for vision-language tasks have achieved competitive performance; however, their interpretability often suffers due to the reliance on unstructured text outputs from large language models (LLMs). This introduces randomness and compromises both transparency and reliability, which are essential for addressing safety issues in AI systems. We introduce \texttt{Hi-CoDe} (Hierarchical Concept Decomposition), a novel framework designed to enhance model interpretability through structured concept analysis. Our approach consists of two main components: (1) We use GPT-4 to decompose an input image into a structured hierarchy of visual concepts, thereby forming a visual concept tree. (2) We then employ an ensemble of simple linear classifiers that operate on concept-specific features derived from CLIP to perform classification. Our approach not only aligns with the performance of state-of-the-art models but also advances transparency by providing clear insights into the decision-making process and highlighting the importance of various concepts. This allows for a detailed analysis of potential failure modes and improves model compactness, therefore setting a new benchmark in interpretability without compromising the accuracy.
Authors: Markus J. Buehler
Abstract: We present Cephalo, a series of multimodal vision large language models (V-LLMs) designed for materials science applications, integrating visual and linguistic data for enhanced understanding and interaction within human-AI and multi-agent AI frameworks. A key innovation of Cephalo is its advanced dataset generation method, which employs a sophisticated algorithm to accurately detect and separate images and their corresponding textual descriptions from PDF documents, such as scientific papers. The method includes a careful refinement of image-text pairs through integrated vision and language processing, ensuring high-quality, contextually relevant, and well reasoned training data. Cephalo is trained on integrated image and text data extracted from thousands of scientific papers and science-focused Wikipedia pages demonstrates can interpret complex visual scenes, generate precise language descriptions, and answer queries about images effectively. The combination of a vision encoder with an autoregressive transformer supports complex natural language understanding in an integrated model, which can be coupled with other generative methods to create an image-to-text-to-image or image-to-text-to-3D pipeline. To explore the development of larger models from smaller ones, we report both mixture-of-expert methods and model merging. These hybrid approaches allow us to leverage the domain-specific expertise and general conversational capabilities to harness the strengths of multiple models. We examine the models in diverse use cases that incorporate biological materials, fracture and engineering analysis, protein biophysics, and bio-inspired design based on insect behavior. Generative applications include bio-inspired designs, including pollen-inspired architected materials, as well as the synthesis of bio-inspired material microstructures from a photograph of a solar eclipse.
Authors: Sean Williams, James Huckle
Abstract: We introduce a comprehensive Linguistic Benchmark designed to evaluate the limitations of Large Language Models (LLMs) in domains such as logical reasoning, spatial intelligence, and linguistic understanding, among others. Through a series of straightforward questions, it uncovers the significant limitations of well-regarded models to perform tasks that humans manage with ease. It also highlights the potential of prompt engineering to mitigate some errors and underscores the necessity for better training methodologies. Our findings stress the importance of grounding LLMs with human reasoning and common sense, emphasising the need for human-in-the-loop for enterprise applications. We hope this work paves the way for future research to enhance the usefulness and reliability of new models.
Authors: Mingda Li, Xinyu Li, Yifan Chen, Wenfeng Xuan, Weinan Zhang
Abstract: Although Retrieval-Augmented Large Language Models (RALMs) demonstrate their superiority in terms of factuality, they do not consistently outperform the original retrieval-free Language Models (LMs). Our experiments reveal that this example-level performance inconsistency exists not only between retrieval-augmented and retrieval-free LM but also among different retrievers. To understand this phenomenon, we investigate the degeneration behavior of RALMs and theoretically decompose it into four categories. Further analysis based on our decomposition reveals that the innate difference in knowledge sources and the unpredictable degeneration of the reader model contribute most to the inconsistency. Drawing from our analysis, we introduce Ensemble of Retrievers (EoR), a trainable framework that can adaptively retrieve from different knowledge sources and effectively decrease unpredictable reader errors. Our experiments on Open Domain Question Answering show that EoR substantially improves performance over the RALM with a single retriever by considerably reducing inconsistent behaviors.
Authors: Ciro Beneduce, Bruno Lepri, Massimiliano Luca
Abstract: Predicting the locations an individual will visit in the future is crucial for solving many societal issues like disease diffusion and reduction of pollution among many others. The models designed to tackle next-location prediction, however, require a significant amount of individual-level information to be trained effectively. Such data may be scarce or even unavailable in some geographic regions or peculiar scenarios (e.g., cold-start in recommendation systems). Moreover, the design of a next-location predictor able to generalize or geographically transfer knowledge is still an open research challenge. Recent advances in natural language processing have led to a rapid diffusion of Large Language Models (LLMs) which have shown good generalization and reasoning capabilities. These insights, coupled with the recent findings that LLMs are rich in geographical knowledge, allowed us to believe that these models can act as zero-shot next-location predictors. This paper evaluates the capabilities of many popular LLMs in this role, specifically Llama, GPT-3.5 and Mistral 7B. After designing a proper prompt, we tested the models on three real-world mobility datasets. The results show that LLMs can obtain accuracies up to 32.4%, a significant relative improvement of over 600% when compared to sophisticated DL models specifically designed for human mobility. Moreover, we show that other LLMs are unable to perform the task properly. To prevent positively biased results, we also propose a framework inspired by other studies to test data contamination. Finally, we explored the possibility of using LLMs as text-based explainers for next-location prediction showing that can effectively provide an explanation for their decision. Notably, 7B models provide more generic, but still reliable, explanations compared to larger counterparts. Code: github.com/ssai-trento/LLM-zero-shot-NL