Authors: Kim Trong Vu, Michael Krumdick, Varshini Reddy, Franck Dernoncourt, Viet Dac Lai
Abstract: FActScore has gained popularity as a metric to estimate the factuality of long-form texts generated by Large Language Models (LLMs) in English. However, there has not been any work in studying the behavior of FActScore in other languages. This paper studies the limitations of each component in the four-component pipeline of FActScore in the multilingual setting. We introduce a new dataset for FActScore on texts generated by strong multilingual LLMs. Our evaluation shows that LLMs exhibit distinct behaviors in both fact extraction and fact scoring tasks. No LLM produces consistent and reliable FActScore across languages with varying levels of resources. We also find that the knowledge source plays an important role in the quality of the estimated FActScore. Using Wikipedia as the knowledge source may hinder the true FActScore of long-form text due to its limited coverage in medium- and low-resource languages. We also incorporate three mitigations to our knowledge source that ultimately improve FActScore estimation across all languages.
Authors: Lekang Jiang, Caiqi Zhang, Pascal A Scherz, Stephan Goetz
Abstract: Large language models (LLMs) have shown exceptional performance across various text generation tasks but remain under-explored in the patent domain, which offers highly structured and precise language. This paper constructs a dataset to investigate the performance of current LLMs in patent claim generation. Our results demonstrate that generating claims based on patent descriptions outperforms previous research relying on abstracts. Interestingly, current patent-specific LLMs perform much worse than state-of-the-art general LLMs, highlighting the necessity for future research on in-domain LLMs. We also find that LLMs can produce high-quality first independent claims, but their performances markedly decrease for subsequent dependent claims. Moreover, fine-tuning can enhance the completeness of inventions' features, conceptual clarity, and feature linkage. Among the tested LLMs, GPT-4 demonstrates the best performance in comprehensive human evaluations by patent experts, with better feature coverage, conceptual clarity, and technical coherence. Despite these capabilities, comprehensive revision and modification are still necessary to pass rigorous patent scrutiny and ensure legal robustness.
Authors: Vipul Gupta, David Pantoja, Candace Ross, Adina Williams, Megan Ung
Abstract: As large language models (LLMs) have grown in prevalence, particular benchmarks have become essential for the evaluation of these models and for understanding model capabilities. Most commonly, we use test accuracy averaged across multiple subtasks in order to rank models on leaderboards, to determine which model is best for our purposes. In this paper, we investigate the robustness of the accuracy measurement on a widely used multiple choice question answering dataset, MMLU. When shuffling the answer label contents, we find that all explored models decrease in accuracy on MMLU, but not every model is equally sensitive. These findings suggest a possible adjustment to the standard practice of leaderboard testing, where we additionally consider the percentage of examples each model answers correctly by random chance.
Authors: Ergun Bi\c{c}ici
Abstract: We use transductive regression techniques to learn mappings between source and target features of given parallel corpora and use these mappings to generate machine translation outputs. We show the effectiveness of $L_1$ regularized regression (\textit{lasso}) to learn the mappings between sparsely observed feature sets versus $L_2$ regularized regression. Proper selection of training instances plays an important role to learn correct feature mappings within limited computational resources and at expected accuracy levels. We introduce \textit{dice} instance selection method for proper selection of training instances, which plays an important role to learn correct feature mappings for improving the source and target coverage of the training set. We show that $L_1$ regularized regression performs better than $L_2$ regularized regression both in regression measurements and in the translation experiments using graph decoding. We present encouraging results when translating from German to English and Spanish to English. We also demonstrate results when the phrase table of a phrase-based decoder is replaced with the mappings we find with the regression model.
Authors: Marcos Treviso, Nuno M. Guerreiro, Sweta Agrawal, Ricardo Rei, Jos\'e Pombal, Tania Vaz, Helena Wu, Beatriz Silva, Daan van Stigt, Andr\'e F. T. Martins
Abstract: While machine translation (MT) systems are achieving increasingly strong performance on benchmarks, they often produce translations with errors and anomalies. Understanding these errors can potentially help improve the translation quality and user experience. This paper introduces xTower, an open large language model (LLM) built on top of TowerBase designed to provide free-text explanations for translation errors in order to guide the generation of a corrected translation. The quality of the generated explanations by xTower are assessed via both intrinsic and extrinsic evaluation. We ask expert translators to evaluate the quality of the explanations across two dimensions: relatedness towards the error span being explained and helpfulness in error understanding and improving translation quality. Extrinsically, we test xTower across various experimental setups in generating translation corrections, demonstrating significant improvements in translation quality. Our findings highlight xTower's potential towards not only producing plausible and helpful explanations of automatic translations, but also leveraging them to suggest corrected translations.
Authors: Shouchang Guo, Sonam Damani, Keng-hao Chang
Abstract: In prompt tuning, a prefix or suffix text is added to the prompt, and the embeddings (soft prompts) or token indices (hard prompts) of the prefix/suffix are optimized to gain more control over language models for specific tasks. This approach eliminates the need for hand-crafted prompt engineering or explicit model fine-tuning. Prompt tuning is significantly more parameter-efficient than model fine-tuning, as it involves optimizing partial inputs of language models to produce desired outputs. In this work, we aim to further reduce the amount of trainable parameters required for a language model to perform well on specific tasks. We propose Low-rank Prompt Tuning (LoPT), a low-rank model for prompts that achieves efficient prompt optimization. The proposed method demonstrates similar outcomes to full parameter prompt tuning while reducing the number of trainable parameters by a factor of 5. It also provides promising results compared to the state-of-the-art methods that would require 10 to 20 times more parameters.
Authors: Anubhab Majumder, Kausik Bhattacharya, Amaresh Chakrabarti
Abstract: Representing systems using the SAPPhIRE causality model is found useful in supporting design-by-analogy. However, creating a SAPPhIRE model of artificial or biological systems is an effort-intensive process that requires human experts to source technical knowledge from multiple technical documents regarding how the system works. This research investigates how to leverage Large Language Models (LLMs) in creating structured descriptions of systems using the SAPPhIRE model of causality. This paper, the second part of the two-part research, presents a new Retrieval-Augmented Generation (RAG) tool for generating information related to SAPPhIRE constructs of artificial systems and reports the results from a preliminary evaluation of the tool's success - focusing on the factual accuracy and reliability of outcomes.
Authors: Naseela Pervez, Alexander J. Titus
Abstract: Large language models (LLMs) are increasingly utilized to assist in scientific and academic writing, helping authors enhance the coherence of their articles. Previous studies have highlighted stereotypes and biases present in LLM outputs, emphasizing the need to evaluate these models for their alignment with human narrative styles and potential gender biases. In this study, we assess the alignment of three prominent LLMs - Claude 3 Opus, Mistral AI Large, and Gemini 1.5 Flash - by analyzing their performance on benchmark text-generation tasks for scientific abstracts. We employ the Linguistic Inquiry and Word Count (LIWC) framework to extract lexical, psychological, and social features from the generated texts. Our findings indicate that, while these models generally produce text closely resembling human authored content, variations in stylistic features suggest significant gender biases. This research highlights the importance of developing LLMs that maintain a diversity of writing styles to promote inclusivity in academic discourse.
Authors: Jiahai Feng, Stuart Russell, Jacob Steinhardt
Abstract: Language models are susceptible to bias, sycophancy, backdoors, and other tendencies that lead to unfaithful responses to the input context. Interpreting internal states of language models could help monitor and correct unfaithful behavior. We hypothesize that language models represent their input contexts in a latent world model, and seek to extract this latent world state from the activations. We do so with 'propositional probes', which compositionally probe tokens for lexical information and bind them into logical propositions representing the world state. For example, given the input context ''Greg is a nurse. Laura is a physicist.'', we decode the propositions ''WorksAs(Greg, nurse)'' and ''WorksAs(Laura, physicist)'' from the model's activations. Key to this is identifying a 'binding subspace' in which bound tokens have high similarity (''Greg'' and ''nurse'') but unbound ones do not (''Greg'' and ''physicist''). We validate propositional probes in a closed-world setting with finitely many predicates and properties. Despite being trained on simple templated contexts, propositional probes generalize to contexts rewritten as short stories and translated to Spanish. Moreover, we find that in three settings where language models respond unfaithfully to the input context -- prompt injections, backdoor attacks, and gender bias -- the decoded propositions remain faithful. This suggests that language models often encode a faithful world model but decode it unfaithfully, which motivates the search for better interpretability tools for monitoring LMs.
Authors: Miyoung Ko, Sue Hyun Park, Joonsuk Park, Minjoon Seo
Abstract: Despite significant advancements, there is a limited understanding of how large language models (LLMs) utilize knowledge for reasoning. To address this, we propose a method that deconstructs complex real-world questions into a graph, representing each question as a node with parent nodes of background knowledge needed to solve the question. We develop the DepthQA dataset, deconstructing questions into three depths: (i) recalling conceptual knowledge, (ii) applying procedural knowledge, and (iii) analyzing strategic knowledge. Based on a hierarchical graph, we quantify forward discrepancy, discrepancies in LLMs' performance on simpler sub-problems versus complex questions. We also measure backward discrepancy, where LLMs answer complex questions but struggle with simpler ones. Our analysis shows that smaller models have more discrepancies than larger models. Additionally, guiding models from simpler to complex questions through multi-turn interactions improves performance across model sizes, highlighting the importance of structured intermediate steps in knowledge reasoning. This work enhances our understanding of LLM reasoning and suggests ways to improve their problem-solving abilities.
Authors: Ibrahim Said Ahmad, Shiran Dudy, Resmi Ramachandranpillai, Kenneth Church
Abstract: Large Language Models (LLMs), such as ChatGPT, are widely used to generate content for various purposes and audiences. However, these models may not reflect the cultural and emotional diversity of their users, especially for low-resource languages. In this paper, we investigate how ChatGPT represents Hausa's culture and emotions. We compare responses generated by ChatGPT with those provided by native Hausa speakers on 37 culturally relevant questions. We conducted experiments using emotion analysis and applied two similarity metrics to measure the alignment between human and ChatGPT responses. We also collected human participants ratings and feedback on ChatGPT responses. Our results show that ChatGPT has some level of similarity to human responses, but also exhibits some gaps and biases in its knowledge and awareness of the Hausa culture and emotions. We discuss the implications and limitations of our methodology and analysis and suggest ways to improve the performance and evaluation of LLMs for low-resource languages.
Authors: Giuseppe Carenini, Jordon Johnson, Ali Salamatian
Abstract: Automatically captioning visualizations is not new, but recent advances in large language models(LLMs) open exciting new possibilities. In this tutorial, after providing a brief review of Information Visualization (InfoVis) principles and past work in captioning, we introduce neural models and the transformer architecture used in generic LLMs. We then discuss their recent applications in InfoVis, with a focus on captioning. Additionally, we explore promising future directions in this field.
Authors: Majd Saleh, Sarra Baghdadi, St\'ephane Paquelet
Abstract: Text segmentation holds paramount importance in the field of Natural Language Processing (NLP). It plays an important role in several NLP downstream tasks like information retrieval and document summarization. In this work, we propose a new solution, namely TocBERT, for segmenting texts using bidirectional transformers. TocBERT represents a supervised solution trained on the detection of titles and sub-titles from their semantic representations. This task was formulated as a named entity recognition (NER) problem. The solution has been applied on a medical text segmentation use-case where the Bio-ClinicalBERT model is fine-tuned to segment discharge summaries of the MIMIC-III dataset. The performance of TocBERT has been evaluated on a human-labeled ground truth corpus of 250 notes. It achieved an F1-score of 84.6% when evaluated on a linear text segmentation problem and 72.8% on a hierarchical text segmentation problem. It outperformed a carefully designed rule-based solution, particularly in distinguishing titles from subtitles.
Authors: Andrea Bacciu, Marco Damonte, Marco Basaldella, Emilio Monti
Abstract: The majority of Neural Semantic Parsing (NSP) models are developed with the assumption that there are no concepts outside the ones such models can represent with their target symbols (closed-world assumption). This assumption leads to generate hallucinated outputs rather than admitting their lack of knowledge. Hallucinations can lead to wrong or potentially offensive responses to users. Hence, a mechanism to prevent this behavior is crucial to build trusted NSP-based Question Answering agents. To that end, we propose the Hallucination Simulation Framework (HSF), a general setting for stimulating and analyzing NSP model hallucinations. The framework can be applied to any NSP task with a closed-ontology. Using the proposed framework and KQA Pro as the benchmark dataset, we assess state-of-the-art techniques for hallucination detection. We then present a novel hallucination detection strategy that exploits the computational graph of the NSP model to detect the NSP hallucinations in the presence of ontology gaps, out-of-domain utterances, and to recognize NSP errors, improving the F1-Score respectively by ~21, ~24% and ~1%. This is the first work in closed-ontology NSP that addresses the problem of recognizing ontology gaps. We release our code and checkpoints at https://github.com/amazon-science/handling-ontology-gaps-in-semantic-parsing.
URLs: https://github.com/amazon-science/handling-ontology-gaps-in-semantic-parsing.
Authors: Dan Schumacher, Fatemeh Haji, Tara Grey, Niharika Bandlamudi, Nupoor Karnik, Gagana Uday Kumar, Jason Cho-Yu Chiang, Paul Rad, Nishant Vishwamitra, Anthony Rios
Abstract: Large language models (LLMs) often struggle with temporal reasoning, crucial for tasks like historical event analysis and time-sensitive information retrieval. Despite advancements, state-of-the-art models falter in handling temporal information, especially when faced with irrelevant or noisy contexts. This paper addresses this gap by empirically examining the robustness of temporal question-answering (TQA) systems trained on various context types, including relevant, irrelevant, slightly altered, and no context. Our findings indicate that training with a mix of these contexts enhances model robustness and accuracy. Additionally, we show that the position of context relative to the question significantly impacts performance, with question-first positioning yielding better results. We introduce two new context-rich TQA datasets, ContextAQA and ContextTQE, and provide comprehensive evaluations and guidelines for training robust TQA models. Our work lays the foundation for developing reliable and context-aware temporal QA systems, with broader implications for enhancing LLM robustness against diverse and potentially adversarial information.
Authors: Seid Muhie Yimam, Daryna Dementieva, Tim Fischer, Daniil Moskovskiy, Naquee Rizwan, Punyajoy Saha, Sarthak Roy, Martin Semmann, Alexander Panchenko, Chris Biemann, Animesh Mukherjee
Abstract: Despite regulations imposed by nations and social media platforms, such as recent EU regulations targeting digital violence, abusive content persists as a significant challenge. Existing approaches primarily rely on binary solutions, such as outright blocking or banning, yet fail to address the complex nature of abusive speech. In this work, we propose a more comprehensive approach called Demarcation scoring abusive speech based on four aspect -- (i) severity scale; (ii) presence of a target; (iii) context scale; (iv) legal scale -- and suggesting more options of actions like detoxification, counter speech generation, blocking, or, as a final measure, human intervention. Through a thorough analysis of abusive speech regulations across diverse jurisdictions, platforms, and research papers we highlight the gap in preventing measures and advocate for tailored proactive steps to combat its multifaceted manifestations. Our work aims to inform future strategies for effectively addressing abusive speech online.
Authors: Ritam Dutt, Zhen Wu, Kelly Shi, Divyanshu Sheth, Prakhar Gupta, Carolyn Penstein Rose
Abstract: We present a generalizable classification approach that leverages Large Language Models (LLMs) to facilitate the detection of implicitly encoded social meaning in conversations. We design a multi-faceted prompt to extract a textual explanation of the reasoning that connects visible cues to underlying social meanings. These extracted explanations or rationales serve as augmentations to the conversational text to facilitate dialogue understanding and transfer. Our empirical results over 2,340 experimental settings demonstrate the significant positive impact of adding these rationales. Our findings hold true for in-domain classification, zero-shot, and few-shot domain transfer for two different social meaning detection tasks, each spanning two different corpora.
Authors: Florin Pop, Judd Rosenblatt, Diogo Schwerz de Lucena, Michael Vaiana
Abstract: In this paper, we investigate the degree to which fine-tuning in Large Language Models (LLMs) effectively mitigates versus merely conceals undesirable behavior. Through the lens of semi-realistic role-playing exercises designed to elicit such behaviors, we explore the response dynamics of LLMs post fine-tuning interventions. Our methodology involves prompting models for Chain-of-Thought (CoT) reasoning and analyzing the coherence between the reasoning traces and the resultant outputs. Notably, we identify a pervasive phenomenon we term \emph{reason-based deception}, where models either stop producing reasoning traces or produce seemingly ethical reasoning traces that belie the unethical nature of their final outputs. We further examine the efficacy of response strategies (polite refusal versus explicit rebuttal) in curbing the occurrence of undesired behavior in subsequent outputs of multi-turn interactions. Our findings reveal that explicit rebuttals significantly outperform polite refusals in preventing the continuation of undesired outputs and nearly eliminate reason-based deception, challenging current practices in model fine-tuning. Accordingly, the two key contributions of this paper are (1) defining and studying reason-based deception, a new type of hidden behavior, and (2) demonstrating that rebuttals provide a more robust response model to harmful requests than refusals, thereby highlighting the need to reconsider the response strategies in fine-tuning approaches.
Authors: Orevaoghene Ahia, Anuoluwapo Aremu, Diana Abagyan, Hila Gonen, David Ifeoluwa Adelani, Daud Abolade, Noah A. Smith, Yulia Tsvetkov
Abstract: Yor\`ub\'a an African language with roughly 47 million speakers encompasses a continuum with several dialects. Recent efforts to develop NLP technologies for African languages have focused on their standard dialects, resulting in disparities for dialects and varieties for which there are little to no resources or tools. We take steps towards bridging this gap by introducing a new high-quality parallel text and speech corpus YOR\`ULECT across three domains and four regional Yor\`ub\'a dialects. To develop this corpus, we engaged native speakers, travelling to communities where these dialects are spoken, to collect text and speech data. Using our newly created corpus, we conducted extensive experiments on (text) machine translation, automatic speech recognition, and speech-to-text translation. Our results reveal substantial performance disparities between standard Yor\`ub\'a and the other dialects across all tasks. However, we also show that with dialect-adaptive finetuning, we are able to narrow this gap. We believe our dataset and experimental analysis will contribute greatly to developing NLP tools for Yor\`ub\'a and its dialects, and potentially for other African languages, by improving our understanding of existing challenges and offering a high-quality dataset for further development. We release YOR\`ULECT dataset and models publicly under an open license.
Authors: Xin Su, Man Luo, Kris W Pan, Tien Pei Chou, Vasudev Lal, Phillip Howard
Abstract: Synthetic data generation has gained significant attention recently for its utility in training large vision and language models. However, the application of synthetic data to the training of multimodal context-augmented generation systems has been relatively unexplored. This gap in existing work is important because existing vision and language models (VLMs) are not trained specifically for context-augmented generation. Resources for adapting such models are therefore crucial for enabling their use in retrieval-augmented generation (RAG) settings, where a retriever is used to gather relevant information that is then subsequently provided to a generative model via context augmentation. To address this challenging problem, we generate SK-VQA: a large synthetic multimodal dataset containing over 2 million question-answer pairs which require external knowledge to determine the final answer. Our dataset is both larger and significantly more diverse than existing resources of its kind, possessing over 11x more unique questions and containing images from a greater variety of sources than previously-proposed datasets. Through extensive experiments, we demonstrate that our synthetic dataset can not only serve as a challenging benchmark, but is also highly effective for adapting existing generative multimodal models for context-augmented generation.
Authors: Hongzhan Lin, Ang Lv, Yuhan Chen, Chen Zhu, Yang Song, Hengshu Zhu, Rui Yan
Abstract: Many studies have revealed that large language models (LLMs) exhibit uneven awareness of different contextual positions.Their limited context awareness can lead to overlooking critical information and subsequent task failures. While several approaches have been proposed to enhance LLMs' context awareness, achieving both effectiveness and efficiency remains challenging.In this paper, for LLMs utilizing RoPE as position embeddings, we introduce a novel method called ``Mixture of In-Context Experts'' (MoICE) to address this challenge. MoICE comprises two key components: a router integrated into each attention head within LLMs and a lightweight router-only training optimization strategy: (1) MoICE views each RoPE angle as an `in-context' expert, demonstrated to be capable of directing the attention of a head to specific contextual positions. Consequently, each attention head flexibly processes tokens using multiple RoPE angles dynamically selected by the router to attend to the needed positions. This approach mitigates the risk of overlooking essential contextual information. (2) The router-only training strategy entails freezing LLM parameters and exclusively updating routers for only a few steps. When applied to open-source LLMs including Llama and Mistral, MoICE surpasses prior methods across multiple tasks on long context understanding and generation, all while maintaining commendable inference efficiency.
Authors: Pedro Faustini, Shakila Mahjabin Tonni, Annabelle McIver, Qiongkai Xu, Mark Dras
Abstract: Natural language processing (NLP) models may leak private information in different ways, including membership inference, reconstruction or attribute inference attacks. Sensitive information may not be explicit in the text, but hidden in underlying writing characteristics. Methods to protect privacy can involve using representations inside models that are demonstrated not to detect sensitive attributes or -- for instance, in cases where users might not trust a model, the sort of scenario of interest here -- changing the raw text before models can have access to it. The goal is to rewrite text to prevent someone from inferring a sensitive attribute (e.g. the gender of the author, or their location by the writing style) whilst keeping the text useful for its original intention (e.g. the sentiment of a product review). The few works tackling this have focused on generative techniques. However, these often create extensively different texts from the original ones or face problems such as mode collapse. This paper explores a novel adaptation of adversarial attack techniques to manipulate a text to deceive a classifier w.r.t one task (privacy) whilst keeping the predictions of another classifier trained for another task (utility) unchanged. We propose IDT, a method that analyses predictions made by auxiliary and interpretable models to identify which tokens are important to change for the privacy task, and which ones should be kept for the utility task. We evaluate different datasets for NLP suitable for different tasks. Automatic and human evaluations show that IDT retains the utility of text, while also outperforming existing methods when deceiving a classifier w.r.t privacy task.
Authors: Zhe Hu, Hou Pong Chan, Jing Li, Yu Yin
Abstract: Writing persuasive arguments is a challenging task for both humans and machines. It entails incorporating high-level beliefs from various perspectives on the topic, along with deliberate reasoning and planning to construct a coherent narrative. Current language models often generate surface tokens autoregressively, lacking explicit integration of these underlying controls, resulting in limited output diversity and coherence. In this work, we propose a persona-based multi-agent framework for argument writing. Inspired by the human debate, we first assign each agent a persona representing its high-level beliefs from a unique perspective, and then design an agent interaction process so that the agents can collaboratively debate and discuss the idea to form an overall plan for argument writing. Such debate process enables fluid and nonlinear development of ideas. We evaluate our framework on argumentative essay writing. The results show that our framework can generate more diverse and persuasive arguments through both automatic and human evaluations.
Authors: Xuanming Zhang, Anthony Diaz, Zixun Chen, Qingyang Wu, Kun Qian, Erik Voss, Zhou Yu
Abstract: Coherence in writing, an aspect that second-language (L2) English learners often struggle with, is crucial in assessing L2 English writing. Existing automated writing evaluation systems primarily use basic surface linguistic features to detect coherence in writing. However, little effort has been made to correct the detected incoherence, which could significantly benefit L2 language learners seeking to improve their writing. To bridge this gap, we introduce DECOR, a novel benchmark that includes expert annotations for detecting incoherence in L2 English writing, identifying the underlying reasons, and rewriting the incoherent sentences. To our knowledge, DECOR is the first coherence assessment dataset specifically designed for improving L2 English writing, featuring pairs of original incoherent sentences alongside their expert-rewritten counterparts. Additionally, we fine-tuned models to automatically detect and rewrite incoherence in student essays. We find that incorporating specific reasons for incoherence during fine-tuning consistently improves the quality of the rewrites, achieving a result that is favored in both automatic and human evaluations.
Authors: Krishna C. Puvvada, Piotr \.Zelasko, He Huang, Oleksii Hrinchuk, Nithin Rao Koluguri, Kunal Dhawan, Somshubra Majumdar, Elena Rastorgueva, Zhehuai Chen, Vitaly Lavrukhin, Jagadeesh Balam, Boris Ginsburg
Abstract: Recent advances in speech recognition and translation rely on hundreds of thousands of hours of Internet speech data. We argue that state-of-the art accuracy can be reached without relying on web-scale data. Canary - multilingual ASR and speech translation model, outperforms current state-of-the-art models - Whisper, OWSM, and Seamless-M4T on English, French, Spanish, and German languages, while being trained on an order of magnitude less data than these models. Three key factors enables such data-efficient model: (1) a FastConformer-based attention encoder-decoder architecture (2) training on synthetic data generated with machine translation and (3) advanced training techniques: data-balancing, dynamic data blending, dynamic bucketing and noise-robust fine-tuning. The model, weights, and training code will be open-sourced.
Authors: Ludovic Tanguy (CLLE), C\'ecile Fabre (CLLE), Nabil Hathout (UT, CNRS, CLLE), Lydia-Mai Ho-Dac (CLLE)
Abstract: In this study, we propose a corpus analysis of an area of the French lexicon that is both dense and highly structured: the vocabulary of family relationships. Starting with a lexicon of 25 nouns designating the main relationships (son, cousin, mother, grandfather, sister-in-law etc.), we examine how these terms are positioned in relation to each other through distributional analyses based on the use of these terms in corpora. We show that distributional information can capture certain features that organize this vocabulary (descent, alliance, siblings, genre), in ways that vary according to the different corpora compared.
Authors: Ludovic Tanguy (CLLE), C\'eline Poudat (BCL), Lydia-Mai Ho-Dac (CLLE)
Abstract: Our study focuses on Wikipedia talk pages, from a global perspective analyzing contributors' behaviors in online interactions. Using a corpus comprising all Wikipedia talk pages in French, totaling more than 300,000 discussion threads, we examine how discussions with more than two participants (multiparty conversation) unfold and we specifically investigate the role of a third participant's intervention when two Wikipedians have already initiated an exchange. In this regard, we concentrate on the sequential structure of these interactions in terms of articulation among different participants and aim to specify this third message by exploring its lexical particularities, while also proposing an initial typology of the third participant's message role and how it aligns with preceding messages.
Authors: Orgest Xhelili, Yihong Liu, Hinrich Sch\"utze
Abstract: Multilingual pre-trained models (mPLMs) have shown impressive performance on cross-lingual transfer tasks. However, the transfer performance is often hindered when a low-resource target language is written in a different script than the high-resource source language, even though the two languages may be related or share parts of their vocabularies. Inspired by recent work that uses transliteration to address this problem, our paper proposes a transliteration-based post-pretraining alignment (PPA) method aiming to improve the cross-lingual alignment between languages using diverse scripts. We select two areal language groups, $\textbf{Mediterranean-Amharic-Farsi}$ and $\textbf{South+East Asian Languages}$, wherein the languages are mutually influenced but use different scripts. We apply our method to these language groups and conduct extensive experiments on a spectrum of downstream tasks. The results show that after PPA, models consistently outperform the original model (up to 50% for some tasks) in English-centric transfer. In addition, when we use languages other than English as sources in transfer, our method obtains even larger improvements. We will make our code and models publicly available at \url{https://github.com/cisnlp/Transliteration-PPA}.
Authors: Bryan Wilie, Samuel Cahyawijaya, Etsuko Ishii, Junxian He, Pascale Fung
Abstract: The capability to reason from text is crucial for real-world NLP applications. Real-world scenarios often involve incomplete or evolving data. In response, individuals update their beliefs and understandings accordingly. However, most existing evaluations assume that language models (LMs) operate with consistent information. We introduce Belief-R, a new dataset designed to test LMs' belief revision ability when presented with new evidence. Inspired by how humans suppress prior inferences, this task assesses LMs within the newly proposed delta reasoning ($\Delta R$) framework. Belief-R features sequences of premises designed to simulate scenarios where additional information could necessitate prior conclusions drawn by LMs. We evaluate $\sim$30 LMs across diverse prompting strategies and found that LMs generally struggle to appropriately revise their beliefs in response to new information. Further, models adept at updating often underperformed in scenarios without necessary updates, highlighting a critical trade-off. These insights underscore the importance of improving LMs' adaptiveness to changing information, a step toward more reliable AI systems.
Authors: Yixing Li, Yuxian Gu, Li Dong, Dequan Wang, Yu Cheng, Furu Wei
Abstract: In the field of large language models (LLMs), Knowledge Distillation (KD) is a critical technique for transferring capabilities from teacher models to student models. However, existing KD methods face limitations and challenges in distillation of LLMs, including efficiency and insufficient measurement capabilities of traditional KL divergence. It is shown that LLMs can serve as an implicit reward function, which we define as a supplement to KL divergence. In this work, we propose Direct Preference Knowledge Distillation (DPKD) for LLMs. DPKD utilizes distribution divergence to represent the preference loss and implicit reward function. We re-formulate KD of LLMs into two stages: first optimizing and objective consisting of implicit reward and reverse KL divergence and then improving the preference probability of teacher outputs over student outputs. We conducted experiments and analysis on various datasets with LLM parameters ranging from 120M to 13B and demonstrate the broad applicability and effectiveness of our DPKD approach. Meanwhile, we prove the value and effectiveness of the introduced implicit reward and output preference in KD through experiments and theoretical analysis. The DPKD method outperforms the baseline method in both output response precision and exact match percentage. Code and data are available at https://aka.ms/dpkd.
URLs: https://aka.ms/dpkd.
Authors: Mohammad Javad Hosseini, Yang Gao, Tim Baumg\"artner, Alex Fabrikant, Reinald Kim Amplayo
Abstract: Segmenting text into fine-grained units of meaning is important to a wide range of NLP applications. The default approach of segmenting text into sentences is often insufficient, especially since sentences are usually complex enough to include multiple units of meaning that merit separate treatment in the downstream task. We focus on the task of abstractive proposition segmentation: transforming text into simple, self-contained, well-formed sentences. Several recent works have demonstrated the utility of proposition segmentation with few-shot prompted LLMs for downstream tasks such as retrieval-augmented grounding and fact verification. However, this approach does not scale to large amounts of text and may not always extract all the facts from the input text. In this paper, we first introduce evaluation metrics for the task to measure several dimensions of quality. We then propose a scalable, yet accurate, proposition segmentation model. We model proposition segmentation as a supervised task by training LLMs on existing annotated datasets and show that training yields significantly improved results. We further show that by using the fine-tuned LLMs as teachers for annotating large amounts of multi-domain synthetic distillation data, we can train smaller student models with results similar to the teacher LLMs. We then demonstrate that our technique leads to effective domain generalization, by annotating data in two domains outside the original training data and evaluating on them. Finally, as a key contribution of the paper, we share an easy-to-use API for NLP practitioners to use.
Authors: Zheng Chu, Jingchang Chen, Qianglong Chen, Haotian Wang, Kun Zhu, Xiyuan Du, Weijiang Yu, Ming Liu, Bing Qin
Abstract: Large language models (LLMs) have demonstrated strong reasoning capabilities. Nevertheless, they still suffer from factual errors when tackling knowledge-intensive tasks. Retrieval-augmented reasoning represents a promising approach. However, significant challenges still persist, including inaccurate and insufficient retrieval for complex questions, as well as difficulty in integrating multi-source knowledge. To address this, we propose Beam Aggregation Reasoning, BeamAggR, a reasoning framework for knowledge-intensive multi-hop QA. BeamAggR explores and prioritizes promising answers at each hop of question. Concretely, we parse the complex questions into trees, which include atom and composite questions, followed by bottom-up reasoning. For atomic questions, the LLM conducts reasoning on multi-source knowledge to get answer candidates. For composite questions, the LLM combines beam candidates, explores multiple reasoning paths through probabilistic aggregation, and prioritizes the most promising trajectory. Extensive experiments on four open-domain multi-hop reasoning datasets show that our method significantly outperforms SOTA methods by 8.5%. Furthermore, our analysis reveals that BeamAggR elicits better knowledge collaboration and answer aggregation.
Authors: Walig\'ora Witold
Abstract: This paper introduces AnomaLLMy, a novel technique for the automatic detection of anomalous tokens in black-box Large Language Models (LLMs) with API-only access. Utilizing low-confidence single-token predictions as a cost-effective indicator, AnomaLLMy identifies irregularities in model behavior, addressing the issue of anomalous tokens degrading the quality and reliability of models. Validated on the cl100k_base dataset, the token set of GPT-4, AnomaLLMy detected 413 major and 65 minor anomalies, demonstrating the method's efficiency with just \$24.39 spent in API credits. The insights from this research are expected to be beneficial for enhancing the robustness of and accuracy of LLMs, particularly in the development and assessment of tokenizers.
Authors: Yutao Zhu, Kun Zhou, Kelong Mao, Wentong Chen, Yiding Sun, Zhipeng Chen, Qian Cao, Yihan Wu, Yushuo Chen, Feng Wang, Lei Zhang, Junyi Li, Xiaolei Wang, Lei Wang, Beichen Zhang, Zican Dong, Xiaoxue Cheng, Yuhan Chen, Xinyu Tang, Yupeng Hou, Qiangqiang Ren, Xincheng Pang, Shufang Xie, Wayne Xin Zhao, Zhicheng Dou, Jiaxin Mao, Yankai Lin, Ruihua Song, Jun Xu, Xu Chen, Rui Yan, Zhewei Wei, Di Hu, Wenbing Huang, Ze-Feng Gao, Yueguo Chen, Weizheng Lu, Ji-Rong Wen
Abstract: Large language models (LLMs) have become the foundation of many applications, leveraging their extensive capabilities in processing and understanding natural language. While many open-source LLMs have been released with technical reports, the lack of training details hinders further research and development. This paper presents the development of YuLan, a series of open-source LLMs with $12$ billion parameters. The base model of YuLan is pre-trained on approximately $1.7$T tokens derived from a diverse corpus, including massive English, Chinese, and multilingual texts. We design a three-stage pre-training method to enhance YuLan's overall capabilities. Subsequent phases of training incorporate instruction-tuning and human alignment, employing a substantial volume of high-quality synthesized data. To facilitate the learning of complex and long-tail knowledge, we devise a curriculum-learning framework throughout across these stages, which helps LLMs learn knowledge in an easy-to-hard manner. YuLan's training is finished on Jan, 2024 and has achieved performance on par with state-of-the-art LLMs across various English and Chinese benchmarks. This paper outlines a comprehensive technical roadmap for developing LLMs from scratch. Our model and codes are available at https://github.com/RUC-GSAI/YuLan-Chat.
Authors: Yang Xu, Yu Wang, Hao An, Zhichen Liu, Yongyuan Li
Abstract: Human and model-generated texts can be distinguished by examining the magnitude of likelihood in language. However, it is becoming increasingly difficult as language model's capabilities of generating human-like texts keep evolving. This study provides a new perspective by using the relative likelihood values instead of absolute ones, and extracting useful features from the spectrum-view of likelihood for the human-model text detection task. We propose a detection procedure with two classification methods, supervised and heuristic-based, respectively, which results in competitive performances with previous zero-shot detection methods and a new state-of-the-art on short-text detection. Our method can also reveal subtle differences between human and model languages, which find theoretical roots in psycholinguistics studies. Our code is available at https://github.com/CLCS-SUSTech/FourierGPT
Authors: Davide Turco, Conor Houghton
Abstract: This study explores the temporal dynamics of language processing by examining the alignment between word representations from a pre-trained transformer-based language model, and EEG data. Using a Temporal Response Function (TRF) model, we investigate how neural activity corresponds to model representations across different layers, revealing insights into the interaction between artificial language models and brain responses during language comprehension. Our analysis reveals patterns in TRFs from distinct layers, highlighting varying contributions to lexical and compositional processing. Additionally, we used linear discriminant analysis (LDA) to isolate part-of-speech (POS) representations, offering insights into their influence on neural responses and the underlying mechanisms of syntactic processing. These findings underscore EEG's utility for probing language processing dynamics with high temporal resolution. By bridging artificial language models and neural activity, this study advances our understanding of their interaction at fine timescales.
Authors: Erik Henriksson, Amanda Myntti, Anni Eskelinen, Selcen Erten-Johansson, Saara Hellstr\"om, Veronika Laippala
Abstract: This article explores deep learning models for the automatic identification of registers - text varieties such as news reports and discussion forums - in web-based datasets across 16 languages. Web register (or genre) identification would provide a robust solution for understanding the content of web-scale datasets, which have become crucial in computational linguistics. Despite recent advances, the potential of register classifiers on the noisy web remains largely unexplored, particularly in multilingual settings and when targeting the entire unrestricted web. We experiment with a range of deep learning models using the new Multilingual CORE corpora, which includes 16 languages annotated using a detailed, hierarchical taxonomy of 25 registers designed to cover the entire unrestricted web. Our models achieve state-of-the-art results, showing that a detailed taxonomy in a hierarchical multi-label setting can yield competitive classification performance. However, all models hit a glass ceiling at approximately 80% F1 score, which we attribute to the non-discrete nature of web registers and the inherent uncertainty in labeling some documents. By pruning ambiguous examples, we improve model performance to over 90%. Finally, multilingual models outperform monolingual ones, particularly benefiting languages with fewer training examples and smaller registers. Although a zero-shot setting decreases performance by an average of 7%, these drops are not linked to specific registers or languages. Instead, registers show surprising similarity across languages.
Authors: Jan Philip Wahle, Terry Ruas, Yang Xu, Bela Gipp
Abstract: Much of the success of modern language models depends on finding a suitable prompt to instruct the model. Until now, it has been largely unknown how variations in the linguistic expression of prompts affect these models. This study systematically and empirically evaluates which linguistic features influence models through paraphrase types, i.e., different linguistic changes at particular positions. We measure behavioral changes for five models across 120 tasks and six families of paraphrases (i.e., morphology, syntax, lexicon, lexico-syntax, discourse, and others). We also control for other prompt engineering factors (e.g., prompt length, lexical diversity, and proximity to training data). Our results show a potential for language models to improve tasks when their prompts are adapted in specific paraphrase types (e.g., 6.7% median gain in Mixtral 8x7B; 5.5% in LLaMA 3 8B). In particular, changes in morphology and lexicon, i.e., the vocabulary used, showed promise in improving prompts. These findings contribute to developing more robust language models capable of handling variability in linguistic expression.
Authors: Garima Dhanania, Sheshera Mysore, Chau Minh Pham, Mohit Iyyer, Hamed Zamani, Andrew McCallum
Abstract: Topic models are widely used to analyze document collections. While they are valuable for discovering latent topics in a corpus when analysts are unfamiliar with the corpus, analysts also commonly start with an understanding of the content present in a corpus. This may be through categories obtained from an initial pass over the corpus or a desire to analyze the corpus through a predefined set of categories derived from a high level theoretical framework (e.g. political ideology). In these scenarios analysts desire a topic modeling approach which incorporates their understanding of the corpus while supporting various forms of interaction with the model. In this work, we present EdTM, as an approach for label name supervised topic modeling. EdTM models topic modeling as an assignment problem while leveraging LM/LLM based document-topic affinities and using optimal transport for making globally coherent topic-assignments. In experiments, we show the efficacy of our framework compared to few-shot LLM classifiers, and topic models based on clustering and LDA. Further, we show EdTM's ability to incorporate various forms of analyst feedback and while remaining robust to noisy analyst inputs.
Authors: Chuanqi Cheng, Jian Guan, Wei Wu, Rui Yan
Abstract: We explore multi-step reasoning in vision-language models (VLMs). The problem is challenging, as reasoning data consisting of multiple steps of visual and language processing are barely available. To overcome the challenge, we first introduce a least-to-most visual reasoning paradigm, which interleaves steps of decomposing a question into sub-questions and invoking external tools for resolving sub-questions. Based on the paradigm, we further propose a novel data synthesis approach that can automatically create questions and multi-step reasoning paths for an image in a bottom-up manner. Our approach divides the complex synthesis task into a few simple sub-tasks, and (almost entirely) relies on open-sourced models to accomplish the sub-tasks. Therefore, the entire synthesis process is reproducible and cost-efficient, and the synthesized data is quality guaranteed. With the approach, we construct $50$k visual reasoning examples. Then, we develop a visual reasoner through supervised fine-tuning, which is capable of generally enhancing the reasoning abilities of a wide range of existing VLMs in a plug-and-play fashion. Extensive experiments indicate that the visual reasoner can consistently and significantly improve four VLMs on four VQA benchmarks. Our code and dataset are available at https://github.com/steven-ccq/VisualReasoner.
Authors: Jiazheng Li, Hainiu Xu, Zhaoyue Sun, Yuxiang Zhou, David West, Cesare Aloisi, Yulan He
Abstract: Generating rationales that justify scoring decisions has been a promising way to facilitate explainability in automated scoring systems. However, existing methods do not match the accuracy of classifier-based methods. Plus, the generated rationales often contain hallucinated information. To address these issues, we propose a novel framework capable of generating more faithful rationales and, more importantly, matching performance with classifier-based black-box scoring systems. We first mimic the human assessment process by querying Large Language Models (LLMs) to generate a thought tree. We then summarise intermediate assessment decisions from each thought tree path for creating synthetic rationale data and rationale preference data. Finally, we utilise the generated synthetic data to calibrate LLMs through a two-step training process: supervised fine-tuning and preference optimization. Extensive experimental results demonstrate that our framework achieves a 38% assessment performance improvement in the QWK score compared to prior work while producing higher-quality rationales, as recognised by human evaluators and LLMs. Our work sheds light on the effectiveness of performing preference optimization using synthetic preference data obtained from thought tree paths.
Authors: Dami\'an Ariel Furman, Juan Junqueras, Z. Bur\c{c}e G\"um\"usl\"u, Edgar Altszyler, Joaquin Navajas, Ophelia Deroy, Justin Sulik
Abstract: We present Reasons For and Against Vaccination (RFAV), a dataset for predicting reasons for and against vaccination, and scientific authorities used to justify them, annotated through nichesourcing and augmented using GPT4 and GPT3.5-Turbo. We show how it is possible to mine these reasons in non-structured text, under different task definitions, despite the high level of subjectivity involved and explore the impact of artificially augmented data using in-context learning with GPT4 and GPT3.5-Turbo. We publish the dataset and the trained models along with the annotation manual used to train annotators and define the task.
Authors: Zhehuai Chen, He Huang, Oleksii Hrinchuk, Krishna C. Puvvada, Nithin Rao Koluguri, Piotr \.Zelasko, Jagadeesh Balam, Boris Ginsburg
Abstract: Incorporating speech understanding capabilities into pretrained large-language models has become a vital research direction (SpeechLLM). The previous architectures can be categorized as: i) GPT-style, prepend speech prompts to the text prompts as a sequence of LLM inputs like a decoder-only model; ii) T5-style, introduce speech cross-attention to each layer of the pretrained LLMs. We propose BESTOW architecture to bring the BESt features from TwO Worlds into a single model that is highly efficient and has strong multitask capabilities. Moreover, there is no clear streaming solution for either style, especially considering the solution should generalize to speech multitask. We reformulate streamable SpeechLLM as a read-write policy problem and unifies the offline and streaming research with BESTOW architecture. Hence we demonstrate the first open-source SpeechLLM solution that enables Streaming and Multitask at scale (beyond ASR) at the same time. This streamable solution achieves very strong performance on a wide range of speech tasks (ASR, AST, SQA, unseen DynamicSuperb). It is end-to-end optimizable, with lower training/inference cost, and demonstrates LLM knowledge transferability to speech.
Authors: Shen Gao, Yuntao Wen, Minghang Zhu, Jianing Wei, Yuhan Cheng, Qunzi Zhang, Shuo Shang
Abstract: Most economic theories typically assume that financial market participants are fully rational individuals and use mathematical models to simulate human behavior in financial markets. However, human behavior is often not entirely rational and is challenging to predict accurately with mathematical models. In this paper, we propose \textbf{A}gent-based \textbf{S}imulated \textbf{F}inancial \textbf{M}arket (ASFM), which first constructs a simulated stock market with a real order matching system. Then, we propose a large language model based agent as the stock trader, which contains the profile, observation, and tool-learning based action module. The trading agent can comprehensively understand current market dynamics and financial policy information, and make decisions that align with their trading strategy. In the experiments, we first verify that the reactions of our ASFM are consistent with the real stock market in two controllable scenarios. In addition, we also conduct experiments in two popular economics research directions, and we find that conclusions drawn in our \model align with the preliminary findings in economics research. Based on these observations, we believe our proposed ASFM provides a new paradigm for economic research.
Authors: Tzuf Paz-Argaman, John Palowitch, Sayali Kulkarni, Reut Tsarfaty, Jason Baldridge
Abstract: Similar to vision-and-language navigation (VLN) tasks that focus on bridging the gap between vision and language for embodied navigation, the new Rendezvous (RVS) task requires reasoning over allocentric spatial relationships (independent of the observer's viewpoint) using non-sequential navigation instructions and maps. However, performance substantially drops in new environments with no training data. Using opensource descriptions paired with coordinates (e.g., Wikipedia) provides training data but suffers from limited spatially-oriented text resulting in low geolocation resolution. We propose a large-scale augmentation method for generating high-quality synthetic data for new environments using readily available geospatial data. Our method constructs a grounded knowledge-graph, capturing entity relationships. Sampled entities and relations (`shop north of school') generate navigation instructions via (i) generating numerous templates using context-free grammar (CFG) to embed specific entities and relations; (ii) feeding the entities and relation into a large language model (LLM) for instruction generation. A comprehensive evaluation on RVS, showed that our approach improves the 100-meter accuracy by 45.83% on unseen environments. Furthermore, we demonstrate that models trained with CFG-based augmentation achieve superior performance compared with those trained with LLM-based augmentation, both in unseen and seen environments. These findings suggest that the potential advantages of explicitly structuring spatial information for text-based geospatial reasoning in previously unknown, can unlock data-scarce scenarios.
Authors: Habib Hajimolahoseini, Mohammad Hassanpour, Foozhan Ataiefard, Boxing Chen, Yang Liu
Abstract: This paper introduces a novel method of Progressive Low Rank Decomposition (PLRD) tailored for the compression of large language models. Our approach leverages a pre-trained model, which is then incrementally decompressed to smaller sizes using progressively lower ranks. This method allows for significant reductions in computational overhead and energy consumption, as subsequent models are derived from the original without the need for retraining from scratch. We detail the implementation of PLRD, which strategically decreases the tensor ranks, thus optimizing the trade-off between model performance and resource usage. The efficacy of PLRD is demonstrated through extensive experiments showing that models trained with PLRD method on only 1B tokens maintain comparable performance with traditionally trained models while using 0.1% of the tokens. The versatility of PLRD is highlighted by its ability to generate multiple model sizes from a single foundational model, adapting fluidly to varying computational and memory budgets. Our findings suggest that PLRD could set a new standard for the efficient scaling of LLMs, making advanced AI more feasible on diverse platforms.
Authors: Xinyi Chen, Baohao Liao, Jirui Qi, Panagiotis Eustratiadis, Christof Monz, Arianna Bisazza, Maarten de Rijke
Abstract: Following multiple instructions is a crucial ability for large language models (LLMs). Evaluating this ability comes with significant challenges: (i) limited coherence between multiple instructions, (ii) positional bias where the order of instructions affects model performance, and (iii) a lack of objectively verifiable tasks. To address these issues, we introduce a benchmark designed to evaluate models' abilities to follow multiple instructions through sequential instruction following (SIFo) tasks. In SIFo, the successful completion of multiple instructions is verifiable by examining only the final instruction. Our benchmark evaluates instruction following using four tasks (text modification, question answering, mathematics, and security rule following), each assessing different aspects of sequential instruction following. Our evaluation of popular LLMs, both closed-source and open-source, shows that more recent and larger models significantly outperform their older and smaller counterparts on the SIFo tasks, validating the benchmark's effectiveness. All models struggle with following sequences of instructions, hinting at an important lack of robustness of today's language models.
Authors: Yuxiang Zhang, Jing Chen, Junjie Wang, Yaxin Liu, Cheng Yang, Chufan Shi, Xinyu Zhu, Zihao Lin, Hanwen Wan, Yujiu Yang, Tetsuya Sakai, Tian Feng, Hayato Yamana
Abstract: Tool-augmented large language models (LLMs) are rapidly being integrated into real-world applications. Due to the lack of benchmarks, the community still needs to fully understand the hallucination issues within these models. To address this challenge, we introduce a comprehensive diagnostic benchmark, ToolBH. Specifically, we assess the LLM's hallucinations through two perspectives: depth and breadth. In terms of depth, we propose a multi-level diagnostic process, including (1) solvability detection, (2) solution planning, and (3) missing-tool analysis. For breadth, we consider three scenarios based on the characteristics of the toolset: missing necessary tools, potential tools, and limited functionality tools. Furthermore, we developed seven tasks and collected 700 evaluation samples through multiple rounds of manual annotation. The results show the significant challenges presented by the ToolBH benchmark. The current advanced models Gemini-1.5-Pro and GPT-4o only achieve a total score of 45.3 and 37.0, respectively, on a scale of 100. In this benchmark, larger model parameters do not guarantee better performance; the training data and response strategies also play a crucial role in tool-enhanced LLM scenarios. Our diagnostic analysis indicates that the primary reason for model errors lies in assessing task solvability. Additionally, open-weight models suffer from performance drops with verbose replies, whereas proprietary models excel with longer reasoning.
Authors: Renzhi Wang, Piji Li
Abstract: Large language models (LLMs) require continual knowledge updates to stay abreast of the ever-changing world facts, prompting the formulation of lifelong model editing task. While recent years have witnessed the development of various techniques for single and batch editing, these methods either fail to apply or perform sub-optimally when faced with lifelong editing. In this paper, we introduce LEMoE, an advanced Mixture of Experts (MoE) adaptor for lifelong model editing. We first analyze the factors influencing the effectiveness of conventional MoE adaptor in lifelong editing, including catastrophic forgetting, inconsistent routing and order sensitivity. Based on these insights, we propose a tailored module insertion method to achieve lifelong editing, incorporating a novel KV anchor routing to enhance routing consistency between training and inference stage, along with a concise yet effective clustering-based editing order planning. Experimental results demonstrate the effectiveness of our method in lifelong editing, surpassing previous model editing techniques while maintaining outstanding performance in batch editing task. Our code will be available.
Authors: Chen Tang, Bohao Yang, Kun Zhao, Bo Lv, Chenghao Xiao, Frank Guerin, Chenghua Lin
Abstract: Named entity recognition (NER) stands as a fundamental and pivotal task within the realm of Natural Language Processing. Particularly within the domain of Biomedical Method NER, this task presents notable challenges, stemming from the continual influx of domain-specific terminologies in scholarly literature. Current research in Biomedical Method (BioMethod) NER suffers from a scarcity of resources, primarily attributed to the intricate nature of methodological concepts, which necessitate a profound understanding for precise delineation. In this study, we propose a novel dataset for biomedical method entity recognition, employing an automated BioMethod entity recognition and information retrieval system to assist human annotation. Furthermore, we comprehensively explore a range of conventional and contemporary open-domain NER methodologies, including the utilization of cutting-edge large-scale language models (LLMs) customised to our dataset. Our empirical findings reveal that the large parameter counts of language models surprisingly inhibit the effective assimilation of entity extraction patterns pertaining to biomedical methods. Remarkably, the approach, leveraging the modestly sized ALBERT model (only 11MB), in conjunction with conditional random fields (CRF), achieves state-of-the-art (SOTA) performance.
Authors: Kelly Marchisio, Wei-Yin Ko, Alexandre B\'erard, Th\'eo Dehaze, Sebastian Ruder
Abstract: We investigate a surprising limitation of LLMs: their inability to consistently generate text in a user's desired language. We create the Language Confusion Benchmark (LCB) to evaluate such failures, covering 15 typologically diverse languages with existing and newly-created English and multilingual prompts. We evaluate a range of LLMs on monolingual and cross-lingual generation reflecting practical use cases, finding that Llama Instruct and Mistral models exhibit high degrees of language confusion and even the strongest models fail to consistently respond in the correct language. We observe that base and English-centric instruct models are more prone to language confusion, which is aggravated by complex prompts and high sampling temperatures. We find that language confusion can be partially mitigated via few-shot prompting, multilingual SFT and preference tuning. We release our language confusion benchmark, which serves as a first layer of efficient, scalable multilingual evaluation at https://github.com/for-ai/language-confusion.
Authors: Bastien Li\'etard, Pascal Denis, Mikaella Keller
Abstract: Polysemy and synonymy are two crucial interrelated facets of lexical ambiguity. While both phenomena have been studied extensively in NLP, leading to dedicated systems, they are often been considered independently. While many tasks dealing with polysemy (e.g. Word Sense Disambiguiation or Induction) highlight the role of a word's senses, the study of synonymy is rooted in the study of concepts, i.e. meaning shared across the lexicon. In this paper, we introduce Concept Induction, the unsupervised task of learning a soft clustering among words that defines a set of concepts directly from data. This task generalizes that of Word Sense Induction. We propose a bi-level approach to Concept Induction that leverages both a local lemma-centric view and a global cross-lexicon perspective to induce concepts. We evaluate the obtained clustering on SemCor's annotated data and obtain good performances (BCubed F1 above 0.60). We find that the local and the global levels are mutually beneficial to induce concepts and also senses in our setting. Finally, we create static embeddings representing our induced concepts and use them on the Word-in-Context task, obtaining competitive performances with the State-of-the-Art.
Authors: Sujan Dutta, Sayantan Mahinder, Raviteja Anantha, Bortik Bandyopadhyay
Abstract: Reinforcement Learning from AI Feedback (RLAIF) has demonstrated significant potential across various domains, including mitigating harm in LLM outputs, enhancing text summarization, and mathematical reasoning. This paper introduces an RLAIF framework for improving the code generation abilities of lightweight (<1B parameters) LLMs. We specifically focus on code generation tasks that require writing appropriate API calls, which is challenging due to the well-known issue of hallucination in LLMs. Our framework extracts AI feedback from a larger LLM (e.g., GPT-3.5) through a specialized prompting strategy and uses this data to train a reward model towards better alignment from smaller LLMs. We run our experiments on the Gorilla dataset and meticulously assess the quality of the model-generated code across various metrics, including AST, ROUGE, and Code-BLEU, and develop a pipeline to compute its executability rate accurately. Our approach significantly enhances the fine-tuned LLM baseline's performance, achieving a 4.5% improvement in executability rate. Notably, a smaller LLM model (780M parameters) trained with RLAIF surpasses a much larger fine-tuned baseline with 7B parameters, achieving a 1.0% higher code executability rate.
Authors: Anisha Gunjal, Greg Durrett
Abstract: Automatic factuality verification of large language model (LLM) generations is becoming more and more widely used to combat hallucinations. A major point of tension in the literature is the granularity of this fact-checking: larger chunks of text are hard to fact-check, but more atomic facts like propositions may lack context to interpret correctly. In this work, we assess the role of context in these atomic facts. We argue that fully atomic facts are not the right representation, and define two criteria for molecular facts: decontextuality, or how well they can stand alone, and minimality, or how little extra information is added to achieve decontexuality. We quantify the impact of decontextualization on minimality, then present a baseline methodology for generating molecular facts automatically, aiming to add the right amount of information. We compare against various methods of decontextualization and find that molecular facts balance minimality with fact verification accuracy in ambiguous settings.
Authors: Sheridan Feucht, David Atkinson, Byron Wallace, David Bau
Abstract: LLMs process text as sequences of tokens that roughly correspond to words, where less common words are represented by multiple tokens. However, individual tokens are often semantically unrelated to the meanings of the words/concepts they comprise. For example, Llama-2-7b's tokenizer splits the word "northeastern" into the tokens ['_n', 'ort', 'he', 'astern'], none of which correspond to semantically meaningful units like "north" or "east." Similarly, the overall meanings of named entities like "Neil Young" and multi-word expressions like "break a leg" cannot be directly inferred from their constituent tokens. Mechanistically, how do LLMs convert such arbitrary groups of tokens into useful higher-level representations? In this work, we find that last token representations of named entities and multi-token words exhibit a pronounced "erasure" effect, where information about previous and current tokens is rapidly forgotten in early layers. Using this observation, we propose a method to "read out" the implicit vocabulary of an autoregressive LLM by examining differences in token representations across layers, and present results of this method for Llama-2-7b and Llama-3-8B. To our knowledge, this is the first attempt to probe the implicit vocabulary of an LLM.
Authors: Xin Chan, Xiaoyang Wang, Dian Yu, Haitao Mi, Dong Yu
Abstract: We propose a novel persona-driven data synthesis methodology that leverages various perspectives within a large language model (LLM) to create diverse synthetic data. To fully exploit this methodology at scale, we introduce Persona Hub -- a collection of 1 billion diverse personas automatically curated from web data. These 1 billion personas (~13% of the world's total population), acting as distributed carriers of world knowledge, can tap into almost every perspective encapsulated within the LLM, thereby facilitating the creation of diverse synthetic data at scale for various scenarios. By showcasing Persona Hub's use cases in synthesizing high-quality mathematical and logical reasoning problems, instructions (i.e., user prompts), knowledge-rich texts, game NPCs and tools (functions) at scale, we demonstrate persona-driven data synthesis is versatile, scalable, flexible, and easy to use, potentially driving a paradigm shift in synthetic data creation and applications in practice, which may have a profound impact on LLM research and development.
Authors: Hetvi Waghela, Jaydip Sen, Sneha Rakshit
Abstract: In this paper, we introduce an enhanced textual adversarial attack method, known as Saliency Attention and Semantic Similarity driven adversarial Perturbation (SASSP). The proposed scheme is designed to improve the effectiveness of contextual perturbations by integrating saliency, attention, and semantic similarity. Traditional adversarial attack methods often struggle to maintain semantic consistency and coherence while effectively deceiving target models. Our proposed approach addresses these challenges by incorporating a three-pronged strategy for word selection and perturbation. First, we utilize a saliency-based word selection to prioritize words for modification based on their importance to the model's prediction. Second, attention mechanisms are employed to focus perturbations on contextually significant words, enhancing the attack's efficacy. Finally, an advanced semantic similarity-checking method is employed that includes embedding-based similarity and paraphrase detection. By leveraging models like Sentence-BERT for embedding similarity and fine-tuned paraphrase detection models from the Sentence Transformers library, the scheme ensures that the perturbed text remains contextually appropriate and semantically consistent with the original. Empirical evaluations demonstrate that SASSP generates adversarial examples that not only maintain high semantic fidelity but also effectively deceive state-of-the-art natural language processing models. Moreover, in comparison to the original scheme of contextual perturbation CLARE, SASSP has yielded a higher attack success rate and lower word perturbation rate.
Authors: Selene Baez Santamaria, Shihan Wang, Piek Vossen
Abstract: We develop an artificial agent motivated to augment its knowledge base beyond its initial training. The agent actively participates in dialogues with other agents, strategically acquiring new information. The agent models its knowledge as an RDF knowledge graph, integrating new beliefs acquired through conversation. Responses in dialogue are generated by identifying graph patterns around these new integrated beliefs. We show that policies can be learned using reinforcement learning to select effective graph patterns during an interaction, without relying on explicit user feedback. Within this context, our study is a proof of concept for leveraging users as effective sources of information.
Authors: Faruk Ahmed, Andrew Sellergren, Lin Yang, Shawn Xu, Boris Babenko, Abbi Ward, Niels Olson, Arash Mohtashamian, Yossi Matias, Greg S. Corrado, Quang Duong, Dale R. Webster, Shravya Shetty, Daniel Golden, Yun Liu, David F. Steiner, Ellery Wulczyn
Abstract: Microscopic interpretation of histopathology images underlies many important diagnostic and treatment decisions. While advances in vision-language modeling raise new opportunities for analysis of such images, the gigapixel-scale size of whole slide images (WSIs) introduces unique challenges. Additionally, pathology reports simultaneously highlight key findings from small regions while also aggregating interpretation across multiple slides, often making it difficult to create robust image-text pairs. As such, pathology reports remain a largely untapped source of supervision in computational pathology, with most efforts relying on region-of-interest annotations or self-supervision at the patch-level. In this work, we develop a vision-language model based on the BLIP-2 framework using WSIs paired with curated text from pathology reports. This enables applications utilizing a shared image-text embedding space, such as text or image retrieval for finding cases of interest, as well as integration of the WSI encoder with a frozen large language model (LLM) for WSI-based generative text capabilities such as report generation or AI-in-the-loop interactions. We utilize a de-identified dataset of over 350,000 WSIs and diagnostic text pairs, spanning a wide range of diagnoses, procedure types, and tissue types. We present pathologist evaluation of text generation and text retrieval using WSI embeddings, as well as results for WSI classification and workflow prioritization (slide-level triaging). Model-generated text for WSIs was rated by pathologists as accurate, without clinically significant error or omission, for 78% of WSIs on average. This work demonstrates exciting potential capabilities for language-aligned WSI embeddings.
Authors: Sion Yoon, Tae Eun Kim, Yoo Jung Oh
Abstract: The dynamics of human-AI communication have been reshaped by language models such as ChatGPT. However, extant research has primarily focused on dyadic communication, leaving much to be explored regarding the dynamics of human-AI communication in group settings. The availability of multiple language model chatbots presents a unique opportunity for scholars to better understand the interaction between humans and multiple chatbots. This study examines the impact of multi-chatbot communication in a specific persuasion setting: promoting charitable donations. We developed an online environment that enables multi-chatbot communication and conducted a pilot experiment utilizing two GPT-based chatbots, Save the Children and UNICEF chatbots, to promote charitable donations. In this study, we present our development process of the multi-chatbot interface and present preliminary findings from a pilot experiment. Analysis of qualitative and quantitative feedback are presented, and limitations are addressed.
Authors: Ferhat Ozgur Catak, Murat Kuzlu
Abstract: Uncertainty quantification approaches have been more critical in large language models (LLMs), particularly high-risk applications requiring reliable outputs. However, traditional methods for uncertainty quantification, such as probabilistic models and ensemble techniques, face challenges when applied to the complex and high-dimensional nature of LLM-generated outputs. This study proposes a novel geometric approach to uncertainty quantification using convex hull analysis. The proposed method leverages the spatial properties of response embeddings to measure the dispersion and variability of model outputs. The prompts are categorized into three types, i.e., `easy', `moderate', and `confusing', to generate multiple responses using different LLMs at varying temperature settings. The responses are transformed into high-dimensional embeddings via a BERT model and subsequently projected into a two-dimensional space using Principal Component Analysis (PCA). The Density-Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm is utilized to cluster the embeddings and compute the convex hull for each selected cluster. The experimental results indicate that the uncertainty of the model for LLMs depends on the prompt complexity, the model, and the temperature setting.
Authors: Jihao Liu, Xin Huang, Jinliang Zheng, Boxiao Liu, Jia Wang, Osamu Yoshie, Yu Liu, Hongsheng Li
Abstract: This paper introduces MM-Instruct, a large-scale dataset of diverse and high-quality visual instruction data designed to enhance the instruction-following capabilities of large multimodal models (LMMs). While existing visual instruction datasets often focus on question-answering, they struggle to generalize to broader application scenarios such as creative writing, summarization, or image analysis. To address these limitations, we propose a novel approach to constructing MM-Instruct that leverages the strong instruction-following capabilities of existing LLMs to generate novel visual instruction data from large-scale but conventional image captioning datasets. MM-Instruct first leverages ChatGPT to automatically generate diverse instructions from a small set of seed instructions through augmenting and summarization. It then matches these instructions with images and uses an open-sourced large language model (LLM) to generate coherent answers to the instruction-image pairs. The LLM is grounded by the detailed text descriptions of images in the whole answer generation process to guarantee the alignment of the instruction data. Moreover, we introduce a benchmark based on the generated instruction data to evaluate the instruction-following capabilities of existing LMMs. We demonstrate the effectiveness of MM-Instruct by training a LLaVA-1.5 model on the generated data, denoted as LLaVA-Instruct, which exhibits significant improvements in instruction-following capabilities compared to LLaVA-1.5 models. The MM-Instruct dataset, benchmark, and pre-trained models are available at https://github.com/jihaonew/MM-Instruct.
Authors: Chenlong Deng, Kelong Mao, Zhicheng Dou
Abstract: Legal case retrieval for sourcing similar cases is critical in upholding judicial fairness. Different from general web search, legal case retrieval involves processing lengthy, complex, and highly specialized legal documents. Existing methods in this domain often overlook the incorporation of legal expert knowledge, which is crucial for accurately understanding and modeling legal cases, leading to unsatisfactory retrieval performance. This paper introduces KELLER, a legal knowledge-guided case reformulation approach based on large language models (LLMs) for effective and interpretable legal case retrieval. By incorporating professional legal knowledge about crimes and law articles, we enable large language models to accurately reformulate the original legal case into concise sub-facts of crimes, which contain the essential information of the case. Extensive experiments on two legal case retrieval benchmarks demonstrate superior retrieval performance and robustness on complex legal case queries of KELLER over existing methods.
Authors: Junkai Chen, Zhenhao Li, Xing Hu, Xin Xia
Abstract: Large language models (LLMs) achieve promising results in code generation based on a given natural language description. They have been integrated into open-source projects and commercial products to facilitate daily coding activities. The natural language description in the prompt is crucial for LLMs to comprehend users' requirements. Prior studies uncover that LLMs are sensitive to the changes in the prompts, including slight changes that look inconspicuous. However, the natural language descriptions often vary in real-world scenarios (e.g., different formats, grammar, and wording). Prior studies on the robustness of LLMs are often based on random perturbations and such perturbations may not actually happen. In this paper, we conduct a comprehensive study to investigate how are code LLMs robust to variations of natural language description in real-world scenarios. We summarize 18 categories of perturbations of natural language and 3 combinations of co-occurred categories based on our literature review and an online survey with practitioners. We propose an automated framework, NLPerturbator, which can perform perturbations of each category given a set of prompts. Through a series of experiments on code generation using six code LLMs, we find that the perturbed prompts can decrease the performance of code generation by a considerable margin (e.g., up to 21.2%, and 4.8% to 6.1% on average). Our study highlights the importance of enhancing the robustness of LLMs to real-world variations in the prompts, as well as the essentiality of attentively constructing the prompts.
Authors: Danny Halawi, Alexander Wei, Eric Wallace, Tony T. Wang, Nika Haghtalab, Jacob Steinhardt
Abstract: Black-box finetuning is an emerging interface for adapting state-of-the-art language models to user needs. However, such access may also let malicious actors undermine model safety. To demonstrate the challenge of defending finetuning interfaces, we introduce covert malicious finetuning, a method to compromise model safety via finetuning while evading detection. Our method constructs a malicious dataset where every individual datapoint appears innocuous, but finetuning on the dataset teaches the model to respond to encoded harmful requests with encoded harmful responses. Applied to GPT-4, our method produces a finetuned model that acts on harmful instructions 99% of the time and avoids detection by defense mechanisms such as dataset inspection, safety evaluations, and input/output classifiers. Our findings question whether black-box finetuning access can be secured against sophisticated adversaries.
Authors: Tianyi Qiu, Yang Zhang, Xuchuan Huang, Jasmine Xinze Li, Jiaming Ji, Yaodong Yang
Abstract: Frontier AI systems, including large language models (LLMs), hold increasing influence over the epistemology of human users. Such influence can reinforce prevailing societal values, potentially contributing to the lock-in of misguided moral beliefs and, consequently, the perpetuation of problematic moral practices on a broad scale. We introduce progress alignment as a technical solution to mitigate this imminent risk. Progress alignment algorithms learn to emulate the mechanics of human moral progress, thereby addressing the susceptibility of existing alignment methods to contemporary moral blindspots. To empower research in progress alignment, we introduce ProgressGym, an experimental framework allowing the learning of moral progress mechanics from history, in order to facilitate future progress in real-world moral decisions. Leveraging 9 centuries of historical text and 18 historical LLMs, ProgressGym enables codification of real-world progress alignment challenges into concrete benchmarks. Specifically, we introduce three core challenges: tracking evolving values (PG-Follow), preemptively anticipating moral progress (PG-Predict), and regulating the feedback loop between human and AI value shifts (PG-Coevolve). Alignment methods without a temporal dimension are inapplicable to these tasks. In response, we present lifelong and extrapolative algorithms as baseline methods of progress alignment, and build an open leaderboard soliciting novel algorithms and challenges. The framework and the leaderboard are available at https://github.com/PKU-Alignment/ProgressGym and https://huggingface.co/spaces/PKU-Alignment/ProgressGym-LeaderBoard respectively.
URLs: https://github.com/PKU-Alignment/ProgressGym, https://huggingface.co/spaces/PKU-Alignment/ProgressGym-LeaderBoard
Authors: Xiang Li, Cristina Mata, Jongwoo Park, Kumara Kahatapitiya, Yoo Sung Jang, Jinghuan Shang, Kanchana Ranasinghe, Ryan Burgert, Mu Cai, Yong Jae Lee, Michael S. Ryoo
Abstract: Large Language Models (LLMs) equipped with extensive world knowledge and strong reasoning skills can tackle diverse tasks across domains, often by posing them as conversation-style instruction-response pairs. In this paper, we propose LLaRA: Large Language and Robotics Assistant, a framework which formulates robot action policy as conversations, and provides improved responses when trained with auxiliary data that complements policy learning. LLMs with visual inputs, i.e., Vision Language Models (VLMs), have the capacity to process state information as visual-textual prompts and generate optimal policy decisions in text. To train such action policy VLMs, we first introduce an automated pipeline to generate diverse high-quality robotics instruction data from existing behavior cloning data. A VLM finetuned with the resulting collection of datasets based on a conversation-style formulation tailored for robotics tasks, can generate meaningful robot action policy decisions. Our experiments across multiple simulated and real-world environments demonstrate the state-of-the-art performance of the proposed LLaRA framework. The code, datasets, and pretrained models are available at https://github.com/LostXine/LLaRA.
Authors: Sukmin Yun, Haokun Lin, Rusiru Thushara, Mohammad Qazim Bhat, Yongxin Wang, Zutao Jiang, Mingkai Deng, Jinhong Wang, Tianhua Tao, Junbo Li, Haonan Li, Preslav Nakov, Timothy Baldwin, Zhengzhong Liu, Eric P. Xing, Xiaodan Liang, Zhiqiang Shen
Abstract: Multimodal large language models (MLLMs) have shown impressive success across modalities such as image, video, and audio in a variety of understanding and generation tasks. However, current MLLMs are surprisingly poor at understanding webpage screenshots and generating their corresponding HTML code. To address this problem, we propose Web2Code, a benchmark consisting of a new large-scale webpage-to-code dataset for instruction tuning and an evaluation framework for the webpage understanding and HTML code translation abilities of MLLMs. For dataset construction, we leverage pretrained LLMs to enhance existing webpage-to-code datasets as well as generate a diverse pool of new webpages rendered into images. Specifically, the inputs are webpage images and instructions, while the responses are the webpage's HTML code. We further include diverse natural language QA pairs about the webpage content in the responses to enable a more comprehensive understanding of the web content. To evaluate model performance in these tasks, we develop an evaluation framework for testing MLLMs' abilities in webpage understanding and web-to-code generation. Extensive experiments show that our proposed dataset is beneficial not only to our proposed tasks but also in the general visual domain, while previous datasets result in worse performance. We hope our work will contribute to the development of general MLLMs suitable for web-based content generation and task automation. Our data and code will be available at https://github.com/MBZUAI-LLM/web2code.
Authors: Junyu Mao, Stuart E. Middleton, Mahesan Niranjan
Abstract: Prompt-based models have gathered a lot of attention from researchers due to their remarkable advancements in the fields of zero-shot and few-shot learning. Developing an effective prompt template plays a critical role. However, prior studies have mainly focused on prompt vocabulary searching or embedding initialization within a predefined template with the prompt position fixed. In this empirical study, we conduct the most comprehensive analysis to date of prompt position for diverse Natural Language Processing (NLP) tasks. Our findings quantify the substantial impact prompt position has on model performance. We observe that the prompt positions used in prior studies are often sub-optimal, and this observation is consistent even in widely used instruction-tuned models. These findings suggest prompt position optimisation as a valuable research direction to augment prompt engineering methodologies and prompt position-aware instruction tuning as a potential way to build more robust models in the future.
Authors: Yiran Wu, Feiran Jia, Shaokun Zhang, Hangyu Li, Erkang Zhu, Yue Wang, Yin Tat Lee, Richard Peng, Qingyun Wu, Chi Wang
Abstract: Employing Large Language Models (LLMs) to address mathematical problems is an intriguing research endeavor, considering the abundance of math problems expressed in natural language across numerous science and engineering fields. LLMs, with their generalized ability, are used as a foundation model to build AI agents for different tasks. In this paper, we study the effectiveness of utilizing LLM agents to solve math problems through conversations. We propose MathChat, a conversational problem-solving framework designed for math problems. MathChat consists of an LLM agent and a user proxy agent which is responsible for tool execution and additional guidance. This synergy facilitates a collaborative problem-solving process, where the agents engage in a dialogue to solve the problems. We perform evaluation on difficult high school competition problems from the MATH dataset. Utilizing Python, we show that MathChat can further improve previous tool-using prompting methods by 6%.
Authors: Xiaoxuan Wang, Ziniu Hu, Pan Lu, Yanqiao Zhu, Jieyu Zhang, Satyen Subramaniam, Arjun R. Loomba, Shichang Zhang, Yizhou Sun, Wei Wang
Abstract: Most of the existing Large Language Model (LLM) benchmarks on scientific problem reasoning focus on problems grounded in high-school subjects and are confined to elementary algebraic operations. To systematically examine the reasoning capabilities required for solving complex scientific problems, we introduce an expansive benchmark suite SciBench for LLMs. SciBench contains a carefully curated dataset featuring a range of collegiate-level scientific problems from mathematics, chemistry, and physics domains. Based on the dataset, we conduct an in-depth benchmarking study of representative open-source and proprietary LLMs with various prompting strategies. The results reveal that the current LLMs fall short of delivering satisfactory performance, with the best overall score of merely 43.22%. Furthermore, through a detailed user study, we categorize the errors made by LLMs into ten problem-solving abilities. Our analysis indicates that no single prompting strategy significantly outperforms the others and some strategies that demonstrate improvements in certain problem-solving skills could result in declines in other skills. We envision that SciBench will catalyze further developments in the reasoning abilities of LLMs, thereby ultimately contributing to scientific research and discovery.
Authors: Tongtao Ling, Yutao Lai, Lei Chen, Shilei Huang, Yi Liu
Abstract: In clinical dictation, utterances after automatic speech recognition (ASR) without explicit punctuation marks may lead to the misunderstanding of dictated reports. To give a precise and understandable clinical report with ASR, automatic punctuation restoration is required. Considering a practical scenario, we propose a fast and light pre-trained model for Chinese medical punctuation restoration based on 'pretraining and fine-tuning' paradigm. In this work, we distill pre-trained models by incorporating supervised contrastive learning and a novel auxiliary pre-training task (Punctuation Mark Prediction) to make it well-suited for punctuation restoration. Our experiments on various distilled models reveal that our model can achieve 95% performance while 10% model size relative to state-of-the-art Chinese RoBERTa.
Authors: Yucheng Shi, Shaochen Xu, Tianze Yang, Zhengliang Liu, Tianming Liu, Xiang Li, Ninghao Liu
Abstract: Large Language Models (LLMs), although powerful in general domains, often perform poorly on domain-specific tasks like medical question answering (QA). Moreover, they tend to function as "black-boxes," making it challenging to modify their behavior. To address the problem, our study delves into retrieval augmented generation (RAG), aiming to improve LLM responses without the need for fine-tuning or retraining. Specifically, we propose a comprehensive retrieval strategy to extract medical facts from an external knowledge base, and then inject them into the query prompt for LLMs. Focusing on medical QA using the MedQA-SMILE dataset, we evaluate the impact of different retrieval models and the number of facts provided to the LLM. Notably, our retrieval-augmented Vicuna-7B model exhibited an accuracy improvement from 44.46% to 48.54%. This work underscores the potential of RAG to enhance LLM performance, offering a practical approach to mitigate the challenges of black-box LLMs.
Authors: Pranjal Aggarwal, Aman Madaan, Ankit Anand, Srividya Pranavi Potharaju, Swaroop Mishra, Pei Zhou, Aditya Gupta, Dheeraj Rajagopal, Karthik Kappaganthu, Yiming Yang, Shyam Upadhyay, Manaal Faruqui, Mausam
Abstract: Large language models (LLMs) are now available from cloud API providers in various sizes and configurations. While this diversity offers a broad spectrum of choices, effectively leveraging the options to optimize computational cost and performance remains challenging. In this work, we present Automix, an approach that strategically routes queries to larger LMs, based on the approximate correctness of outputs from a smaller LM. Central to Automix are two key technical contributions. First, it has a few-shot self-verification mechanism, which estimates the reliability of its own outputs without requiring extensive training. Second, given that self-verification can be noisy, it employs a POMDP based router that can effectively select an appropriately sized model, based on answer confidence. Experiments across five language models and five challenging datasets show that Automix consistently surpasses strong baselines, reducing computational cost by over 50% for comparable performance.
Authors: Junda Wang, Zonghai Yao, Zhichao Yang, Huixue Zhou, Rumeng Li, Xun Wang, Yucheng Xu, Hong Yu
Abstract: We introduce NoteChat, a novel cooperative multi-agent framework leveraging Large Language Models (LLMs) to generate patient-physician dialogues. NoteChat embodies the principle that an ensemble of role-specific LLMs, through structured role-play and strategic prompting, can perform their assigned roles more effectively. The synergy among these role-playing LLMs results in a cohesive and efficient dialogue generation. Evaluation on MTS-dialogue, a benchmark dataset for patient-physician dialogues-note pairs, shows that models trained with the augmented synthetic patient-physician dialogues by NoteChat outperforms other state-of-the-art models for generating clinical notes. Our comprehensive automatic and human evaluation demonstrates that NoteChat substantially surpasses state-of-the-art models like ChatGPT and GPT-4 up to 22.78% by domain experts in generating superior synthetic patient-physician dialogues based on clinical notes. NoteChat has the potential to engage patients directly and help clinical documentation, a leading cause of physician burnout.
Authors: Mohan Raj Chanthran, Lay-Ki Soon, Huey Fang Ong, Bhawani Selvaretnam
Abstract: Recently, ChatGPT has attracted a lot of interest from both researchers and the general public. While the performance of ChatGPT in named entity recognition and relation extraction from Standard English texts is satisfactory, it remains to be seen if it can perform similarly for Malaysian English. Malaysian English is unique as it exhibits morphosyntactic and semantical adaptation from local contexts. In this study, we assess ChatGPT's capability in extracting entities and relations from the Malaysian English News (MEN) dataset. We propose a three-step methodology referred to as \textbf{\textit{educate-predict-evaluate}}. The performance of ChatGPT is assessed using F1-Score across 18 unique prompt settings, which were carefully engineered for a comprehensive review. From our evaluation, we found that ChatGPT does not perform well in extracting entities from Malaysian English news articles, with the highest F1-Score of 0.497. Further analysis shows that the morphosyntactic adaptation in Malaysian English caused the limitation. However, interestingly, this morphosyntactic adaptation does not impact the performance of ChatGPT for relation extraction.
Authors: Zheng Chu, Jingchang Chen, Qianglong Chen, Weijiang Yu, Haotian Wang, Ming Liu, Bing Qin
Abstract: Grasping the concept of time is a fundamental facet of human cognition, indispensable for truly comprehending the intricacies of the world. Previous studies typically focus on specific aspects of time, lacking a comprehensive temporal reasoning benchmark. To address this, we propose TimeBench, a comprehensive hierarchical temporal reasoning benchmark that covers a broad spectrum of temporal reasoning phenomena. TimeBench provides a thorough evaluation for investigating the temporal reasoning capabilities of large language models. We conduct extensive experiments on GPT-4, LLaMA2, and other popular LLMs under various settings. Our experimental results indicate a significant performance gap between the state-of-the-art LLMs and humans, highlighting that there is still a considerable distance to cover in temporal reasoning. Besides, LLMs exhibit capability discrepancies across different reasoning categories. Furthermore, we thoroughly analyze the impact of multiple aspects on temporal reasoning and emphasize the associated challenges. We aspire for TimeBench to serve as a comprehensive benchmark, fostering research in temporal reasoning. Resources are available at: https://github.com/zchuz/TimeBench
Authors: Armand Stricker, Patrick Paroubek
Abstract: In current text-based task-oriented dialogue (TOD) systems, user emotion detection (ED) is often overlooked or is typically treated as a separate and independent task, requiring additional training. In contrast, our work demonstrates that seamlessly unifying ED and TOD modeling brings about mutual benefits, and is therefore an alternative to be considered. Our method consists in augmenting SimpleToD, an end-to-end TOD system, by extending belief state tracking to include ED, relying on a single language model. We evaluate our approach using GPT-2 and Llama-2 on the EmoWOZ benchmark, a version of MultiWOZ annotated with emotions. Our results reveal a general increase in performance for ED and task results. Our findings also indicate that user emotions provide useful contextual conditioning for system responses, and can be leveraged to further refine responses in terms of empathy.
Authors: Jianlv Chen, Shitao Xiao, Peitian Zhang, Kun Luo, Defu Lian, Zheng Liu
Abstract: In this paper, we present a new embedding model, called M3-Embedding, which is distinguished for its versatility in Multi-Linguality, Multi-Functionality, and Multi-Granularity. It can support more than 100 working languages, leading to new state-of-the-art performances on multi-lingual and cross-lingual retrieval tasks. It can simultaneously perform the three common retrieval functionalities of embedding model: dense retrieval, multi-vector retrieval, and sparse retrieval, which provides a unified model foundation for real-world IR applications. It is able to process inputs of different granularities, spanning from short sentences to long documents of up to 8192 tokens. The effective training of M3-Embedding involves the following technical contributions. We propose a novel self-knowledge distillation approach, where the relevance scores from different retrieval functionalities can be integrated as the teacher signal to enhance the training quality. We also optimize the batching strategy, enabling a large batch size and high training throughput to ensure the discriminativeness of embeddings. To the best of our knowledge, M3-Embedding is the first embedding model which realizes such a strong versatility. The model and code will be publicly available at https://github.com/FlagOpen/FlagEmbedding.
Authors: Zhihao Fan, Jialong Tang, Wei Chen, Siyuan Wang, Zhongyu Wei, Jun Xi, Fei Huang, Jingren Zhou
Abstract: Artificial intelligence has significantly advanced healthcare, particularly through large language models (LLMs) that excel in medical question answering benchmarks. However, their real-world clinical application remains limited due to the complexities of doctor-patient interactions. To address this, we introduce \textbf{AI Hospital}, a multi-agent framework simulating dynamic medical interactions between \emph{Doctor} as player and NPCs including \emph{Patient}, \emph{Examiner}, \emph{Chief Physician}. This setup allows for realistic assessments of LLMs in clinical scenarios. We develop the Multi-View Medical Evaluation (MVME) benchmark, utilizing high-quality Chinese medical records and NPCs to evaluate LLMs' performance in symptom collection, examination recommendations, and diagnoses. Additionally, a dispute resolution collaborative mechanism is proposed to enhance diagnostic accuracy through iterative discussions. Despite improvements, current LLMs exhibit significant performance gaps in multi-turn interactions compared to one-step approaches. Our findings highlight the need for further research to bridge these gaps and improve LLMs' clinical diagnostic capabilities. Our data, code, and experimental results are all open-sourced at \url{https://github.com/LibertFan/AI_Hospital}.
Authors: Guiming Hardy Chen, Shunian Chen, Ziche Liu, Feng Jiang, Benyou Wang
Abstract: Adopting human and large language models (LLM) as judges (a.k.a human- and LLM-as-a-judge) for evaluating the performance of LLMs has recently gained attention. Nonetheless, this approach concurrently introduces potential biases from human and LLMs, questioning the reliability of the evaluation results. In this paper, we propose a novel framework that is free from referencing groundtruth annotations for investigating Misinformation Oversight Bias, Gender Bias, Authority Bias and Beauty Bias on LLM and human judges. We curate a dataset referring to the revised Bloom's Taxonomy and conduct thousands of evaluations. Results show that human and LLM judges are vulnerable to perturbations to various degrees, and that even the cutting-edge judges possess considerable biases. We further exploit these biases to conduct attacks on LLM judges. We hope that our work can notify the community of the bias and vulnerability of human- and LLM-as-a-judge, as well as the urgency of developing robust evaluation systems.
Authors: Xinyu Hu, Mingqi Gao, Sen Hu, Yang Zhang, Yicheng Chen, Teng Xu, Xiaojun Wan
Abstract: Some prior work has shown that LLMs perform well in NLG evaluation for different tasks. However, we discover that LLMs seem to confuse different evaluation criteria, which reduces their reliability. For further verification, we first consider avoiding issues of inconsistent conceptualization and vague expression in existing NLG quality criteria themselves. So we summarize a clear hierarchical classification system for 11 common aspects with corresponding different criteria from previous studies involved. Inspired by behavioral testing, we elaborately design 18 types of aspect-targeted perturbation attacks for fine-grained analysis of the evaluation behaviors of different LLMs. We also conduct human annotations beyond the guidance of the classification system to validate the impact of the perturbations. Our experimental results reveal confusion issues inherent in LLMs, as well as other noteworthy phenomena, and necessitate further research and improvements for LLM-based evaluation.
Authors: Mohammad Javad Hosseini, Andrey Petrov, Alex Fabrikant, Annie Louis
Abstract: Natural Language Inference (NLI) remains an important benchmark task for LLMs. NLI datasets are a springboard for transfer learning to other semantic tasks, and NLI models are standard tools for identifying the faithfulness of model-generated text. There are several large scale NLI datasets today, and models have improved greatly by hill-climbing on these collections. Yet their realistic performance on out-of-distribution/domain data is less well-understood. We explore the opportunity for synthetic high-quality datasets to adapt NLI models for zero-shot use in downstream applications across new and unseen text domains. We demonstrate a new approach for generating NLI data in diverse domains and lengths, so far not covered by existing training sets. The resulting examples have meaningful premises, the hypotheses are formed in creative ways rather than simple edits to a few premise tokens, and the labels have high accuracy. We show that models trained on this data ($685$K synthetic examples) have the best generalization to completely new downstream test settings. On the TRUE benchmark, a T5-small model trained with our data improves around $7\%$ on average compared to training on the best alternative dataset. The improvements are more pronounced for smaller models, while still meaningful on a T5 XXL model. We also demonstrate gains on test sets when in-domain training data is augmented with our domain-general synthetic data.
Authors: Armand Stricker, Patrick Paroubek
Abstract: During task-oriented dialogues (TODs), human users naturally introduce chitchat that is beyond the immediate scope of the task, interfering with the flow of the conversation. To address this issue without the need for expensive manual data creation, we use few-shot prompting with Llama-2-70B to enhance the MultiWOZ dataset with user backstories, a typical example of chitchat interference in TODs. We assess the impact of this addition by testing two models: one trained solely on TODs and another trained on TODs with a preliminary chitchat interaction. Our analysis demonstrates that our enhanced dataset poses a challenge for these systems. Moreover, we demonstrate that our dataset can be effectively used for training purposes, enabling a system to consistently acknowledge the user's backstory while also successfully moving the task forward in the same turn, as confirmed by human evaluation. These findings highlight the benefits of generating novel chitchat-TOD scenarios to test TOD systems more thoroughly and improve their resilience to natural user interferences
Authors: Junda Wang, Zhichao Yang, Zonghai Yao, Hong Yu
Abstract: Large Language Models (LLMs) have demonstrated a remarkable potential in medical knowledge acquisition and question-answering. However, LLMs can potentially hallucinate and yield factually incorrect outcomes, even with domain-specific pretraining. Previously, retrieval augmented generation (RAG) has limited success in addressing hallucinations. Unlike previous methods in RAG where the retrieval model was trained separately from the LLM, we introduce JMLR (for Jointly trains LLM and information Retrieval) during the fine-tuning phase. The synchronized training mechanism enhances JMLR's ability to retrieve clinical guidelines and leverage medical knowledge to reason and answer questions and reduces the demand for computational resources. We evaluated JMLR on the important medical question-answering application. Our experimental results demonstrate that JMLR-13B (70.5%) outperforms a previous state-of-the-art open-source model using conventional pre-training and fine-tuning Meditron-70B (68.9%) and Llama2-13B with RAG (67.7%) on a medical question-answering dataset. Comprehensive evaluations reveal JMLR-13B enhances reasoning quality and reduces hallucinations better than Claude3-Opus. Additionally, JMLR-13B (148 GPU hours) also trains much faster than Meditron-70B (42630 GPU hours). Through this work, we provide a new and efficient knowledge enhancement method for healthcare, demonstrating the potential of integrating retrieval and LLM training for medical question-answering systems.
Authors: Guangming Huang, Yunfei Long, Cunjin Luo, Jiaxing Shen, Xia Sun
Abstract: Pre-trained language models (PLMs) leverage chains-of-thought (CoT) to simulate human reasoning and inference processes, achieving proficient performance in multi-hop QA. However, a gap persists between PLMs' reasoning abilities and those of humans when tackling complex problems. Psychological studies suggest a vital connection between explicit information in passages and human prior knowledge during reading. Nevertheless, current research has given insufficient attention to linking input passages and PLMs' pre-training-based knowledge from the perspective of human cognition studies. In this study, we introduce a Prompting Explicit and Implicit knowledge (PEI) framework, which uses prompts to connect explicit and implicit knowledge, aligning with human reading process for multi-hop QA. We consider the input passages as explicit knowledge, employing them to elicit implicit knowledge through unified prompt reasoning. Furthermore, our model incorporates type-specific reasoning via prompts, a form of implicit knowledge. Experimental results show that PEI performs comparably to the state-of-the-art on HotpotQA. Ablation studies confirm the efficacy of our model in bridging and integrating explicit and implicit knowledge.
Authors: Bosheng Ding, Chengwei Qin, Ruochen Zhao, Tianze Luo, Xinze Li, Guizhen Chen, Wenhan Xia, Junjie Hu, Anh Tuan Luu, Shafiq Joty
Abstract: In the rapidly evolving field of large language models (LLMs), data augmentation (DA) has emerged as a pivotal technique for enhancing model performance by diversifying training examples without the need for additional data collection. This survey explores the transformative impact of LLMs on DA, particularly addressing the unique challenges and opportunities they present in the context of natural language processing (NLP) and beyond. From both data and learning perspectives, we examine various strategies that utilize LLMs for data augmentation, including a novel exploration of learning paradigms where LLM-generated data is used for diverse forms of further training. Additionally, this paper highlights the primary open challenges faced in this domain, ranging from controllable data augmentation to multi-modal data augmentation. This survey highlights a paradigm shift introduced by LLMs in DA, and aims to serve as a comprehensive guide for researchers and practitioners.
Authors: Xidong Wang, Nuo Chen, Junyin Chen, Yan Hu, Yidong Wang, Xiangbo Wu, Anningzhe Gao, Xiang Wan, Haizhou Li, Benyou Wang
Abstract: Despite the vast repository of global medical knowledge predominantly being in English, local languages are crucial for delivering tailored healthcare services, particularly in areas with limited medical resources. To extend the reach of medical AI advancements to a broader population, we aim to develop medical LLMs across the six most widely spoken languages, encompassing a global population of 6.1 billion. This effort culminates in the creation of the ApolloCorpora multilingual medical dataset and the XMedBench benchmark. In the multilingual medical benchmark, the released Apollo models, at various relatively-small sizes (i.e., 0.5B, 1.8B, 2B, 6B, and 7B), achieve the best performance among models of equivalent size. Especially, Apollo-7B is the state-of-the-art multilingual medical LLMs up to 70B. Additionally, these lite models could be used to improve the multi-lingual medical capabilities of larger models without fine-tuning in a proxy-tuning fashion. We will open-source training corpora, code, model weights and evaluation benchmark.
Authors: Michal \v{S}tef\'anik, Marek Kadl\v{c}\'ik, Petr Sojka
Abstract: Many recent language models (LMs) are capable of in-context learning (ICL), manifested in the LMs' ability to perform a new task solely from natural-language instruction. Previous work curating in-context learners assumes that ICL emerges from a vast over-parametrization or the scale of multi-task training. However, recent theoretical work attributes the ICL ability to concept-dependent training data and creates functional in-context learners even in small-scale, synthetic settings. In this work, we practically explore this newly identified axis of ICL quality. We propose Concept-aware Training (CoAT), a framework for constructing training scenarios that make it beneficial for the LM to learn to utilize the analogical reasoning concepts from demonstrations. We find that by using CoAT, pre-trained transformers can learn to better utilise new latent concepts from demonstrations and that such ability makes ICL more robust to the functional deficiencies of the previous models. Finally, we show that concept-aware in-context learning is more effective for a majority of new tasks when compared to traditional instruction tuning, resulting in a performance comparable to the previous in-context learners using magnitudes of more training data.
Authors: Yaowei Zheng, Richong Zhang, Junhao Zhang, Yanhan Ye, Zheyan Luo, Zhangchi Feng, Yongqiang Ma
Abstract: Efficient fine-tuning is vital for adapting large language models (LLMs) to downstream tasks. However, it requires non-trivial efforts to implement these methods on different models. We present LlamaFactory, a unified framework that integrates a suite of cutting-edge efficient training methods. It provides a solution for flexibly customizing the fine-tuning of 100+ LLMs without the need for coding through the built-in web UI LlamaBoard. We empirically validate the efficiency and effectiveness of our framework on language modeling and text generation tasks. It has been released at https://github.com/hiyouga/LLaMA-Factory and received over 25,000 stars and 3,000 forks.
Authors: Nazanin Jafari, James Allan, Sheikh Muhammad Sarwar
Abstract: Identifying the targets of hate speech is a crucial step in grasping the nature of such speech and, ultimately, in improving the detection of offensive posts on online forums. Much harmful content on online platforms uses implicit language especially when targeting vulnerable and protected groups such as using stereotypical characteristics instead of explicit target names, making it harder to detect and mitigate the language. In this study, we focus on identifying implied targets of hate speech, essential for recognizing subtler hate speech and enhancing the detection of harmful content on digital platforms. We define a new task aimed at identifying the targets even when they are not explicitly stated. To address that task, we collect and annotate target spans in three prominent implicit hate speech datasets: SBIC, DynaHate, and IHC. We call the resulting merged collection Implicit-Target-Span. The collection is achieved using an innovative pooling method with matching scores based on human annotations and Large Language Models (LLMs). Our experiments indicate that Implicit-Target-Span provides a challenging test bed for target span detection methods.
Authors: Tobias Schnabel, Jennifer Neville
Abstract: In many modern LLM applications, such as retrieval augmented generation, prompts have become programs themselves. In these settings, prompt programs are repeatedly called with different user queries or data instances. A big practical challenge is optimizing such prompt programs. Recent work has mostly focused on either simple prompt programs or assumed that the general structure of a prompt program is fixed. We introduce SAMMO, a framework to perform symbolic prompt program search for compile-time optimizations of prompt programs. SAMMO represents prompt programs on a symbolic level which allows for a rich set of transformations that can be searched over during optimization. We show that SAMMO generalizes previous methods and improves the performance of complex prompts on (1) instruction tuning, (2) RAG pipeline tuning, and (3) prompt compression, across several different LLMs. We make all code available open-source at https://github.com/microsoft/sammo .
Authors: Christopher Clarke, Roland Daynauth, Charlene Wilkinson, Hubert Devonish, Jason Mars
Abstract: While major languages often enjoy substantial attention and resources, the linguistic diversity across the globe encompasses a multitude of smaller, indigenous, and regional languages that lack the same level of computational support. One such region is the Caribbean. While commonly labeled as "English speaking", the ex-British Caribbean region consists of a myriad of Creole languages thriving alongside English. In this paper, we present Guylingo: a comprehensive corpus designed for advancing NLP research in the domain of Creolese (Guyanese English-lexicon Creole), the most widely spoken language in the culturally rich nation of Guyana. We first outline our framework for gathering and digitizing this diverse corpus, inclusive of colloquial expressions, idioms, and regional variations in a low-resource language. We then demonstrate the challenges of training and evaluating NLP models for machine translation in Creole. Lastly, we discuss the unique opportunities presented by recent NLP advancements for accelerating the formal adoption of Creole languages as official languages in the Caribbean.
Authors: Shaina Raza, Ananya Raval, Veronica Chatrath
Abstract: The deployment of Large Language Models (LLMs) in diverse applications necessitates an assurance of safety without compromising the contextual integrity of the generated content. Traditional approaches, including safety-specific fine-tuning or adversarial testing, often yield safe outputs at the expense of contextual meaning. This can result in a diminished capacity to handle nuanced aspects of bias and toxicity, such as underrepresentation or negative portrayals across various demographics. To address these challenges, we introduce MBIAS, an LLM framework carefully instruction fine-tuned on a custom dataset designed specifically for safety interventions. MBIAS is designed to significantly reduce biases and toxic elements in LLM outputs while preserving the main information. This work also details our further use of LLMs: as annotator under human supervision and as evaluator of generated content. Empirical analysis reveals that MBIAS achieves a reduction in bias and toxicity by over 30\% in standard evaluations, and by more than 90\% in diverse demographic tests, highlighting the robustness of our approach. We make the dataset and the fine-tuned model available to the research community for further investigation and ensure reproducibility. The code for this project can be accessed here https://github.com/shainarazavi/MBIAS/tree/main. Warning: This paper contains examples that may be offensive or upsetting.
Authors: Felix B Mueller, Rebekka G\"orge, Anna K Bernzen, Janna C Pirk, Maximilian Poretschkin
Abstract: Memorization in large language models (LLMs) is a growing concern. LLMs have been shown to easily reproduce parts of their training data, including copyrighted work. This is an important problem to solve, as it may violate existing copyright laws as well as the European AI Act. In this work, we propose a systematic analysis to quantify the extent of potential copyright infringements in LLMs using European law as an example. Unlike previous work, we evaluate instruction-finetuned models in a realistic end-user scenario. Our analysis builds on a proposed threshold of 160 characters, which we borrow from the German Copyright Service Provider Act and a fuzzy text matching algorithm to identify potentially copyright-infringing textual reproductions. The specificity of countermeasures against copyright infringement is analyzed by comparing model behavior on copyrighted and public domain data. We investigate what behaviors models show instead of producing protected text (such as refusal or hallucination) and provide a first legal assessment of these behaviors. We find that there are huge differences in copyright compliance, specificity, and appropriate refusal among popular LLMs. Alpaca, GPT 4, GPT 3.5, and Luminous perform best in our comparison, with OpenGPT-X, Alpaca, and Luminous producing a particularly low absolute number of potential copyright violations. Code will be published soon.
Authors: Yongkang Liu, Ercong Nie, Shi Feng, Zheng Hua, Zifeng Ding, Daling Wang, Yifei Zhang, Hinrich Sch\"utze
Abstract: Current state-of-the-art dialogue systems heavily rely on extensive training datasets. However, challenges arise in domains where domain-specific training datasets are insufficient or entirely absent. To tackle this challenge, we propose a novel data \textbf{A}ugmentation framework for \textbf{M}ulti-\textbf{D}omain \textbf{D}ialogue \textbf{G}eneration, referred to as \textbf{AMD$^2$G}. The AMD$^2$G framework consists of a data augmentation process and a two-stage training approach: domain-agnostic training and domain adaptation training. We posit that domain corpora are a blend of domain-agnostic and domain-specific features, with certain representation patterns shared among diverse domains. Domain-agnostic training aims to enable models to learn these common expressive patterns. To construct domain-agnostic dialogue corpora, we employ a \textit{\textbf{de-domaining}} data processing technique used to remove domain-specific features. By mitigating the effects of domain-specific features, the model trained on the de-domained corpora can effectively learn common expression patterns in different domains. Subsequently, we adapt the learned domain-agnostic features to the target domain through domain adaptation training. We conduct experiments on Chinese dialogue datasets from five different domains and show that AMD$^2$G achieves superior performance compared to both direct training on the target domain corpus and collective training on all five domain corpora. Our work underscores AMD$^2$G as a viable alternative solution for low-resource multi-domain dialogue generation. Code and data associated with our work are available on GitHub repository$^{\text 1}$.
Authors: Adane Nega Tarekegn
Abstract: The news landscape is continuously evolving, with an ever-increasing volume of information from around the world. Automated event detection within this vast data repository is essential for monitoring, identifying, and categorizing significant news occurrences across diverse platforms. This paper presents an event detection framework that leverages Large Language Models (LLMs) combined with clustering analysis to detect news events from the Global Database of Events, Language, and Tone (GDELT). The framework enhances event clustering through both pre-event detection tasks (keyword extraction and text embedding) and post-event detection tasks (event summarization and topic labelling). We also evaluate the impact of various textual embeddings on the quality of clustering outcomes, ensuring robust news categorization. Additionally, we introduce a novel Cluster Stability Assessment Index (CSAI) to assess the validity and robustness of clustering results. CSAI utilizes multiple feature vectors to provide a new way of measuring clustering quality. Our experiments indicate that the use of LLM embedding in the event detection framework has significantly improved the results, demonstrating greater robustness in terms of CSAI scores. Moreover, post-event detection tasks generate meaningful insights, facilitating effective interpretation of event clustering results. Overall, our experimental results indicate that the proposed framework offers valuable insights and could enhance the accuracy in news analysis and reporting.
Authors: Lucy Horowitz, Ryan Hathaway
Abstract: In this paper, we fine-tuned three pre-trained BERT models on the task of "definition extraction" from mathematical English written in LaTeX. This is presented as a binary classification problem, where either a sentence contains a definition of a mathematical term or it does not. We used two original data sets, "Chicago" and "TAC," to fine-tune and test these models. We also tested on WFMALL, a dataset presented by Vanetik and Litvak in 2021 and compared the performance of our models to theirs. We found that a high-performance Sentence-BERT transformer model performed best based on overall accuracy, recall, and precision metrics, achieving comparable results to the earlier models with less computational effort.
Authors: Qianchao Zhu, Jiangfei Duan, Chang Chen, Siran Liu, Xiuhong Li, Guanyu Feng, Xin Lv, Huanqi Cao, Xiao Chuanfu, Xingcheng Zhang, Dahua Lin, Chao Yang
Abstract: Large language models (LLMs) now support extremely long context windows, but the quadratic complexity of vanilla attention results in significantly long Time-to-First-Token (TTFT) latency. Existing approaches to address this complexity require additional pretraining or finetuning, and often sacrifice model accuracy. In this paper, we first provide both theoretical and empirical foundations for near-lossless sparse attention. We find dynamically capturing head-specific sparse patterns at runtime with low overhead is crucial. To address this, we propose SampleAttention, an adaptive structured and near-lossless sparse attention. Leveraging observed significant sparse patterns, SampleAttention attends to a fixed percentage of adjacent tokens to capture local window patterns, and employs a two-stage query-guided key-value filtering approach, which adaptively select a minimum set of key-values with low overhead, to capture column stripe patterns. Comprehensive evaluations show that SampleAttention can seamlessly replace vanilla attention in off-the-shelf LLMs with nearly no accuracy loss, and reduces TTFT by up to $2.42\times$ compared with FlashAttention.
Authors: Rishabh Maheshwary, Vikas Yadav, Hoang Nguyen, Khyati Mahajan, Sathwik Tejaswi Madhusudhan
Abstract: Instruction finetuning (IFT) is critical for aligning Large Language Models (LLMs) to follow instructions. While many effective IFT datasets have been introduced recently, they predominantly focus on high-resource languages like English. To better align LLMs across a broad spectrum of languages and tasks, we propose a fully synthetic, novel taxonomy (Evol) guided Multilingual, Multi-turn instruction finetuning dataset, called M2Lingual. It is constructed by first selecting a diverse set of seed examples and then utilizing the proposed Evol taxonomy to convert these seeds into complex and challenging multi-turn instructions. We demonstrate the effectiveness of M2Lingual by training LLMs of varying sizes and showcasing the enhanced performance across a diverse set of languages. We contribute the 2 step Evol taxonomy with the guided generation code: https://github.com/ServiceNow/M2Lingual, as well as the first fully synthetic, general and task-oriented, multi-turn, multilingual dataset built with Evol - M2Lingual: https://huggingface.co/datasets/ServiceNow-AI/ M2Lingual - containing 182K total IFT pairs, covering 70 languages and 17+ NLP tasks.
URLs: https://github.com/ServiceNow/M2Lingual,, https://huggingface.co/datasets/ServiceNow-AI/
Authors: Mihai Masala, Denis C. Ilie-Ablachim, Alexandru Dima, Dragos Corlatescu, Miruna Zavelca, Ovio Olaru, Simina Terian, Andrei Terian, Marius Leordeanu, Horia Velicu, Marius Popescu, Mihai Dascalu, Traian Rebedea
Abstract: In recent years, Large Language Models (LLMs) have achieved almost human-like performance on various tasks. While some LLMs have been trained on multilingual data, most of the training data is in English; hence, their performance in English greatly exceeds other languages. To our knowledge, we are the first to collect and translate a large collection of texts, instructions, and benchmarks and train, evaluate, and release open-source LLMs tailored for Romanian. We evaluate our methods on four different categories, including academic benchmarks, MT-Bench (manually translated), and a professionally built historical, cultural, and social benchmark adapted to Romanian. We argue for the usefulness and high performance of RoLLMs by obtaining state-of-the-art results across the board. We publicly release all resources (i.e., data, training and evaluation code, models) to support and encourage research on Romanian LLMs while concurrently creating a generalizable recipe, adequate for other low or less-resourced languages.
Authors: Siyuan Wu, Yue Huang, Chujie Gao, Dongping Chen, Qihui Zhang, Yao Wan, Tianyi Zhou, Xiangliang Zhang, Jianfeng Gao, Chaowei Xiao, Lichao Sun
Abstract: Large Language Models (LLMs) such as GPT-4 and Llama3 have significantly impacted various fields by enabling high-quality synthetic data generation and reducing dependence on expensive human-generated datasets. Despite this, challenges remain in the areas of generalization, controllability, diversity, and truthfulness within the existing generative frameworks. To address these challenges, this paper presents UniGen, a comprehensive LLM-powered framework designed to produce diverse, accurate, and highly controllable datasets. UniGen is adaptable, supporting all types of text datasets and enhancing the generative process through innovative mechanisms. To augment data diversity, UniGen incorporates an attribute-guided generation module and a group checking feature. For accuracy, it employs a code-based mathematical assessment for label verification alongside a retrieval-augmented generation technique for factual validation. The framework also allows for user-specified constraints, enabling customization of the data generation process to suit particular requirements. Extensive experiments demonstrate the superior quality of data generated by UniGen, and each module within UniGen plays a critical role in this enhancement. Additionally, UniGen is applied in two practical scenarios: benchmarking LLMs and data augmentation. The results indicate that UniGen effectively supports dynamic and evolving benchmarking, and that data augmentation improves LLM capabilities in various domains, including agent-oriented abilities and reasoning skills.
Authors: Ekaterina Taktasheva, Maxim Bazhukov, Kirill Koncha, Alena Fenogenova, Ekaterina Artemova, Vladislav Mikhailov
Abstract: Minimal pairs are a well-established approach to evaluating the grammatical knowledge of language models. However, existing resources for minimal pairs address a limited number of languages and lack diversity of language-specific grammatical phenomena. This paper introduces the Russian Benchmark of Linguistic Minimal Pairs (RuBLiMP), which includes 45k pairs of sentences that differ in grammaticality and isolate a morphological, syntactic, or semantic phenomenon. In contrast to existing benchmarks of linguistic minimal pairs, RuBLiMP is created by applying linguistic perturbations to automatically annotated sentences from open text corpora and carefully curating test data. We describe the data collection protocol and present the results of evaluating 25 language models in various scenarios. We find that the widely used language models for Russian are sensitive to morphological and agreement-oriented contrasts but fall behind humans on phenomena requiring understanding of structural relations, negation, transitivity, and tense. RuBLiMP, the codebase, and other materials are publicly available.
Authors: Shubhankar Singh, Purvi Chaurasia, Yerram Varun, Pranshu Pandya, Vatsal Gupta, Vivek Gupta, Dan Roth
Abstract: Existing benchmarks for visual question answering lack in visual grounding and complexity, particularly in evaluating spatial reasoning skills. We introduce FlowVQA, a novel benchmark aimed at assessing the capabilities of visual question-answering multimodal language models in reasoning with flowcharts as visual contexts. FlowVQA comprises 2,272 carefully generated and human-verified flowchart images from three distinct content sources, along with 22,413 diverse question-answer pairs, to test a spectrum of reasoning tasks, including information localization, decision-making, and logical progression. We conduct a thorough baseline evaluation on a suite of both open-source and proprietary multimodal language models using various strategies, followed by an analysis of directional bias. The results underscore the benchmark's potential as a vital tool for advancing the field of multimodal modeling, providing a focused and challenging environment for enhancing model performance in visual and logical reasoning tasks.
Authors: William Muldrew, Peter Hayes, Mingtian Zhang, David Barber
Abstract: As large language models (LLMs) become more capable, fine-tuning techniques for aligning with human intent are increasingly important. A key consideration for aligning these models is how to most effectively use human resources, or model resources in the case where LLMs themselves are used as oracles. Reinforcement learning from Human or AI preferences (RLHF/RLAIF) is the most prominent example of such a technique, but is complex and often unstable. Direct Preference Optimization (DPO) has recently been proposed as a simpler and more stable alternative. In this work, we develop an active learning strategy for DPO to make better use of preference labels. We propose a practical acquisition function for prompt/completion pairs based on the predictive entropy of the language model and a measure of certainty of the implicit preference model optimized by DPO. We demonstrate how our approach improves both the rate of learning and final performance of fine-tuning on pairwise preference data.
Authors: Junfei Wu, Qiang Liu, Ding Wang, Jinghao Zhang, Shu Wu, Liang Wang, Tieniu Tan
Abstract: Object hallucination has been an Achilles' heel which hinders the broader applications of large vision-language models (LVLMs). Object hallucination refers to the phenomenon that the LVLMs claim non-existent objects in the image. To mitigate the object hallucinations, instruction tuning and external model-based detection methods have been proposed, which either require large-scare computational resources or depend on the detection result of external models. However, there remains an under-explored field to utilize the LVLM itself to alleviate object hallucinations. In this work, we adopt the intuition that the LVLM tends to respond logically consistently for existent objects but inconsistently for hallucinated objects. Therefore, we propose a Logical Closed Loop-based framework for Object Hallucination Detection and Mitigation, namely LogicCheckGPT. In specific, we devise logical consistency probing to raise questions with logical correlations, inquiring about attributes from objects and vice versa. Whether their responses can form a logical closed loop serves as an indicator of object hallucination. As a plug-and-play method, it can be seamlessly applied to all existing LVLMs. Comprehensive experiments conducted on three benchmarks across four LVLMs have demonstrated significant improvements brought by our method, indicating its effectiveness and generality.
Authors: Hanlei Zhang, Xin Wang, Hua Xu, Qianrui Zhou, Kai Gao, Jianhua Su, jinyue Zhao, Wenrui Li, Yanting Chen
Abstract: Multimodal intent recognition poses significant challenges, requiring the incorporation of non-verbal modalities from real-world contexts to enhance the comprehension of human intentions. Existing benchmark datasets are limited in scale and suffer from difficulties in handling out-of-scope samples that arise in multi-turn conversational interactions. We introduce MIntRec2.0, a large-scale benchmark dataset for multimodal intent recognition in multi-party conversations. It contains 1,245 dialogues with 15,040 samples, each annotated within a new intent taxonomy of 30 fine-grained classes. Besides 9,304 in-scope samples, it also includes 5,736 out-of-scope samples appearing in multi-turn contexts, which naturally occur in real-world scenarios. Furthermore, we provide comprehensive information on the speakers in each utterance, enriching its utility for multi-party conversational research. We establish a general framework supporting the organization of single-turn and multi-turn dialogue data, modality feature extraction, multimodal fusion, as well as in-scope classification and out-of-scope detection. Evaluation benchmarks are built using classic multimodal fusion methods, ChatGPT, and human evaluators. While existing methods incorporating nonverbal information yield improvements, effectively leveraging context information and detecting out-of-scope samples remains a substantial challenge. Notably, large language models exhibit a significant performance gap compared to humans, highlighting the limitations of machine learning methods in the cognitive intent understanding task. We believe that MIntRec2.0 will serve as a valuable resource, providing a pioneering foundation for research in human-machine conversational interactions, and significantly facilitating related applications. The full dataset and codes are available at https://github.com/thuiar/MIntRec2.0.
Authors: Rohan Padhye
Abstract: The recent proliferation of generative artificial intelligence (AI) technologies such as pre-trained large language models (LLMs) has opened up new frontiers in computational law. An exciting area of development is the use of AI to automate the deductive rule-based reasoning inherent in statutory and contract law. This paper argues that such automated deductive legal reasoning can now be viewed from the lens of software engineering, treating LLMs as interpreters of natural-language programs with natural-language inputs. We show how it is possible to apply principled software engineering techniques to enhance AI-driven legal reasoning of complex statutes and to unlock new applications in automated meta-reasoning such as mutation-guided example generation and metamorphic property-based testing.
Authors: Nadav Timor, Jonathan Mamou, Daniel Korat, Moshe Berchansky, Oren Pereg, Moshe Wasserblat, Tomer Galanti, Michal Gordon, David Harel
Abstract: Accelerating the inference of large language models (LLMs) is an important challenge in artificial intelligence. This paper introduces distributed speculative inference (DSI), a novel distributed inference algorithm that is provably faster than speculative inference (SI) [leviathan2023fast, chen2023accelerating, miao2023specinfer] and traditional autoregressive inference (non-SI). Like other SI algorithms, DSI works on frozen LLMs, requiring no training or architectural modifications, and it preserves the target distribution. Prior studies on SI have demonstrated empirical speedups (compared to non-SI) but require a fast and accurate drafter LLM. In practice, off-the-shelf LLMs often do not have matching drafters that are sufficiently fast and accurate. We show a gap: SI gets slower than non-SI when using slower or less accurate drafters. We close this gap by proving that DSI is faster than both SI and non-SI given any drafters. By orchestrating multiple instances of the target and drafters, DSI is not only faster than SI but also supports LLMs that cannot be accelerated with SI. Our simulations show speedups of off-the-shelf LLMs in realistic settings: DSI is 1.29-1.92x faster than SI.
Authors: Paiheng Xu, Louiqa Raschid, Vanessa Frias-Martinez
Abstract: Social media platforms like Twitter (now X) have been pivotal in information dissemination and public engagement, especially during COVID-19. A key goal for public health experts was to encourage prosocial behavior that could impact local outcomes such as masking and social distancing. Given the importance of local news and guidance during COVID-19, the objective of our research is to analyze the effect of localized engagement, on social media conversations. This study examines the impact of geographic co-location, as a proxy for localized engagement between public health experts (PHEs) and the public, on social media. We analyze a Twitter conversation dataset from January 2020 to November 2021, comprising over 19 K tweets from nearly five hundred PHEs, along with approximately 800 K replies from 350 K participants. Our findings reveal that geo-co-location is associated with higher engagement rates, especially in conversations on topics including masking, lockdowns, and education, and in conversations with academic and medical professionals. Lexical features associated with emotion and personal experiences were more common in geo-co-located contexts. This research provides insights into how geographic co-location influences social media engagement and can inform strategies to improve public health messaging.
Authors: Zitao Song, Chao Yang, Chaojie Wang, Bo An, Shuang Li
Abstract: Modern high-stakes systems, such as healthcare or robotics, often generate vast streaming event sequences. Our goal is to design an efficient, plug-and-play tool to elicit logic tree-based explanations from Large Language Models (LLMs) to provide customized insights into each observed event sequence. Built on the temporal point process model for events, our method employs the likelihood function as a score to evaluate generated logic trees. We propose an amortized Expectation-Maximization (EM) learning framework and treat the logic tree as latent variables. In the E-step, we evaluate the posterior distribution over the latent logic trees using an LLM prior and the likelihood of the observed event sequences. LLM provides a high-quality prior for the latent logic trees, however, since the posterior is built over a discrete combinatorial space, we cannot get the closed-form solution. We propose to generate logic tree samples from the posterior using a learnable GFlowNet, which is a diversity-seeking generator for structured discrete variables. The M-step employs the generated logic rules to approximate marginalization over the posterior, facilitating the learning of model parameters and refining the tunable LLM prior parameters. In the online setting, our locally built, lightweight model will iteratively extract the most relevant rules from LLMs for each sequence using only a few iterations. Empirical demonstrations showcase the promising performance and adaptability of our framework.
Authors: Seyedali Mohammadi, Edward Raff, Jinendra Malekar, Vedant Palit, Francis Ferraro, Manas Gaur
Abstract: Language Models (LMs) are being proposed for mental health applications where the heightened risk of adverse outcomes means predictive performance may not be a sufficient litmus test of a model's utility in clinical practice. A model that can be trusted for practice should have a correspondence between explanation and clinical determination, yet no prior research has examined the attention fidelity of these models and their effect on ground truth explanations. We introduce an evaluation design that focuses on the robustness and explainability of LMs in identifying Wellness Dimensions (WD). We focus on two mental health and well-being datasets: (a) Multi-label Classification-based MultiWD, and (b) WellXplain for evaluating attention mechanism veracity against expert-labeled explanations. The labels are based on Halbert Dunn's theory of wellness, which gives grounding to our evaluation. We reveal four surprising results about LMs/LLMs: (1) Despite their human-like capabilities, GPT-3.5/4 lag behind RoBERTa, and MedAlpaca, a fine-tuned LLM fails to deliver any remarkable improvements in performance or explanations. (2) Re-examining LMs' predictions based on a confidence-oriented loss function reveals a significant performance drop. (3) Across all LMs/LLMs, the alignment between attention and explanations remains low, with LLMs scoring a dismal 0.0. (4) Most mental health-specific LMs/LLMs overlook domain-specific knowledge and undervalue explanations, causing these discrepancies. This study highlights the need for further research into their consistency and explanations in mental health and well-being.
Authors: Mengdan Zhu, Raasikh Kanjiani, Jiahui Lu, Andrew Choi, Qirui Ye, Liang Zhao
Abstract: Deep generative models like VAEs and diffusion models have advanced various generation tasks by leveraging latent variables to learn data distributions and generate high-quality samples. Despite the field of explainable AI making strides in interpreting machine learning models, understanding latent variables in generative models remains challenging. This paper introduces LatentExplainer, a framework for automatically generating semantically meaningful explanations of latent variables in deep generative models. LatentExplainer tackles three main challenges: inferring the meaning of latent variables, aligning explanations with inductive biases, and handling varying degrees of explainability. By perturbing latent variables and interpreting changes in generated data, the framework provides a systematic approach to understanding and controlling the data generation process, enhancing the transparency and interpretability of deep generative models. We evaluate our proposed method on several real-world and synthetic datasets, and the results demonstrate superior performance in generating high-quality explanations of latent variables.
Authors: Caishuang Huang, Wanxu Zhao, Rui Zheng, Huijie Lv, Shihan Dou, Sixian Li, Xiao Wang, Enyu Zhou, Junjie Ye, Yuming Yang, Tao Gui, Qi Zhang, Xuanjing Huang
Abstract: As the development of large language models (LLMs) rapidly advances, securing these models effectively without compromising their utility has become a pivotal area of research. However, current defense strategies against jailbreak attacks (i.e., efforts to bypass security protocols) often suffer from limited adaptability, restricted general capability, and high cost. To address these challenges, we introduce SafeAligner, a methodology implemented at the decoding stage to fortify defenses against jailbreak attacks. We begin by developing two specialized models: the Sentinel Model, which is trained to foster safety, and the Intruder Model, designed to generate riskier responses. SafeAligner leverages the disparity in security levels between the responses from these models to differentiate between harmful and beneficial tokens, effectively guiding the safety alignment by altering the output token distribution of the target model. Extensive experiments show that SafeAligner can increase the likelihood of beneficial tokens, while reducing the occurrence of harmful ones, thereby ensuring secure alignment with minimal loss to generality.
Authors: Yuanxi Lin, Tonglin Zhou, Yang Xiao
Abstract: Accurate recognition of aviation commands is vital for flight safety and efficiency, as pilots must follow air traffic control instructions precisely. This paper addresses challenges in speech command recognition, such as noisy environments and limited computational resources, by advancing keyword spotting technology. We create a dataset of standardized airport tower commands, including routine and emergency instructions. We enhance broadcasted residual learning with squeeze-and-excitation and time-frame frequency-wise squeeze-and-excitation techniques, resulting in our BC-SENet model. This model focuses on crucial information with fewer parameters. Our tests on five keyword spotting models, including BC-SENet, demonstrate superior accuracy and efficiency. These findings highlight the effectiveness of our model advancements in improving speech command recognition for aviation safety and efficiency in noisy, high-stakes environments. Additionally, BC-SENet shows comparable performance on the common Google Speech Command dataset.
Authors: Junfeng Jiao, Saleh Afroogh, Yiming Xu, Connor Phillips
Abstract: This study addresses ethical issues surrounding Large Language Models (LLMs) within the field of artificial intelligence. It explores the common ethical challenges posed by both LLMs and other AI systems, such as privacy and fairness, as well as ethical challenges uniquely arising from LLMs. It highlights challenges such as hallucination, verifiable accountability, and decoding censorship complexity, which are unique to LLMs and distinct from those encountered in traditional AI systems. The study underscores the need to tackle these complexities to ensure accountability, reduce biases, and enhance transparency in the influential role that LLMs play in shaping information dissemination. It proposes mitigation strategies and future directions for LLM ethics, advocating for interdisciplinary collaboration. It recommends ethical frameworks tailored to specific domains and dynamic auditing systems adapted to diverse contexts. This roadmap aims to guide responsible development and integration of LLMs, envisioning a future where ethical considerations govern AI advancements in society.
Authors: Junfeng Jiao, Saleh Afroogh, Kevin Chen, David Atkinson, Amit Dhurandhar
Abstract: The integration of Generative Artificial Intelligence (GAI) and Large Language Models (LLMs) in academia has spurred a global discourse on their potential pedagogical benefits and ethical considerations. Positive reactions highlight some potential, such as collaborative creativity, increased access to education, and empowerment of trainers and trainees. However, negative reactions raise concerns about ethical complexities, balancing innovation and academic integrity, unequal access, and misinformation risks. Through a systematic survey and text-mining-based analysis of global and national directives, insights from independent research, and eighty university-level guidelines, this study provides a nuanced understanding of the opportunities and challenges posed by GAI and LLMs in education. It emphasizes the importance of balanced approaches that harness the benefits of these technologies while addressing ethical considerations and ensuring equitable access and educational outcomes. The paper concludes with recommendations for fostering responsible innovation and ethical practices to guide the integration of GAI and LLMs in academia.