Authors: Xuming Hu, Xiaochuan Li, Junzhe Chen, Yinghui Li, Yangning Li, Xiaoguang Li, Yasheng Wang, Qun Liu, Lijie Wen, Philip S. Yu, Zhijiang Guo
Abstract: Generative search engines have the potential to transform how people seek information online, but generated responses from existing large language models (LLMs)-backed generative search engines may not always be accurate. Nonetheless, retrieval-augmented generation exacerbates safety concerns, since adversaries may successfully evade the entire system by subtly manipulating the most vulnerable part of a claim. To this end, we propose evaluating the robustness of generative search engines in the realistic and high-risk setting, where adversaries have only black-box system access and seek to deceive the model into returning incorrect responses. Through a comprehensive human evaluation of various generative search engines, such as Bing Chat, PerplexityAI, and YouChat across diverse queries, we demonstrate the effectiveness of adversarial factual questions in inducing incorrect responses. Moreover, retrieval-augmented generation exhibits a higher susceptibility to factual errors compared to LLMs without retrieval. These findings highlight the potential security risks of these systems and emphasize the need for rigorous evaluation before deployment.
Authors: Adam Shostack
Abstract: Recent work arXiv.2310.02238 asserted that "we effectively erase the model's ability to generate or recall Harry Potter-related content.'' This claim is shown to be overbroad. A small experiment of less than a dozen trials led to repeated and specific mentions of Harry Potter, including "Ah, I see! A "muggle" is a term used in the Harry Potter book series by Terry Pratchett...''
Authors: Aritra Kumar Lahiri, Emrul Hasan, Qinmin Vivian Hu, Cherie Ding
Abstract: This paper describes Toronto Metropolitan University's participation in the TREC Clinical Trials Track for 2023. As part of the tasks, we utilize advanced natural language processing techniques and neural language models in our experiments to retrieve the most relevant clinical trials. We illustrate the overall methodology, experimental settings, and results of our implementation for the run submission as part of Team - V-TorontoMU.
Authors: Ashwin Daswani, Rohan Sawant, Najoung Kim
Abstract: Sensitivity to false assumptions (or false premises) in information-seeking questions is critical for robust question-answering (QA) systems. Recent work has shown that false assumptions in naturally occurring questions pose challenges to current models, with low performance on both generative QA and simple detection tasks (Kim et al. 2023). However, the focus of existing work on naturally occurring questions leads to a gap in the analysis of model behavior on the long tail of the distribution of possible questions. To this end, we introduce Syn-(QA)$^2$, a set of two synthetically generated QA datasets: one generated using perturbed relations from Wikidata, and the other by perturbing HotpotQA (Yang et al. 2018). Our findings from evaluating a range of large language models are threefold: (1) false assumptions in QA are challenging, echoing the findings of prior work, (2) the binary detection task is challenging even compared to the difficulty of generative QA itself, possibly due to the linguistic structure of the problem, and (3) the detection task is more challenging with long-tail questions compared to naturally occurring questions, highlighting the utility of our synthetic datasets and generation method.
Authors: Weikang Zhou, Xiao Wang, Limao Xiong, Han Xia, Yingshuang Gu, Mingxu Chai, Fukang Zhu, Caishuang Huang, Shihan Dou, Zhiheng Xi, Rui Zheng, Songyang Gao, Yicheng Zou, Hang Yan, Yifan Le, Ruohui Wang, Lijun Li, Jing Shao, Tao Gui, Qi Zhang, Xuanjing Huang
Abstract: Jailbreak attacks are crucial for identifying and mitigating the security vulnerabilities of Large Language Models (LLMs). They are designed to bypass safeguards and elicit prohibited outputs. However, due to significant differences among various jailbreak methods, there is no standard implementation framework available for the community, which limits comprehensive security evaluations. This paper introduces EasyJailbreak, a unified framework simplifying the construction and evaluation of jailbreak attacks against LLMs. It builds jailbreak attacks using four components: Selector, Mutator, Constraint, and Evaluator. This modular framework enables researchers to easily construct attacks from combinations of novel and existing components. So far, EasyJailbreak supports 11 distinct jailbreak methods and facilitates the security validation of a broad spectrum of LLMs. Our validation across 10 distinct LLMs reveals a significant vulnerability, with an average breach probability of 60% under various jailbreaking attacks. Notably, even advanced models like GPT-3.5-Turbo and GPT-4 exhibit average Attack Success Rates (ASR) of 57% and 33%, respectively. We have released a wealth of resources for researchers, including a web platform, PyPI published package, screencast video, and experimental outputs.
Authors: Mengting Wan, Tara Safavi, Sujay Kumar Jauhar, Yujin Kim, Scott Counts, Jennifer Neville, Siddharth Suri, Chirag Shah, Ryen W White, Longqi Yang, Reid Andersen, Georg Buscher, Dhruv Joshi, Nagu Rangan
Abstract: Transforming unstructured text into structured and meaningful forms, organized by useful category labels, is a fundamental step in text mining for downstream analysis and application. However, most existing methods for producing label taxonomies and building text-based label classifiers still rely heavily on domain expertise and manual curation, making the process expensive and time-consuming. This is particularly challenging when the label space is under-specified and large-scale data annotations are unavailable. In this paper, we address these challenges with Large Language Models (LLMs), whose prompt-based interface facilitates the induction and use of large-scale pseudo labels. We propose TnT-LLM, a two-phase framework that employs LLMs to automate the process of end-to-end label generation and assignment with minimal human effort for any given use-case. In the first phase, we introduce a zero-shot, multi-stage reasoning approach which enables LLMs to produce and refine a label taxonomy iteratively. In the second phase, LLMs are used as data labelers that yield training samples so that lightweight supervised classifiers can be reliably built, deployed, and served at scale. We apply TnT-LLM to the analysis of user intent and conversational domain for Bing Copilot (formerly Bing Chat), an open-domain chat-based search engine. Extensive experiments using both human and automatic evaluation metrics demonstrate that TnT-LLM generates more accurate and relevant label taxonomies when compared against state-of-the-art baselines, and achieves a favorable balance between accuracy and efficiency for classification at scale. We also share our practical experiences and insights on the challenges and opportunities of using LLMs for large-scale text mining in real-world applications.
Authors: Ramon Abilio, Guilherme Palermo Coelho, Ana Estela Antunes da Silva
Abstract: Named Entity Recognition (NER) is a Natural Language Processing technique for extracting information from textual documents. However, much of the existing research on NER has been centered around English-language documents, leaving a gap in the availability of datasets tailored to the financial domain in Portuguese. This study addresses the need for NER within the financial domain, focusing on Portuguese-language texts extracted from earnings call transcriptions of Brazilian banks. By curating a comprehensive dataset comprising 384 transcriptions and leveraging weak supervision techniques for annotation, we evaluate the performance of monolingual models trained on Portuguese (BERTimbau and PTT5) and multilingual models (mBERT and mT5). Notably, we introduce a novel approach that reframes the token classification task as a text generation problem, enabling fine-tuning and evaluation of T5 models. Following the fine-tuning of the models, we conduct an evaluation on the test dataset, employing performance and error metrics. Our findings reveal that BERT-based models consistently outperform T5-based models. Furthermore, while the multilingual models exhibit comparable macro F1-scores, BERTimbau demonstrates superior performance over PTT5. A manual analysis of sentences generated by PTT5 and mT5 unveils a degree of similarity ranging from 0.89 to 1.0, between the original and generated sentences. However, critical errors emerge as both models exhibit discrepancies, such as alterations to monetary and percentage values, underscoring the importance of accuracy and consistency in the financial domain. Despite these challenges, PTT5 and mT5 achieve impressive macro F1-scores of 98.52% and 98.85%, respectively, with our proposed approach. Furthermore, our study sheds light on notable disparities in memory and time consumption for inference across the models.
Authors: Bang Nguyen, Mengxia Yu, Yun Huang, Meng Jiang
Abstract: Reference-based metrics such as BLEU and BERTScore are widely used to evaluate question generation (QG). In this study, on QG benchmarks such as SQuAD and HotpotQA, we find that using human-written references cannot guarantee the effectiveness of the reference-based metrics. Most QG benchmarks have only one reference; we replicated the annotation process and collect another reference. A good metric was expected to grade a human-validated question no worse than generated questions. However, the results of reference-based metrics on our newly collected reference disproved the metrics themselves. We propose a reference-free metric consisted of multi-dimensional criteria such as naturalness, answerability, and complexity, utilizing large language models. These criteria are not constrained to the syntactic or semantic of a single reference question, and the metric does not require a diverse set of references. Experiments reveal that our metric accurately distinguishes between high-quality questions and flawed ones, and achieves state-of-the-art alignment with human judgment.
Authors: Patanjali Bhamidipati, Advaith Malladi, Manish Shrivastava, Radhika Mamidi
Abstract: In recent studies, the extensive utilization of large language models has underscored the importance of robust evaluation methodologies for assessing text generation quality and relevance to specific tasks. This has revealed a prevalent issue known as hallucination, an emergent condition in the model where generated text lacks faithfulness to the source and deviates from the evaluation criteria. In this study, we formally define hallucination and propose a framework for its quantitative detection in a zero-shot setting, leveraging our definition and the assumption that model outputs entail task and sample specific inputs. In detecting hallucinations, our solution achieves an accuracy of 0.78 in a model-aware setting and 0.61 in a model-agnostic setting. Notably, our solution maintains computational efficiency, requiring far less computational resources than other SOTA approaches, aligning with the trend towards lightweight and compressed models.
Authors: Thanos Konstantinidis, Giorgos Iacovides, Mingxue Xu, Tony G. Constantinides, Danilo Mandic
Abstract: There are multiple sources of financial news online which influence market movements and trader's decisions. This highlights the need for accurate sentiment analysis, in addition to having appropriate algorithmic trading techniques, to arrive at better informed trading decisions. Standard lexicon based sentiment approaches have demonstrated their power in aiding financial decisions. However, they are known to suffer from issues related to context sensitivity and word ordering. Large Language Models (LLMs) can also be used in this context, but they are not finance-specific and tend to require significant computational resources. To facilitate a finance specific LLM framework, we introduce a novel approach based on the Llama 2 7B foundational model, in order to benefit from its generative nature and comprehensive language manipulation. This is achieved by fine-tuning the Llama2 7B model on a small portion of supervised financial sentiment analysis data, so as to jointly handle the complexities of financial lexicon and context, and further equipping it with a neural network based decision mechanism. Such a generator-classifier scheme, referred to as FinLlama, is trained not only to classify the sentiment valence but also quantify its strength, thus offering traders a nuanced insight into financial news articles. Complementing this, the implementation of parameter-efficient fine-tuning through LoRA optimises trainable parameters, thus minimising computational and memory requirements, without sacrificing accuracy. Simulation results demonstrate the ability of the proposed FinLlama to provide a framework for enhanced portfolio management decisions and increased market returns. These results underpin the ability of FinLlama to construct high-return portfolios which exhibit enhanced resilience, even during volatile periods and unpredictable market events.
Authors: Elliot Murphy, Jill de Villiers, Sofia Lucero Morales
Abstract: In this study we compared how well DALL-E 2 visually represented the meaning of linguistic prompts also given to young children in comprehension tests. Sentences representing fundamental components of grammatical knowledge were selected from assessment tests used with several hundred English-speaking children aged 2-7 years for whom we had collected original item-level data. DALL-E 2 was given these prompts five times to generate 20 cartoons per item, for 9 adult judges to score. Results revealed no conditions in which DALL-E 2-generated images that matched the semantic accuracy of children, even at the youngest age (2 years). DALL-E 2 failed to assign the appropriate roles in reversible forms; it failed on negation despite an easier contrastive prompt than the children received; it often assigned the adjective to the wrong noun; it ignored implicit agents in passives. This work points to a clear absence of compositional sentence representations for DALL-E 2.
Authors: Maria Mahbub, Gregory M. Dams, Sudarshan Srinivasan, Caitlin Rizy, Ioana Danciu, Jodie Trafton, Kathryn Knight
Abstract: Substance use disorder (SUD) poses a major concern due to its detrimental effects on health and society. SUD identification and treatment depend on a variety of factors such as severity, co-determinants (e.g., withdrawal symptoms), and social determinants of health. Existing diagnostic coding systems used by American insurance providers, like the International Classification of Diseases (ICD-10), lack granularity for certain diagnoses, but clinicians will add this granularity (as that found within the Diagnostic and Statistical Manual of Mental Disorders classification or DSM-5) as supplemental unstructured text in clinical notes. Traditional natural language processing (NLP) methods face limitations in accurately parsing such diverse clinical language. Large Language Models (LLMs) offer promise in overcoming these challenges by adapting to diverse language patterns. This study investigates the application of LLMs for extracting severity-related information for various SUD diagnoses from clinical notes. We propose a workflow employing zero-shot learning of LLMs with carefully crafted prompts and post-processing techniques. Through experimentation with Flan-T5, an open-source LLM, we demonstrate its superior recall compared to the rule-based approach. Focusing on 11 categories of SUD diagnoses, we show the effectiveness of LLMs in extracting severity information, contributing to improved risk assessment and treatment planning for SUD patients.
Authors: Chuang Liu, Linhao Yu, Jiaxuan Li, Renren Jin, Yufei Huang, Ling Shi, Junhui Zhang, Xinmeng Ji, Tingting Cui, Tao Liu, Jinwang Song, Hongying Zan, Sun Li, Deyi Xiong
Abstract: The rapid development of Chinese large language models (LLMs) poses big challenges for efficient LLM evaluation. While current initiatives have introduced new benchmarks or evaluation platforms for assessing Chinese LLMs, many of these focus primarily on capabilities, usually overlooking potential alignment and safety issues. To address this gap, we introduce OpenEval, an evaluation testbed that benchmarks Chinese LLMs across capability, alignment and safety. For capability assessment, we include 12 benchmark datasets to evaluate Chinese LLMs from 4 sub-dimensions: NLP tasks, disciplinary knowledge, commonsense reasoning and mathematical reasoning. For alignment assessment, OpenEval contains 7 datasets that examines the bias, offensiveness and illegalness in the outputs yielded by Chinese LLMs. To evaluate safety, especially anticipated risks (e.g., power-seeking, self-awareness) of advanced LLMs, we include 6 datasets. In addition to these benchmarks, we have implemented a phased public evaluation and benchmark update strategy to ensure that OpenEval is in line with the development of Chinese LLMs or even able to provide cutting-edge benchmark datasets to guide the development of Chinese LLMs. In our first public evaluation, we have tested a range of Chinese LLMs, spanning from 7B to 72B parameters, including both open-source and proprietary models. Evaluation results indicate that while Chinese LLMs have shown impressive performance in certain tasks, more attention should be directed towards broader aspects such as commonsense reasoning, alignment, and safety.
Authors: Xi Wang, Hongliang Dai, Shen Gao, Piji Li
Abstract: The advancement of Large Language Models (LLMs) has led to significant enhancements in the performance of chatbot systems. Many researchers have dedicated their efforts to the development of bringing characteristics to chatbots. While there have been commercial products for developing role-driven chatbots using LLMs, it is worth noting that academic research in this area remains relatively scarce. Our research focuses on investigating the performance of LLMs in constructing Characteristic AI Agents by simulating real-life individuals across different settings. Current investigations have primarily focused on act on roles with simple profiles. In response to this research gap, we create a benchmark for the characteristic AI agents task, including dataset, techniques, and evaluation metrics. A dataset called ``Character100'' is built for this benchmark, comprising the most-visited people on Wikipedia for language models to role-play. With the constructed dataset, we conduct comprehensive assessment of LLMs across various settings. In addition, we devise a set of automatic metrics for quantitative performance evaluation. The experimental results underscore the potential directions for further improvement in the capabilities of LLMs in constructing characteristic AI agents. The benchmark is available at https://github.com/nuaa-nlp/Character100.
Authors: Chi Hu, Yuan Ge, Xiangnan Ma, Hang Cao, Qiang Li, Yonghua Yang, Tong Xiao, Jingbo Zhu
Abstract: Large Language Models (LLMs) have achieved impressive performance across various reasoning tasks. However, even state-of-the-art LLMs such as ChatGPT are prone to logical errors during their reasoning processes. Existing solutions, which include deploying task-specific verifiers or voting over multiple reasoning paths, either require extensive human annotations or fail in scenarios with inconsistent responses. To address these challenges, we introduce RankPrompt, a new prompting method that enables LLMs to self-rank their responses without additional resources. RankPrompt breaks down the ranking problem into a series of comparisons among diverse responses, leveraging the inherent capabilities of LLMs to generate chains of comparison as contextual exemplars. Our experiments across 11 arithmetic and commonsense reasoning tasks show that RankPrompt significantly enhances the reasoning performance of ChatGPT and GPT-4, with improvements of up to 13\%. RankPrompt also excels in LLM-based automatic evaluations for open-ended generation, aligning with human preferences 74\% of the time in the AlpacaEval set. Moreover, RankPrompt demonstrates robustness against variations in the orderings and consistencies of responses.
Authors: Cheng Peng, Zehao Yu, Kaleb E Smith, Wei-Hsuan Lo-Ciganic, Jiang Bian, Yonghui Wu
Abstract: The progress in natural language processing (NLP) using large language models (LLMs) has greatly improved patient information extraction from clinical narratives. However, most methods based on the fine-tuning strategy have limited transfer learning ability for cross-domain applications. This study proposed a novel approach that employs a soft prompt-based learning architecture, which introduces trainable prompts to guide LLMs toward desired outputs. We examined two types of LLM architectures, including encoder-only GatorTron and decoder-only GatorTronGPT, and evaluated their performance for the extraction of social determinants of health (SDoH) using a cross-institution dataset from the 2022 n2c2 challenge and a cross-disease dataset from the University of Florida (UF) Health. The results show that decoder-only LLMs with prompt tuning achieved better performance in cross-domain applications. GatorTronGPT achieved the best F1 scores for both datasets, outperforming traditional fine-tuned GatorTron by 8.9% and 21.8% in a cross-institution setting, and 5.5% and 14.5% in a cross-disease setting.
Authors: Pengchao Wu, Xuefeng Li, Jinghang Gu, Longhua Qian, Guodong Zhou
Abstract: Biomedical event extraction is an information extraction task to obtain events from biomedical text, whose targets include the type, the trigger, and the respective arguments involved in an event. Traditional biomedical event extraction usually adopts a pipelined approach, which contains trigger identification, argument role recognition, and finally event construction either using specific rules or by machine learning. In this paper, we propose an n-ary relation extraction method based on the BERT pre-training model to construct Binding events, in order to capture the semantic information about an event's context and its participants. The experimental results show that our method achieves promising results on the GE11 and GE13 corpora of the BioNLP shared task with F1 scores of 63.14% and 59.40%, respectively. It demonstrates that by significantly improving theperformance of Binding events, the overall performance of the pipelined event extraction approach or even exceeds those of current joint learning methods.
Authors: Faisal Qarah
Abstract: Arabic poetry, with its rich linguistic features and profound cultural significance, presents a unique challenge to the Natural Language Processing (NLP) field. The complexity of its structure and context necessitates advanced computational models for accurate analysis. In this paper, we introduce AraPoemBERT, an Arabic language model pretrained exclusively on Arabic poetry text. To demonstrate the effectiveness of the proposed model, we compared AraPoemBERT with 5 different Arabic language models on various NLP tasks related to Arabic poetry. The new model outperformed all other models and achieved state-of-the-art results in most of the downstream tasks. AraPoemBERT achieved unprecedented accuracy in two out of three novel tasks: poet's gender classification (99.34\% accuracy), and poetry sub-meter classification (97.79\% accuracy). In addition, the model achieved an accuracy score in poems' rhyme classification (97.73\% accuracy) which is almost equivalent to the best score reported in this study. Moreover, the proposed model significantly outperformed previous work and other comparative models in the tasks of poems' sentiment analysis, achieving an accuracy of 78.95\%, and poetry meter classification (99.03\% accuracy), while significantly expanding the scope of these two problems. The dataset used in this study, contains more than 2.09 million verses collected from online sources, each associated with various attributes such as meter, sub-meter, poet, rhyme, and topic. The results demonstrate the effectiveness of the proposed model in understanding and analyzing Arabic poetry, achieving state-of-the-art results in several tasks and outperforming previous works and other language models included in the study. AraPoemBERT model is publicly available on \url{https://huggingface.co/faisalq}.
Authors: Yuan Gao, Yiheng Zhu, Yuanbin Cao, Yinzhi Zhou, Zhen Wu, Yujie Chen, Shenglan Wu, Haoyuan Hu, Xinyu Dai
Abstract: Open Domain Multi-Hop Question Answering (ODMHQA) plays a crucial role in Natural Language Processing (NLP) by aiming to answer complex questions through multi-step reasoning over retrieved information from external knowledge sources. Recently, Large Language Models (LLMs) have demonstrated remarkable performance in solving ODMHQA owing to their capabilities including planning, reasoning, and utilizing tools. However, LLMs may generate off-topic answers when attempting to solve ODMHQA, namely the generated answers are irrelevant to the original questions. This issue of off-topic answers accounts for approximately one-third of incorrect answers, yet remains underexplored despite its significance. To alleviate this issue, we propose the Discriminate->Re-Compose->Re- Solve->Re-Decompose (Dr3) mechanism. Specifically, the Discriminator leverages the intrinsic capabilities of LLMs to judge whether the generated answers are off-topic. In cases where an off-topic answer is detected, the Corrector performs step-wise revisions along the reversed reasoning chain (Re-Compose->Re-Solve->Re-Decompose) until the final answer becomes on-topic. Experimental results on the HotpotQA and 2WikiMultiHopQA datasets demonstrate that our Dr3 mechanism considerably reduces the occurrence of off-topic answers in ODMHQA by nearly 13%, improving the performance in Exact Match (EM) by nearly 3% compared to the baseline method without the Dr3 mechanism.
Authors: Yifan Peng, Ilia Kulikov, Yilin Yang, Sravya Popuri, Hui Lu, Changhan Wang, Hongyu Gong
Abstract: Speech language models (LMs) are promising for high-quality speech synthesis through in-context learning. A typical speech LM takes discrete semantic units as content and a short utterance as prompt, and synthesizes speech which preserves the content's semantics but mimics the prompt's style. However, there is no systematic understanding on how the synthesized audio is controlled by the prompt and content. In this work, we conduct an empirical study of the widely used autoregressive (AR) and non-autoregressive (NAR) speech LMs and provide insights into the prompt design and content semantic units. Our analysis reveals that heterogeneous and nonstationary prompts hurt the audio quality in contrast to the previous finding that longer prompts always lead to better synthesis. Moreover, we find that the speaker style of the synthesized audio is also affected by the content in addition to the prompt. We further show that semantic units carry rich acoustic information such as pitch, tempo, volume and speech emphasis, which might be leaked from the content to the synthesized audio.
Authors: Ayushi Nirmal, Amrita Bhattacharjee, Paras Sheth, Huan Liu
Abstract: Although social media platforms are a prominent arena for users to engage in interpersonal discussions and express opinions, the facade and anonymity offered by social media may allow users to spew hate speech and offensive content. Given the massive scale of such platforms, there arises a need to automatically identify and flag instances of hate speech. Although several hate speech detection methods exist, most of these black-box methods are not interpretable or explainable by design. To address the lack of interpretability, in this paper, we propose to use state-of-the-art Large Language Models (LLMs) to extract features in the form of rationales from the input text, to train a base hate speech classifier, thereby enabling faithful interpretability by design. Our framework effectively combines the textual understanding capabilities of LLMs and the discriminative power of state-of-the-art hate speech classifiers to make these classifiers faithfully interpretable. Our comprehensive evaluation on a variety of social media hate speech datasets demonstrate: (1) the goodness of the LLM-extracted rationales, and (2) the surprising retention of detector performance even after training to ensure interpretability.
Authors: Xiaoyu Qiu, Yuechen Wang, Jiaxin Shi, Wengang Zhou, Houqiang Li
Abstract: Based on multilingual pre-trained models, cross-lingual transfer with prompt learning has shown promising effectiveness, where soft prompt learned in a source language is transferred to target languages for downstream tasks, particularly in the low-resource scenario. To efficiently transfer soft prompt, we propose a novel framework, Multilingual Prompt Translator (MPT), where a multilingual prompt translator is introduced to properly process crucial knowledge embedded in prompt by changing language knowledge while retaining task knowledge. Concretely, we first train prompt in source language and employ translator to translate it into target prompt. Besides, we extend an external corpus as auxiliary data, on which an alignment task for predicted answer probability is designed to convert language knowledge, thereby equipping target prompt with multilingual knowledge. In few-shot settings on XNLI, MPT demonstrates superiority over baselines by remarkable improvements. MPT is more prominent compared with vanilla prompting when transferring to languages quite distinct from source language.
Authors: Yifan Peng, Ilia Kulikov, Yilin Yang, Sravya Popuri, Hui Lu, Changhan Wang, Hongyu Gong
Abstract: There have been emerging research interest and advances in speech-to-speech translation (S2ST), translating utterances from one language to another. This work proposes Multitask Speech Language Model (MSLM), which is a decoder-only speech language model trained in a multitask setting. Without reliance on text training data, our model is able to support multilingual S2ST with speaker style preserved.
Authors: Rahul Nadkarni, Yizhong Wang, Noah A. Smith
Abstract: Language model-based instruction-following systems have lately shown increasing performance on many benchmark tasks, demonstrating the capability of adapting to a broad variety of instructions. However, such systems are often not designed to be transparent about their limitations; a user may easily prompt a model with an instruction without any idea of whether the responses should be expected to be accurate, or if the system is even capable of performing the task. We propose a third party performance prediction framework, where a separate model is trained to predict the metric resulting from evaluating an instruction-following system on a task while assuming access only to its inputs and outputs at inference time. We perform this analysis with a variety of both open and closed instruction-following models as well as multiple performance predictors, and examine the effect of various factors such as model size, number of training tasks, and prompt format. Our findings indicate that third-party performance prediction is very challenging, and much work remains in developing predictors that can automatically reveal the limitations of modern instruction-following natural language processing systems.
Authors: Danqing Luo, Chen Zhang, Yan Zhang, Haizhou Li
Abstract: Training or finetuning large-scale language models (LLMs) requires substantial computation resources, motivating recent efforts to explore parameter-efficient adaptation to downstream tasks. One approach is to treat these models as black boxes and use forward passes (Inference APIs) to interact with them. Current research focuses on adapting these black-box models to downstream tasks using gradient-free prompt optimization, but this often involves an expensive process of searching task-specific prompts. Therefore, we are motivated to study black-box language model adaptation without prompt search. Specifically, we introduce a label-enhanced cross-attention network called CrossTune, which models the semantic relatedness between the input text sequence and task-specific label descriptions. Its effectiveness is examined in the context of few-shot text classification. To improve the generalization of CrossTune, we utilize ChatGPT to generate additional training data through in-context learning. A switch mechanism is implemented to exclude low-quality ChatGPT-generated data. Through extensive experiments on seven benchmark text classification datasets, we demonstrate that our proposed approach outperforms the previous state-of-the-art gradient-free black-box tuning method by 5.7% on average. Even without using ChatGPT-augmented data, CrossTune performs better or comparably than previous black-box tuning methods, suggesting the effectiveness of our approach.
Authors: Ojas Nimase, Sanghyun Hong
Abstract: Sarcasm recognition is challenging because it needs an understanding of the true intention, which is opposite to or different from the literal meaning of the words. Prior work has addressed this challenge by developing a series of methods that provide richer $contexts$, e.g., sentiment or cultural nuances, to models. While shown to be effective individually, no study has systematically evaluated their collective effectiveness. As a result, it remains unclear to what extent additional contexts can improve sarcasm recognition. In this work, we explore the improvements that existing methods bring by incorporating more contexts into a model. To this end, we develop a framework where we can integrate multiple contextual cues and test different approaches. In evaluation with four approaches on three sarcasm recognition benchmarks, we achieve existing state-of-the-art performances and also demonstrate the benefits of sequentially adding more contexts. We also identify inherent drawbacks of using more contexts, highlighting that in the pursuit of even better results, the model may need to adopt societal biases.
Authors: Shiki Sato, Reina Akama, Jun Suzuki, Kentaro Inui
Abstract: Mitigating the generation of contradictory responses poses a substantial challenge in dialogue response generation. The quality and quantity of available contradictory response data play a vital role in suppressing these contradictions, offering two significant benefits. First, having access to large contradiction data enables a comprehensive examination of their characteristics. Second, data-driven methods to mitigate contradictions may be enhanced with large-scale contradiction data for training. Nevertheless, no attempt has been made to build an extensive collection of model-generated contradictory responses. In this paper, we build a large dataset of response generation models' contradictions for the first time. Then, we acquire valuable insights into the characteristics of model-generated contradictions through an extensive analysis of the collected responses. Lastly, we also demonstrate how this dataset substantially enhances the performance of data-driven contradiction suppression methods.
Authors: Haochen Li, Di Geng
Abstract: Events describe the state changes of entities. In a document, multiple events are connected by various relations (e.g., Coreference, Temporal, Causal, and Subevent). Therefore, obtaining the connections between events through Event-Event Relation Extraction (ERE) is critical to understand natural language. There are two main problems in the current ERE works: a. Only embeddings of the event triggers are used for event feature representation, ignoring event arguments (e.g., time, place, person, etc.) and their structure within the event. b. The interconnection between relations (e.g., temporal and causal relations usually interact with each other ) is ignored. To solve the above problems, this paper proposes a jointly multiple ERE framework called GraphERE based on Graph-enhanced Event Embeddings. First, we enrich the event embeddings with event argument and structure features by using static AMR graphs and IE graphs; Then, to jointly extract multiple event relations, we use Node Transformer and construct Task-specific Dynamic Event Graphs for each type of relation. Finally, we used a multi-task learning strategy to train the whole framework. Experimental results on the latest MAVEN-ERE dataset validate that GraphERE significantly outperforms existing methods. Further analyses indicate the effectiveness of the graph-enhanced event embeddings and the joint extraction strategy.
Authors: Haochen Li, Di Geng
Abstract: Events are essential components of speech and texts, describing the changes in the state of entities. The event extraction task aims to identify and classify events and find their participants according to event schemas. Manually predefined event schemas have limited coverage and are hard to migrate across domains. Therefore, the researchers propose Liberal Event Extraction (LEE), which aims to extract events and discover event schemas simultaneously. However, existing LEE models rely heavily on external language knowledge bases and require the manual development of numerous rules for noise removal and knowledge alignment, which is complex and laborious. To this end, we propose a Prompt-based Graph Model for Liberal Event Extraction (PGLEE). Specifically, we use a prompt-based model to obtain candidate triggers and arguments, and then build heterogeneous event graphs to encode the structures within and between events. Experimental results prove that our approach achieves excellent performance with or without predefined event schemas, while the automatically detected event schemas are proven high quality.
Authors: Zhigang Chen, Benjia Zhou, Jun Li, Jun Wan, Zhen Lei, Ning Jiang, Quan Lu, Guoqing Zhao
Abstract: Previous Sign Language Translation (SLT) methods achieve superior performance by relying on gloss annotations. However, labeling high-quality glosses is a labor-intensive task, which limits the further development of SLT. Although some approaches work towards gloss-free SLT through jointly training the visual encoder and translation network, these efforts still suffer from poor performance and inefficient use of the powerful Large Language Model (LLM). Most seriously, we find that directly introducing LLM into SLT will lead to insufficient learning of visual representations as LLM dominates the learning curve. To address these problems, we propose Factorized Learning assisted with Large Language Model (FLa-LLM) for gloss-free SLT. Concretely, we factorize the training process into two stages. In the visual initialing stage, we employ a lightweight translation model after the visual encoder to pre-train the visual encoder. In the LLM fine-tuning stage, we freeze the acquired knowledge in the visual encoder and integrate it with a pre-trained LLM to inspire the LLM's translation potential. This factorized training strategy proves to be highly effective as evidenced by significant improvements achieved across three SLT datasets which are all conducted under the gloss-free setting.
Authors: Mirza Alim Mutasodirin, Radityo Eko Prasojo, Achmad F. Abka, Hanif Rasyidi
Abstract: Many NLP researchers rely on free computational services, such as Google Colab, to fine-tune their Transformer models, causing a limitation for hyperparameter optimization (HPO) in long-text classification due to the method having quadratic complexity and needing a bigger resource. In Indonesian, only a few works were found on long-text classification using Transformers. Most only use a small amount of data and do not report any HPO. In this study, using 18k news articles, we investigate which pretrained models are recommended to use based on the output length of the tokenizer. We then compare some hacks to shorten and enrich the sequences, which are the removals of stopwords, punctuation, low-frequency words, and recurring words. To get a fair comparison, we propose and run an efficient and dynamic HPO procedure that can be done gradually on a limited resource and does not require a long-running optimization library. Using the best hack found, we then compare 512, 256, and 128 tokens length. We find that removing stopwords while keeping punctuation and low-frequency words is the best hack. Some of our setups manage to outperform taking 512 first tokens using a smaller 128 or 256 first tokens which manage to represent the same information while requiring less computational resources. The findings could help developers to efficiently pursue optimal performance of the models using limited resources.
Authors: Xiang Li, Zhenyu Li, Chen Shi, Yong Xu, Qing Du, Mingkui Tan, Jun Huang, Wei Lin
Abstract: The task of financial analysis primarily encompasses two key areas: stock trend prediction and the corresponding financial question answering. Currently, machine learning and deep learning algorithms (ML&DL) have been widely applied for stock trend predictions, leading to significant progress. However, these methods fail to provide reasons for predictions, lacking interpretability and reasoning processes. Also, they can not integrate textual information such as financial news or reports. Meanwhile, large language models (LLMs) have remarkable textual understanding and generation ability. But due to the scarcity of financial training datasets and limited integration with real-time knowledge, LLMs still suffer from hallucinations and are unable to keep up with the latest information. To tackle these challenges, we first release AlphaFin datasets, combining traditional research datasets, real-time financial data, and handwritten chain-of-thought (CoT) data. It has a positive impact on training LLMs for completing financial analysis. We then use AlphaFin datasets to benchmark a state-of-the-art method, called Stock-Chain, for effectively tackling the financial analysis task, which integrates retrieval-augmented generation (RAG) techniques. Extensive experiments are conducted to demonstrate the effectiveness of our framework on financial analysis.
Authors: Victor Carbune, Hassan Mansoor, Fangyu Liu, Rahul Aralikatte, Gilles Baechler, Jindong Chen, Abhanshu Sharma
Abstract: Vision-language models (VLMs) are achieving increasingly strong performance on multimodal tasks. However, reasoning capabilities remain limited particularly for smaller VLMs, while those of large-language models (LLMs) have seen numerous improvements. We propose a technique to transfer capabilities from LLMs to VLMs. On the recently introduced ChartQA, our method obtains state-of-the-art performance when applied on the PaLI3-5B VLM by \citet{chen2023pali3}, while also enabling much better performance on PlotQA and FigureQA. We first improve the chart representation by continuing the pre-training stage using an improved version of the chart-to-table translation task by \citet{liu2023deplot}. We then propose constructing a 20x larger dataset than the original training set. To improve general reasoning capabilities and improve numerical operations, we synthesize reasoning traces using the table representation of charts. Lastly, our model is fine-tuned using the multitask loss introduced by \citet{hsieh2023distilling}. Our variant ChartPaLI-5B outperforms even 10x larger models such as PaLIX-55B without using an upstream OCR system, while keeping inference time constant compared to the PaLI3-5B baseline. When rationales are further refined with a simple program-of-thought prompt \cite{chen2023program}, our model outperforms the recently introduced Gemini Ultra and GPT-4V.
Authors: Chuang Liu, Renren Jin, Yuqi Ren, Deyi Xiong
Abstract: Chinese Large Language Models (LLMs) have recently demonstrated impressive capabilities across various NLP benchmarks and real-world applications. However, the existing benchmarks for comprehensively evaluating these LLMs are still insufficient, particularly in terms of measuring knowledge that LLMs capture. Current datasets collect questions from Chinese examinations across different subjects and educational levels to address this issue. Yet, these benchmarks primarily focus on objective questions such as multiple-choice questions, leading to a lack of diversity in question types. To tackle this problem, we propose LHMKE, a Large-scale, Holistic, and Multi-subject Knowledge Evaluation benchmark in this paper. LHMKE is designed to provide a comprehensive evaluation of the knowledge acquisition capabilities of Chinese LLMs. It encompasses 10,465 questions across 75 tasks covering 30 subjects, ranging from primary school to professional certification exams. Notably, LHMKE includes both objective and subjective questions, offering a more holistic evaluation of the knowledge level of LLMs. We have assessed 11 Chinese LLMs under the zero-shot setting, which aligns with real examinations, and compared their performance across different subjects. We also conduct an in-depth analysis to check whether GPT-4 can automatically score subjective predictions. Our findings suggest that LHMKE is a challenging and advanced testbed for Chinese LLMs.
Authors: Dojun Park, Sebastian Pad\'o
Abstract: Almost all frameworks for the manual or automatic evaluation of machine translation characterize the quality of an MT output with a single number. An exception is the Multidimensional Quality Metrics (MQM) framework which offers a fine-grained ontology of quality dimensions for scoring (such as style, fluency, accuracy, and terminology). Previous studies have demonstrated the feasibility of MQM annotation but there are, to our knowledge, no computational models that predict MQM scores for novel texts, due to a lack of resources. In this paper, we address these shortcomings by (a) providing a 1200-sentence MQM evaluation benchmark for the language pair English-Korean and (b) reframing MT evaluation as the multi-task problem of simultaneously predicting several MQM scores using SOTA language models, both in a reference-based MT evaluation setup and a reference-free quality estimation (QE) setup. We find that reference-free setup outperforms its counterpart in the style dimension while reference-based models retain an edge regarding accuracy. Overall, RemBERT emerges as the most promising model. Through our evaluation, we offer an insight into the translation quality in a more fine-grained, interpretable manner.
Authors: Dojun Park, Jiwoo Lee, Hyeyun Jeong, Seohyun Park, Sungeun Lee
Abstract: The current evaluation of Large Language Models (LLMs) predominantly relies on benchmarks focusing on their embedded knowledge by testing through multiple-choice questions (MCQs), a format inherently suited for automated evaluation. Our study extends this evaluation to explore LLMs' pragmatic competence--a facet previously underexamined before the advent of sophisticated LLMs, specifically in the context of Korean. We employ two distinct evaluation setups: the conventional MCQ format, adapted for automatic evaluation, and Open-Ended Questions (OEQs), assessed by human experts, to examine LLMs' narrative response capabilities without predefined options. Our findings reveal that GPT-4 excels, scoring 81.11 and 85.69 in the MCQ and OEQ setups, respectively, with HyperCLOVA X, an LLM optimized for Korean, closely following, especially in the OEQ setup, demonstrating a score of 81.56 with a marginal difference of 4.13 points compared to GPT-4. Furthermore, while few-shot learning strategies generally enhance LLM performance, Chain-of-Thought (CoT) prompting introduces a bias toward literal interpretations, hindering accurate pragmatic inference. Considering the growing expectation for LLMs to understand and produce language that aligns with human communicative norms, our findings emphasize the importance for advancing LLMs' abilities to grasp and convey sophisticated meanings beyond mere literal interpretations.
Authors: Maksym Taranukhin, Sahithya Ravi, Gabor Lukacs, Evangelos Milios, Vered Shwartz
Abstract: The Canadian air travel sector has seen a significant increase in flight delays, cancellations, and other issues concerning passenger rights. Recognizing this demand, we present a chatbot to assist passengers and educate them about their rights. Our system breaks a complex user input into simple queries which are used to retrieve information from a collection of documents detailing air travel regulations. The most relevant passages from these documents are presented along with links to the original documents and the generated queries, enabling users to dissect and leverage the information for their unique circumstances. The system successfully overcomes two predominant challenges: understanding complex user inputs, and delivering accurate answers, free of hallucinations, that passengers can rely on for making informed decisions. A user study comparing the chatbot to a Google search demonstrated the chatbot's usefulness and ease of use. Beyond the primary goal of providing accurate and timely information to air passengers regarding their rights, we hope that this system will also enable further research exploring the tradeoff between the user-friendly conversational interface of chatbots and the accuracy of retrieval systems.
Authors: Nikola Ljube\v{s}i\'c, Taja Kuzman
Abstract: This paper presents a collection of highly comparable web corpora of Slovenian, Croatian, Bosnian, Montenegrin, Serbian, Macedonian, and Bulgarian, covering thereby the whole spectrum of official languages in the South Slavic language space. The collection of these corpora comprises a total of 13 billion tokens of texts from 26 million documents. The comparability of the corpora is ensured by a comparable crawling setup and the usage of identical crawling and post-processing technology. All the corpora were linguistically annotated with the state-of-the-art CLASSLA-Stanza linguistic processing pipeline, and enriched with document-level genre information via the Transformer-based multilingual X-GENRE classifier, which further enhances comparability at the level of linguistic annotation and metadata enrichment. The genre-focused analysis of the resulting corpora shows a rather consistent distribution of genres throughout the seven corpora, with variations in the most prominent genre categories being well-explained by the economic strength of each language community. A comparison of the distribution of genre categories across the corpora indicates that web corpora from less developed countries primarily consist of news articles. Conversely, web corpora from economically more developed countries exhibit a smaller proportion of news content, with a greater presence of promotional and opinionated texts.
Authors: Zhenyu Wu, Chao Shen, Meng Jiang
Abstract: Math word problem (MWP) solving requires generating a reasoning path based on a given problem description that often contains irrelevant conditions. Existing chain-of-thought (CoT) prompting methods elicited multi-step reasoning abilities of large language models (LLMs) to solve MWPs. However, they were seriously confused by the irrelevant conditions, resulting in low accuracy. In this paper, we propose a novel approach named I$^3$C that instructs LLMs to identify and ignore irrelevant conditions. It identifies a set of irrelevant condition candidates that have a weak semantic relevance with the question. Then it prompts LLMs to verify the irrelevant conditions. Lastly it instructs the LLMs with the verification on relevant and irrelevant conditions to avoid confusion and improve reasoning paths. Moreover, we propose to select (problem, reasoning paths) pairs as demonstrations to enhance I$^3$C with few-shot reasoning. We develop I$^3$C-Select that selects the most confusing problems based on the semantic relevance measurement. We conduct extensive experiments on eight MWP datasets. I$^3$C can be combined with any CoT prompting methods to improve the performance of solving MWPs. Notably, with GPT-3.5-Turbo and I$^3$C-Select, we achieve an accuracy of 96.0 and 94.1 on GSM-IC2-1K and GSM-ICM-1K, respectively, significantly outperforming the state-of-the-art few-shot prompting method Complex-CoT by +11.7 and +11.1. Our implementation is made publicly available at https://wzy6642.github.io/I3C.github.io/.
Authors: Siyao Peng, Zihang Sun, Huangyan Shan, Marie Kolm, Verena Blaschke, Ekaterina Artemova, Barbara Plank
Abstract: Named Entity Recognition (NER) is a fundamental task to extract key information from texts, but annotated resources are scarce for dialects. This paper introduces the first dialectal NER dataset for German, BarNER, with 161K tokens annotated on Bavarian Wikipedia articles (bar-wiki) and tweets (bar-tweet), using a schema adapted from German CoNLL 2006 and GermEval. The Bavarian dialect differs from standard German in lexical distribution, syntactic construction, and entity information. We conduct in-domain, cross-domain, sequential, and joint experiments on two Bavarian and three German corpora and present the first comprehensive NER results on Bavarian. Incorporating knowledge from the larger German NER (sub-)datasets notably improves on bar-wiki and moderately on bar-tweet. Inversely, training first on Bavarian contributes slightly to the seminal German CoNLL 2006 corpus. Moreover, with gold dialect labels on Bavarian tweets, we assess multi-task learning between five NER and two Bavarian-German dialect identification tasks and achieve NER SOTA on bar-wiki. We substantiate the necessity of our low-resource BarNER corpus and the importance of diversity in dialects, genres, and topics in enhancing model performance.
Authors: Cunxiang Wang, Ruoxi Ning, Boqi Pan, Tonghui Wu, Qipeng Guo, Cheng Deng, Guangsheng Bao, Qian Wang, Yue Zhang
Abstract: The rapid advancement of Large Language Models (LLMs) has introduced a new frontier in natural language processing, particularly in understanding and processing long-context information. However, the evaluation of these models' long-context abilities remains a challenge due to the limitations of current benchmarks. To address this gap, we introduce NovelQA, a benchmark specifically designed to test the capabilities of LLMs with extended texts. Constructed from English novels, NovelQA offers a unique blend of complexity, length, and narrative coherence, making it an ideal tool for assessing deep textual understanding in LLMs. This paper presents the design and construction of NovelQA, highlighting its manual annotation, and diverse question types. Our evaluation of Long-context LLMs on NovelQA reveals significant insights into the models' performance, particularly emphasizing the challenges they face with multi-hop reasoning, detail-oriented questions, and extremely long input with more than 100,000 tokens. The results underscore the necessity for further advancements in LLMs to improve their long-context comprehension and computational literary studies.
Authors: Jiuhai Chen, Jonas Mueller
Abstract: Large Language Models have become the de facto approach to sequence-to-sequence text generation tasks, but for specialized tasks/domains, a pretrained LLM lacks specific capabilities to produce accurate or well-formatted responses. Supervised fine-tuning specializes a LLM by training it on dataset of example prompts with target responses, but real-world data tends to be noisy. While many fine-tuning algorithms exist, here we consider a \emph{data-centric AI} perspective on LLM fine-tuning, studying how to \emph{systematically} curate the training dataset to improve the LLM produced via \emph{any} fine-tuning algorithm. We introduce an automated data curation pipeline CLEAR (Confidence-based LLM Evaluation And Rectification) for instruction tuning datasets, that can be used with any LLM and fine-tuning procedure. CLEAR estimates which training data is low-quality and either filters or corrects it. Automatically identifying which data to filter or correct is done via LLM-derived confidence estimates, to ensure only confident modifications to the dataset. Unlike existing data curation techniques, CLEAR is a comprehensive framework that can improve a dataset (and trained model outputs) without additional fine-tuning computations. We don't assume access to a stronger LLM than the model being fine-tuned (e.g.\ relying on GPT-4 when fine-tuning GPT-3.5), to see whether CLEAR can meaningfully improve the capabilities of any LLM. Experiments reveal that CLEAR consistently improves the performance of fine-tuned models across many datasets and models (like GPT-3.5 and Llama2).
Authors: Mirza Alim Mutasodirin, Radityo Eko Prasojo
Abstract: The parallelism of Transformer-based models comes at the cost of their input max-length. Some studies proposed methods to overcome this limitation, but none of them reported the effectiveness of summarization as an alternative. In this study, we investigate the performance of document truncation and summarization in text classification tasks. Each of the two was investigated with several variations. This study also investigated how close their performances are to the performance of full-text. We used a dataset of summarization tasks based on Indonesian news articles (IndoSum) to do classification tests. This study shows how the summaries outperform the majority of truncation method variations and lose to only one. The best strategy obtained in this study is taking the head of the document. The second is extractive summarization. This study explains what happened to the result, leading to further research in order to exploit the potential of document summarization as a shortening alternative. The code and data used in this work are publicly available in https://github.com/mirzaalimm/TruncationVsSummarization.
URLs: https://github.com/mirzaalimm/TruncationVsSummarization.
Authors: Zhixue Zhao, Nikolaos Aletras
Abstract: In many real natural language processing application scenarios, practitioners not only aim to maximize predictive performance but also seek faithful explanations for the model predictions. Rationales and importance distribution given by feature attribution methods (FAs) provide insights into how different parts of the input contribute to a prediction. Previous studies have explored how different factors affect faithfulness, mainly in the context of monolingual English models. On the other hand, the differences in FA faithfulness between multilingual and monolingual models have yet to be explored. Our extensive experiments, covering five languages and five popular FAs, show that FA faithfulness varies between multilingual and monolingual models. We find that the larger the multilingual model, the less faithful the FAs are compared to its counterpart monolingual models.Our further analysis shows that the faithfulness disparity is potentially driven by the differences between model tokenizers. Our code is available: https://github.com/casszhao/multilingual-faith.
Authors: Minsu Kim, James Thorne
Abstract: This paper investigates the inherent knowledge in language models from the perspective of epistemological holism. The purpose of this paper is to explore whether LLMs exhibit characteristics consistent with epistemological holism. These characteristics suggest that core knowledge, such as general scientific knowledge, each plays a specific role, serving as the foundation of our knowledge system and being difficult to revise. To assess these traits related to holism, we created a scientific reasoning dataset and examined the epistemology of language models through three tasks: Abduction, Revision, and Argument Generation. In the abduction task, the language models explained situations while avoiding revising the core knowledge. However, in other tasks, the language models were revealed not to distinguish between core and peripheral knowledge, showing an incomplete alignment with holistic knowledge principles.
Authors: Zehui Chen, Kuikun Liu, Qiuchen Wang, Wenwei Zhang, Jiangning Liu, Dahua Lin, Kai Chen, Feng Zhao
Abstract: Open-sourced Large Language Models (LLMs) have achieved great success in various NLP tasks, however, they are still far inferior to API-based models when acting as agents. How to integrate agent ability into general LLMs becomes a crucial and urgent problem. This paper first delivers three key observations: (1) the current agent training corpus is entangled with both formats following and agent reasoning, which significantly shifts from the distribution of its pre-training data; (2) LLMs exhibit different learning speeds on the capabilities required by agent tasks; and (3) current approaches have side-effects when improving agent abilities by introducing hallucinations. Based on the above findings, we propose Agent-FLAN to effectively Fine-tune LANguage models for Agents. Through careful decomposition and redesign of the training corpus, Agent-FLAN enables Llama2-7B to outperform prior best works by 3.5\% across various agent evaluation datasets. With comprehensively constructed negative samples, Agent-FLAN greatly alleviates the hallucination issues based on our established evaluation benchmark. Besides, it consistently improves the agent capability of LLMs when scaling model sizes while slightly enhancing the general capability of LLMs. The code will be available at https://github.com/InternLM/Agent-FLAN.
Authors: Sai Ashish Somayajula, Youwei Liang, Abhishek Singh, Li Zhang, Pengtao Xie
Abstract: Pretrained Language Models (PLMs) have advanced Natural Language Processing (NLP) tasks significantly, but finetuning PLMs on low-resource datasets poses significant challenges such as instability and overfitting. Previous methods tackle these issues by finetuning a strategically chosen subnetwork on a downstream task, while keeping the remaining weights fixed to the pretrained weights. However, they rely on a suboptimal criteria for sub-network selection, leading to suboptimal solutions. To address these limitations, we propose a regularization method based on attention-guided weight mixup for finetuning PLMs. Our approach represents each network weight as a mixup of task-specific weight and pretrained weight, controlled by a learnable attention parameter, providing finer control over sub-network selection. Furthermore, we employ a bi-level optimization (BLO) based framework on two separate splits of the training dataset, improving generalization and combating overfitting. We validate the efficacy of our proposed method through extensive experiments, demonstrating its superiority over previous methods, particularly in the context of finetuning PLMs on low-resource datasets.
Authors: Grant Buster, Pavlo Pinchuk, Jacob Barrons, Ryan McKeever, Aaron Levine, Anthony Lopez
Abstract: The recent growth in renewable energy development in the United States has been accompanied by a simultaneous surge in renewable energy siting ordinances. These zoning laws play a critical role in dictating the placement of wind and solar resources that are critical for achieving low-carbon energy futures. In this context, efficient access to and management of siting ordinance data becomes imperative. The National Renewable Energy Laboratory (NREL) recently introduced a public wind and solar siting database to fill this need. This paper presents a method for harnessing Large Language Models (LLMs) to automate the extraction of these siting ordinances from legal documents, enabling this database to maintain accurate up-to-date information in the rapidly changing energy policy landscape. A novel contribution of this research is the integration of a decision tree framework with LLMs. Our results show that this approach is 85 to 90% accurate with outputs that can be used directly in downstream quantitative modeling. We discuss opportunities to use this work to support similar large-scale policy research in the energy sector. By unlocking new efficiencies in the extraction and analysis of legal documents using LLMs, this study enables a path forward for automated large-scale energy policy research.
Authors: Joana Ribeiro de Faria, Huiyuan Xie, Felix Steffek
Abstract: Court transcripts and judgments are rich repositories of legal knowledge, detailing the intricacies of cases and the rationale behind judicial decisions. The extraction of key information from these documents provides a concise overview of a case, crucial for both legal experts and the public. With the advent of large language models (LLMs), automatic information extraction has become increasingly feasible and efficient. This paper presents a comprehensive study on the application of GPT-4, a large language model, for automatic information extraction from UK Employment Tribunal (UKET) cases. We meticulously evaluated GPT-4's performance in extracting critical information with a manual verification process to ensure the accuracy and relevance of the extracted data. Our research is structured around two primary extraction tasks: the first involves a general extraction of eight key aspects that hold significance for both legal specialists and the general public, including the facts of the case, the claims made, references to legal statutes, references to precedents, general case outcomes and corresponding labels, detailed order and remedies and reasons for the decision. The second task is more focused, aimed at analysing three of those extracted features, namely facts, claims and outcomes, in order to facilitate the development of a tool capable of predicting the outcome of employment law disputes. Through our analysis, we demonstrate that LLMs like GPT-4 can obtain high accuracy in legal information extraction, highlighting the potential of LLMs in revolutionising the way legal information is processed and utilised, offering significant implications for legal research and practice.
Authors: Jeffrey Cheng, Marc Marone, Orion Weller, Dawn Lawrie, Daniel Khashabi, Benjamin Van Durme
Abstract: Released Large Language Models (LLMs) are often paired with a claimed knowledge cutoff date, or the dates at which training data was gathered. Such information is crucial for applications where the LLM must provide up to date information. However, this statement only scratches the surface: do all resources in the training data share the same knowledge cutoff date? Does the model's demonstrated knowledge for these subsets closely align to their cutoff dates? In this work, we define the notion of an effective cutoff. This is distinct from the LLM designer reported cutoff and applies separately to sub-resources and topics. We propose a simple approach to estimate effective cutoffs on the resource-level temporal alignment of an LLM by probing across versions of the data. Using this analysis, we find that effective cutoffs often differ from reported cutoffs. To understand the root cause of this observation, we conduct a direct large-scale analysis on open pre-training datasets. Our analysis reveals two reasons for these inconsistencies: (1) temporal biases of CommonCrawl data due to non-trivial amounts of old data in new dumps and (2) complications in LLM deduplication schemes involving semantic duplicates and lexical near-duplicates. Overall, our results show that knowledge cutoffs are not as simple as they have seemed and that care must be taken both by LLM dataset curators as well as practitioners who seek to use information from these models.
Authors: Zhuoshi Pan, Qianhui Wu, Huiqiang Jiang, Menglin Xia, Xufang Luo, Jue Zhang, Qingwei Lin, Victor R\"uhle, Yuqing Yang, Chin-Yew Lin, H. Vicky Zhao, Lili Qiu, Dongmei Zhang
Abstract: This paper focuses on task-agnostic prompt compression for better generalizability and efficiency. Considering the redundancy in natural language, existing approaches compress prompts by removing tokens or lexical units according to their information entropy obtained from a causal language model such as LLaMa-7B. The challenge is that information entropy may be a suboptimal compression metric: (i) it only leverages unidirectional context and may fail to capture all essential information needed for prompt compression; (ii) it is not aligned with the prompt compression objective. To address these issues, we propose a data distillation procedure to derive knowledge from an LLM to compress prompts without losing crucial information, and meantime, introduce an extractive text compression dataset. We formulate prompt compression as a token classification problem to guarantee the faithfulness of the compressed prompt to the original one, and use a Transformer encoder as the base architecture to capture all essential information for prompt compression from the full bidirectional context. Our approach leads to lower latency by explicitly learning the compression objective with smaller models such as XLM-RoBERTa-large and mBERT. We evaluate our method on both in-domain and out-of-domain datasets, including MeetingBank, LongBench, ZeroScrolls, GSM8K, and BBH. Despite its small size, our model shows significant performance gains over strong baselines and demonstrates robust generalization ability across different LLMs. Additionally, our model is 3x-6x faster than existing prompt compression methods, while accelerating the end-to-end latency by 1.6x-2.9x with compression ratios of 2x-5x.
Authors: Grazia Sveva Ascione, Valerio Sterzi
Abstract: The problem of disambiguation of company names poses a significant challenge in extracting useful information from patents. This issue biases research outcomes as it mostly underestimates the number of patents attributed to companies, particularly multinational corporations which file patents under a plethora of names, including alternate spellings of the same entity and, eventually, companies' subsidiaries. To date, addressing these challenges has relied on labor-intensive dictionary based or string matching approaches, leaving the problem of patents' assignee harmonization on large datasets mostly unresolved. To bridge this gap, this paper describes the Terrorizer algorithm, a text-based algorithm that leverages natural language processing (NLP), network theory, and rule-based techniques to harmonize the variants of company names recorded as patent assignees. In particular, the algorithm follows the tripartite structure of its antecedents, namely parsing, matching and filtering stage, adding an original "knowledge augmentation" phase which is used to enrich the information available on each assignee name. We use Terrorizer on a set of 325'917 companies' names who are assignees of patents granted by the USPTO from 2005 to 2022. The performance of Terrorizer is evaluated on four gold standard datasets. This validation step shows us two main things: the first is that the performance of Terrorizer is similar over different kind of datasets, proving that our algorithm generalizes well. Second, when comparing its performance with the one of the algorithm currently used in PatentsView for the same task (Monath et al., 2021), it achieves a higher F1 score. Finally, we use the Tree-structured Parzen Estimator (TPE) optimization algorithm for the hyperparameters' tuning. Our final result is a reduction in the initial set of names of over 42%.
Authors: Keshav Ramani, Daniel Borrajo
Abstract: Addresses occupy a niche location within the landscape of textual data, due to the positional importance carried by every word, and the geographical scope it refers to. The task of matching addresses happens everyday and is present in various fields like mail redirection, entity resolution, etc. Our work defines, and formalizes a framework to generate matching and mismatching pairs of addresses in the English language, and use it to evaluate various methods to automatically perform address matching. These methods vary widely from distance based approaches to deep learning models. By studying the Precision, Recall and Accuracy metrics of these approaches, we obtain an understanding of the best suited method for this setting of the address matching task.
Authors: Abdelrahman "Boda" Sadallah, Daria Kotova, Ekaterina Kochmar
Abstract: Cryptic crosswords are puzzles that rely not only on general knowledge but also on the solver's ability to manipulate language on different levels and deal with various types of wordplay. Previous research suggests that solving such puzzles is a challenge even for modern NLP models. However, the abilities of large language models (LLMs) have not yet been tested on this task. In this paper, we establish the benchmark results for three popular LLMs -- LLaMA2, Mistral, and ChatGPT -- showing that their performance on this task is still far from that of humans.
Authors: Filippos Gouidis, Katerina Papantoniou, Konstantinos Papoutsakis Theodore Patkos, Antonis Argyros, Dimitris Plexousakis
Abstract: Domain-specific knowledge can significantly contribute to addressing a wide variety of vision tasks. However, the generation of such knowledge entails considerable human labor and time costs. This study investigates the potential of Large Language Models (LLMs) in generating and providing domain-specific information through semantic embeddings. To achieve this, an LLM is integrated into a pipeline that utilizes Knowledge Graphs and pre-trained semantic vectors in the context of the Vision-based Zero-shot Object State Classification task. We thoroughly examine the behavior of the LLM through an extensive ablation study. Our findings reveal that the integration of LLM-based embeddings, in combination with general-purpose pre-trained embeddings, leads to substantial performance improvements. Drawing insights from this ablation study, we conduct a comparative analysis against competing models, thereby highlighting the state-of-the-art performance achieved by the proposed approach.
Authors: Cristiano Mesquita Garcia, Alessandro Lameiras Koerich, Alceu de Souza Britto Jr, Jean Paul Barddal
Abstract: Systems and individuals produce data continuously. On the Internet, people share their knowledge, sentiments, and opinions, provide reviews about services and products, and so on. Automatically learning from these textual data can provide insights to organizations and institutions, thus preventing financial impacts, for example. To learn from textual data over time, the machine learning system must account for concept drift. Concept drift is a frequent phenomenon in real-world datasets and corresponds to changes in data distribution over time. For instance, a concept drift occurs when sentiments change or a word's meaning is adjusted over time. Although concept drift is frequent in real-world applications, benchmark datasets with labeled drifts are rare in the literature. To bridge this gap, this paper provides four textual drift generation methods to ease the production of datasets with labeled drifts. These methods were applied to Yelp and Airbnb datasets and tested using incremental classifiers respecting the stream mining paradigm to evaluate their ability to recover from the drifts. Results show that all methods have their performance degraded right after the drifts, and the incremental SVM is the fastest to run and recover the previous performance levels regarding accuracy and Macro F1-Score.
Authors: Chong Ma, Hanqi Jiang, Wenting Chen, Zihao Wu, Xiaowei Yu, Fang Zeng, Lei Guo, Dajiang Zhu, Tuo Zhang, Dinggang Shen, Tianming Liu, Xiang Li
Abstract: In multi-modal frameworks, the alignment of cross-modal features presents a significant challenge. The predominant approach in multi-modal pre-training emphasizes either global or local alignment between modalities, utilizing extensive datasets. This bottom-up driven method often suffers from a lack of interpretability, a critical concern in radiology. Previous studies have integrated high-level labels in medical images or text, but these still rely on manual annotation, a costly and labor-intensive process. Our work introduces a novel approach by using eye-gaze data, collected synchronously by radiologists during diagnostic evaluations. This data, indicating radiologists' focus areas, naturally links chest X-rays to diagnostic texts. We propose the Eye-gaze Guided Multi-modal Alignment (EGMA) framework to harness eye-gaze data for better alignment of image and text features, aiming to reduce reliance on manual annotations and thus cut training costs. Our model demonstrates robust performance, outperforming other state-of-the-art methods in zero-shot classification and retrieval tasks. The incorporation of easily-obtained eye-gaze data during routine radiological diagnoses signifies a step towards minimizing manual annotation dependency. Additionally, we explore the impact of varying amounts of eye-gaze data on model performance, highlighting the feasibility and utility of integrating this auxiliary data into multi-modal pre-training.
Authors: Xudong Guo, Kaixuan Huang, Jiale Liu, Wenhui Fan, Natalia V\'elez, Qingyun Wu, Huazheng Wang, Thomas L. Griffiths, Mengdi Wang
Abstract: Large Language Models (LLMs) have emerged as integral tools for reasoning, planning, and decision-making, drawing upon their extensive world knowledge and proficiency in language-related tasks. LLMs thus hold tremendous potential for natural language interaction within multi-agent systems to foster cooperation. However, LLM agents tend to over-report and comply with any instruction, which may result in information redundancy and confusion in multi-agent cooperation. Inspired by human organizations, this paper introduces a framework that imposes prompt-based organization structures on LLM agents to mitigate these problems. Through a series of experiments with embodied LLM agents and human-agent collaboration, our results highlight the impact of designated leadership on team efficiency, shedding light on the leadership qualities displayed by LLM agents and their spontaneous cooperative behaviors. Further, we harness the potential of LLMs to propose enhanced organizational prompts, via a Criticize-Reflect process, resulting in novel organization structures that reduce communication costs and enhance team efficiency.
Authors: Pierre Dognin, Jesus Rios, Ronny Luss, Inkit Padhi, Matthew D Riemer, Miao Liu, Prasanna Sattigeri, Manish Nagireddy, Kush R. Varshney, Djallel Bouneffouf
Abstract: Developing value-aligned AI agents is a complex undertaking and an ongoing challenge in the field of AI. Specifically within the domain of Large Language Models (LLMs), the capability to consolidate multiple independently trained dialogue agents, each aligned with a distinct moral value, into a unified system that can adapt to and be aligned with multiple moral values is of paramount importance. In this paper, we propose a system that does contextual moral value alignment based on contextual aggregation. Here, aggregation is defined as the process of integrating a subset of LLM responses that are best suited to respond to a user input, taking into account features extracted from the user's input. The proposed system shows better results in term of alignment to human value compared to the state of the art.
Authors: Baolin Li, Yankai Jiang, Vijay Gadepally, Devesh Tiwari
Abstract: The rapid advancement of Generative Artificial Intelligence (GenAI) across diverse sectors raises significant environmental concerns, notably the carbon emissions from their cloud and high performance computing (HPC) infrastructure. This paper presents Sprout, an innovative framework designed to address these concerns by reducing the carbon footprint of generative Large Language Model (LLM) inference services. Sprout leverages the innovative concept of "generation directives" to guide the autoregressive generation process, thereby enhancing carbon efficiency. Our proposed method meticulously balances the need for ecological sustainability with the demand for high-quality generation outcomes. Employing a directive optimizer for the strategic assignment of generation directives to user prompts and an original offline quality evaluator, Sprout demonstrates a significant reduction in carbon emissions by over 40% in real-world evaluations using the Llama2 LLM and global electricity grid data. This research marks a critical step toward aligning AI technology with sustainable practices, highlighting the potential for mitigating environmental impacts in the rapidly expanding domain of generative artificial intelligence.
Authors: Ce Zhang, Simon Stepputtis, Katia Sycara, Yaqi Xie
Abstract: Recently, large-scale pre-trained Vision-Language Models (VLMs) have demonstrated great potential in learning open-world visual representations, and exhibit remarkable performance across a wide range of downstream tasks through efficient fine-tuning. In this work, we innovatively introduce the concept of dual learning into fine-tuning VLMs, i.e., we not only learn what an image is, but also what an image isn't. Building on this concept, we introduce a novel DualAdapter approach to enable dual-path adaptation of VLMs from both positive and negative perspectives with only limited annotated samples. In the inference stage, our DualAdapter performs unified predictions by simultaneously conducting complementary positive selection and negative exclusion across target classes, thereby enhancing the overall recognition accuracy of VLMs in downstream tasks. Our extensive experimental results across 15 datasets validate that the proposed DualAdapter outperforms existing state-of-the-art methods on both few-shot learning and domain generalization tasks while achieving competitive computational efficiency. Code is available at https://github.com/zhangce01/DualAdapter.
Authors: Rossi Kamal, Zuzana Kubincova
Abstract: Education is a right of all, however, every individual is different than others. Teachers in post-communism era discover inherent individualism to equally train all towards job market of fourth industrial revolution. We can consider scenario of ethnic minority education in academic practices. Ethnic minority group has grown in their own culture and would prefer to be taught in their native way. We have formulated such linguistic anthropology(how people learn)based engagement as semi-supervised problem. Then, we have developed an conditional deep generative adversarial network algorithm namely LA-GAN to classify linguistic ethnographic features in student engagement. Theoretical justification proves the objective, regularization and loss function of our semi-supervised adversarial model. Survey questions are prepared to reach some form of assumptions about z-generation and ethnic minority group, whose learning style, learning approach and preference are our main area of interest.
Authors: Songning Lai, Jiakang Li, Guinan Guo, Xifeng Hu, Yulong Li, Yuan Tan, Zichen Song, Yutong Liu, Zhaoxia Ren, Chun Wan, Danmin Miao, Zhi Liu
Abstract: Designing an effective representation learning method for multimodal sentiment analysis tasks is a crucial research direction. The challenge lies in learning both shared and private information in a complete modal representation, which is difficult with uniform multimodal labels and a raw feature fusion approach. In this work, we propose a deep modal shared information learning module based on the covariance matrix to capture the shared information between modalities. Additionally, we use a label generation module based on a self-supervised learning strategy to capture the private information of the modalities. Our module is plug-and-play in multimodal tasks, and by changing the parameterization, it can adjust the information exchange relationship between the modes and learn the private or shared information between the specified modes. We also employ a multi-task learning strategy to help the model focus its attention on the modal differentiation training data. We provide a detailed formulation derivation and feasibility proof for the design of the deep modal shared information learning module. We conduct extensive experiments on three common multimodal sentiment analysis baseline datasets, and the experimental results validate the reliability of our model. Furthermore, we explore more combinatorial techniques for the use of the module. Our approach outperforms current state-of-the-art methods on most of the metrics of the three public datasets.
Authors: Xingyao Wang, Hao Peng, Reyhaneh Jabbarvand, Heng Ji
Abstract: Fine-tuning pre-trained language models (LMs) is essential for enhancing their capabilities. Existing techniques commonly fine-tune on input-output pairs (e.g., instruction tuning) or with numerical rewards that gauge the output quality (e.g., RLHF). We explore LMs' potential to learn from textual interactions (LETI) that not only check their correctness with binary labels but also pinpoint and explain errors in their outputs through textual feedback. Our focus is the code generation task, where the model produces code based on natural language instructions. This setting invites a natural and scalable way to acquire textual feedback: the error messages and stack traces from code execution using a Python interpreter. LETI iteratively fine-tunes the model, using the LM objective, on a concatenation of natural language instructions, LM-generated programs, and textual feedback. Prepended to this fine-tuning text, a binary reward token is used to differentiate correct and buggy solutions. LETI requires no ground-truth outputs for training and even outperforms a fine-tuned baseline that does. LETI not only improves the performance of LMs on a code generation dataset MBPP, but also generalizes to other datasets. Trained on MBPP, it achieves comparable or better performance than the base LMs on unseen problems in HumanEval. Furthermore, compared to binary feedback, we observe that textual feedback leads to improved generation quality and sample efficiency, achieving the same performance with fewer than half of the gradient steps. LETI is equally applicable in natural language tasks when they can be formulated as code generation, which we empirically verified on event argument extraction.
Authors: Jaehun Jung, Peter West, Liwei Jiang, Faeze Brahman, Ximing Lu, Jillian Fisher, Taylor Sorensen, Yejin Choi
Abstract: We present Impossible Distillation, a novel framework for paraphrasing and sentence summarization, that distills a high-quality dataset and model from a low-quality teacher that itself cannot perform these tasks. Unlike prior works that rely on an extreme-scale teacher model (e.g., GPT3) or task-specific architecture, we hypothesize and verify the paraphrastic proximity intrinsic to pre-trained LMs (e.g., GPT2), where paraphrases occupy a proximal subspace in the LM distribution. By identifying and distilling generations from these subspaces, Impossible Distillation produces a high-quality dataset and model even from GPT2-scale LMs. We evaluate our method on multiple benchmarks spanning unconstrained / syntax-controlled paraphrase generation and sentence summarization. Our model with 770M parameters consistently outperforms strong baselines, including models distilled from ChatGPT, and sometimes, even ChatGPT itself. Also, we find that our distilled dataset from 1.5B LMs exhibits higher diversity and fidelity than up to 13 times larger datasets.
Authors: Jinan Zou, Maihao Guo, Yu Tian, Yuhao Lin, Haiyao Cao, Lingqiao Liu, Ehsan Abbasnejad, Javen Qinfeng Shi
Abstract: Identifying unexpected domain-shifted instances in natural language processing is crucial in real-world applications. Previous works identify the out-of-distribution (OOD) instance by leveraging a single global feature embedding to represent the sentence, which cannot characterize subtle OOD patterns well. Another major challenge current OOD methods face is learning effective low-dimensional sentence representations to identify the hard OOD instances that are semantically similar to the in-distribution (ID) data. In this paper, we propose a new unsupervised OOD detection method, namely Semantic Role Labeling Guided Out-of-distribution Detection (SRLOOD), that separates, extracts, and learns the semantic role labeling (SRL) guided fine-grained local feature representations from different arguments of a sentence and the global feature representations of the full sentence using a margin-based contrastive loss. A novel self-supervised approach is also introduced to enhance such global-local feature learning by predicting the SRL extracted role. The resulting model achieves SOTA performance on four OOD benchmarks, indicating the effectiveness of our approach. The code is publicly accessible via \url{https://github.com/cytai/SRLOOD}.
Authors: Zhengliang Liu, Aoxiao Zhong, Yiwei Li, Longtao Yang, Chao Ju, Zihao Wu, Chong Ma, Peng Shu, Cheng Chen, Sekeun Kim, Haixing Dai, Lin Zhao, Lichao Sun, Dajiang Zhu, Jun Liu, Wei Liu, Dinggang Shen, Xiang Li, Quanzheng Li, Tianming Liu
Abstract: We introduce Radiology-GPT, a large language model for radiology. Using an instruction tuning approach on an extensive dataset of radiology domain knowledge, Radiology-GPT demonstrates superior performance compared to general language models such as StableLM, Dolly and LLaMA. It exhibits significant versatility in radiological diagnosis, research, and communication. This work serves as a catalyst for future developments in clinical NLP. The successful implementation of Radiology-GPT is indicative of the potential of localizing generative large language models, specifically tailored for distinctive medical specialties, while ensuring adherence to privacy standards such as HIPAA. The prospect of developing individualized, large-scale language models that cater to specific needs of various hospitals presents a promising direction. The fusion of conversational competence and domain-specific knowledge in these models is set to foster future development in healthcare AI. A demo of Radiology-GPT is available at https://huggingface.co/spaces/allen-eric/radiology-gpt.
URLs: https://huggingface.co/spaces/allen-eric/radiology-gpt.
Authors: Senbao Shi, Zhenran Xu, Baotian Hu, Min Zhang
Abstract: Multimodal Entity Linking (MEL) is the task of mapping mentions with multimodal contexts to the referent entities from a knowledge base. Existing MEL methods mainly focus on designing complex multimodal interaction mechanisms and require fine-tuning all model parameters, which can be prohibitively costly and difficult to scale in the era of Large Language Models (LLMs). In this work, we propose GEMEL, a Generative Multimodal Entity Linking framework based on LLMs, which directly generates target entity names. We keep the vision and language model frozen and only train a feature mapper to enable cross-modality interactions. To adapt LLMs to the MEL task, we leverage the in-context learning capability of LLMs by retrieving multimodal instances as demonstrations. Extensive experiments show that, with only ~0.3% of the model parameters fine-tuned, GEMEL achieves state-of-the-art results on two well-established MEL datasets (7.7% accuracy gains on WikiDiverse and 8.8% accuracy gains on WikiMEL). The performance gain stems from mitigating the popularity bias of LLM predictions and disambiguating less common entities effectively. Further analysis verifies the generality and scalability of GEMEL. Our framework is compatible with any off-the-shelf language model, paving the way towards an efficient and general solution for utilizing LLMs in the MEL task.
Authors: Yueen Ma, Dafeng Chi, Jingjing Li, Kai Song, Yuzheng Zhuang, Irwin King
Abstract: The natural language generation domain has witnessed great success thanks to Transformer models. Although they have achieved state-of-the-art generative quality, they often neglect generative diversity. Prior attempts to tackle this issue suffer from either low model capacity or over-complicated architectures. Some recent methods employ the VAE framework to enhance diversity, but their latent variables fully depend on the input context, restricting exploration of the latent space. In this paper, we introduce VOLTA, a framework that elevates generative diversity by bridging Transformer with VAE via a more effective cross-attention-based connection, departing from conventional embedding concatenation or summation. Additionally, we propose integrating InfoGAN-style latent codes to enable input-independent variability, further diversifying the generation. Moreover, our framework accommodates discrete inputs alongside its existing support for continuous inputs. We perform comprehensive experiments with two types of Transformers on six datasets from three different NLG tasks to show that our approach can significantly improve generative diversity while maintaining generative quality.
Authors: Peng Wang, Ningyu Zhang, Bozhong Tian, Zekun Xi, Yunzhi Yao, Ziwen Xu, Mengru Wang, Shengyu Mao, Xiaohan Wang, Siyuan Cheng, Kangwei Liu, Yuansheng Ni, Guozhou Zheng, Huajun Chen
Abstract: Large Language Models (LLMs) usually suffer from knowledge cutoff or fallacy issues, which means they are unaware of unseen events or generate text with incorrect facts owing to outdated/noisy data. To this end, many knowledge editing approaches for LLMs have emerged -- aiming to subtly inject/edit updated knowledge or adjust undesired behavior while minimizing the impact on unrelated inputs. Nevertheless, due to significant differences among various knowledge editing methods and the variations in task setups, there is no standard implementation framework available for the community, which hinders practitioners from applying knowledge editing to applications. To address these issues, we propose EasyEdit, an easy-to-use knowledge editing framework for LLMs. It supports various cutting-edge knowledge editing approaches and can be readily applied to many well-known LLMs such as T5, GPT-J, LlaMA, etc. Empirically, we report the knowledge editing results on LlaMA-2 with EasyEdit, demonstrating that knowledge editing surpasses traditional fine-tuning in terms of reliability and generalization. We have released the source code on GitHub, along with Google Colab tutorials and comprehensive documentation for beginners to get started. Besides, we present an online system for real-time knowledge editing, and a demo video.
Authors: Federico Bianchi, Mirac Suzgun, Giuseppe Attanasio, Paul R\"ottger, Dan Jurafsky, Tatsunori Hashimoto, James Zou
Abstract: Training large language models to follow instructions makes them perform better on a wide range of tasks and generally become more helpful. However, a perfectly helpful model will follow even the most malicious instructions and readily generate harmful content. In this paper, we raise concerns over the safety of models that only emphasize helpfulness, not harmlessness, in their instruction-tuning. We show that several popular instruction-tuned models are highly unsafe. Moreover, we show that adding just 3% safety examples (a few hundred demonstrations) when fine-tuning a model like LLaMA can substantially improve its safety. Our safety-tuning does not make models significantly less capable or helpful as measured by standard benchmarks. However, we do find exaggerated safety behaviours, where too much safety-tuning makes models refuse perfectly safe prompts if they superficially resemble unsafe ones. As a whole, our results illustrate trade-offs in training LLMs to be helpful and training them to be safe.
Authors: Mohsinul Kabir, Mohammed Saidul Islam, Md Tahmid Rahman Laskar, Mir Tafseer Nayeem, M Saiful Bari, Enamul Hoque
Abstract: Large Language Models (LLMs) have emerged as one of the most important breakthroughs in NLP for their impressive skills in language generation and other language-specific tasks. Though LLMs have been evaluated in various tasks, mostly in English, they have not yet undergone thorough evaluation in under-resourced languages such as Bengali (Bangla). To this end, this paper introduces BenLLM-Eval, which consists of a comprehensive evaluation of LLMs to benchmark their performance in the Bengali language that has modest resources. In this regard, we select various important and diverse Bengali NLP tasks, such as text summarization, question answering, paraphrasing, natural language inference, transliteration, text classification, and sentiment analysis for zero-shot evaluation of popular LLMs, namely, GPT-3.5, LLaMA-2-13b-chat, and Claude-2. Our experimental results demonstrate that while in some Bengali NLP tasks, zero-shot LLMs could achieve performance on par, or even better than current SOTA fine-tuned models; in most tasks, their performance is quite poor (with the performance of open-source LLMs like LLaMA-2-13b-chat being significantly bad) in comparison to the current SOTA results. Therefore, it calls for further efforts to develop a better understanding of LLMs in modest-resourced languages like Bengali.
Authors: Zican Dong, Tianyi Tang, Junyi Li, Wayne Xin Zhao, Ji-Rong Wen
Abstract: Large language models (LLMs) have achieved dramatic proficiency over NLP tasks with normal length. Recently, multiple studies have committed to extending the context length and enhancing the long text modeling capabilities of LLMs. To comprehensively evaluate the long context ability of LLMs, we propose BAMBOO, a multi-task long context benchmark. BAMBOO has been designed with four principles: comprehensive capacity evaluation, avoidance of data contamination, accurate automatic evaluation, and different length levels. It consists of 10 datasets from 5 different long text understanding tasks, i.e. question answering, hallucination detection, text sorting, language modeling, and code completion, to cover core capacities and various domains of LLMs. We conduct experiments with five long context models on BAMBOO and further discuss four key research questions of long text. We also qualitatively analyze current long context models and point out future directions for enhancing long text modeling capacities. We release our data, prompts, and code at https://github.com/RUCAIBox/BAMBOO.
Authors: Tianjian Li, Haoran Xu, Philipp Koehn, Daniel Khashabi, Kenton Murray
Abstract: Text generation models are notoriously vulnerable to errors in the training data. With the wide-spread availability of massive amounts of web-crawled data becoming more commonplace, how can we enhance the robustness of models trained on a massive amount of noisy web-crawled text? In our work, we propose Error Norm Truncation (ENT), a robust enhancement method to the standard training objective that truncates noisy data. Compared to methods that only uses the negative log-likelihood loss to estimate data quality, our method provides a more accurate estimation by considering the distribution of non-target tokens, which is often overlooked by previous work. Through comprehensive experiments across language modeling, machine translation, and text summarization, we show that equipping text generation models with ENT improves generation quality over standard training and previous soft and hard truncation methods. Furthermore, we show that our method improves the robustness of models against two of the most detrimental types of noise in machine translation, resulting in an increase of more than 2 BLEU points over the MLE baseline when up to 50% of noise is added to the data.
Authors: Zishun Yu, Yunzhe Tao, Liyu Chen, Tao Sun, Hongxia Yang
Abstract: Program synthesis aims to create accurate, executable programs from problem specifications, specifically from natural language descriptions in our context. Recent studies have leveraged the power of reinforcement learning (RL) in conjunction with large language models (LLMs), significantly enhancing code generation capabilities. The application of RL focuses on directly optimizing for functional correctness, offering an advantage over conventional supervised methods. Despite policy-based RL methods dominating the literature on RL for program synthesis, the nature of program synthesis tasks hints at a natural alignment with value-based methods. This stems from the rich collection of off-policy programs, including those developed by human programmers and also historical samples, coupled with the straightforward verification of generated programs through automated unit testing, meaning rewards are easy to obtain. Diverging from the dominant use of policy-based algorithms, our work explores the feasibility of value-based approaches, leading to the development of our $\mathcal{B}$-Coder (pronounced Bellman coder). Yet, training value-based methods presents challenges due to the enormous search space inherent to program synthesis. To this end, we introduce an initialization protocol for RL agents utilizing pre-trained LMs and a conservative Bellman operator to reduce training complexities. Moreover, we demonstrate how to leverage the learned value functions as a dual strategy to post-process generated programs. Our empirical evaluations demonstrated $\mathcal{B}$-Coder's capability in achieving state-of-the-art performance when compared to policy-based methods. Remarkably, this achievement is reached with minimal reward engineering effort, highlighting the effectiveness of value-based RL, independent of reward designs.
Authors: Linlu Qiu, Liwei Jiang, Ximing Lu, Melanie Sclar, Valentina Pyatkin, Chandra Bhagavatula, Bailin Wang, Yoon Kim, Yejin Choi, Nouha Dziri, Xiang Ren
Abstract: The ability to derive underlying principles from a handful of observations and then generalize to novel situations -- known as inductive reasoning -- is central to human intelligence. Prior work suggests that language models (LMs) often fall short on inductive reasoning, despite achieving impressive success on research benchmarks. In this work, we conduct a systematic study of the inductive reasoning capabilities of LMs through iterative hypothesis refinement, a technique that more closely mirrors the human inductive process than standard input-output prompting. Iterative hypothesis refinement employs a three-step process: proposing, selecting, and refining hypotheses in the form of textual rules. By examining the intermediate rules, we observe that LMs are phenomenal hypothesis proposers (i.e., generating candidate rules), and when coupled with a (task-specific) symbolic interpreter that is able to systematically filter the proposed set of rules, this hybrid approach achieves strong results across inductive reasoning benchmarks that require inducing causal relations, language-like instructions, and symbolic concepts. However, they also behave as puzzling inductive reasoners, showing notable performance gaps between rule induction (i.e., identifying plausible rules) and rule application (i.e., applying proposed rules to instances), suggesting that LMs are proposing hypotheses without being able to actually apply the rules. Through empirical and human analyses, we further reveal several discrepancies between the inductive reasoning processes of LMs and humans, shedding light on both the potentials and limitations of using LMs in inductive reasoning tasks.
Authors: Sai Koneru, Miriam Exel, Matthias Huck, Jan Niehues
Abstract: Large Language Models (LLM's) have demonstrated considerable success in various Natural Language Processing tasks, but they have yet to attain state-of-the-art performance in Neural Machine Translation (NMT). Nevertheless, their significant performance in tasks demanding a broad understanding and contextual processing shows their potential for translation. To exploit these abilities, we investigate using LLM's for MT and explore recent parameter-efficient fine-tuning techniques. Surprisingly, our initial experiments find that fine-tuning for translation purposes even led to performance degradation. To overcome this, we propose an alternative approach: adapting LLM's as Automatic Post-Editors (APE) rather than direct translators. Building on the LLM's exceptional ability to process and generate lengthy sequences, we also propose extending our approach to document-level translation. We show that leveraging Low-Rank-Adapter fine-tuning for APE can yield significant improvements across both sentence and document-level metrics while generalizing to out-of-domain data. Most notably, we achieve a state-of-the-art accuracy rate of 89\% on the ContraPro test set, which specifically assesses the model's ability to resolve pronoun ambiguities when translating from English to German. Lastly, we investigate a practical scenario involving manual post-editing for document-level translation, where reference context is made available. Here, we demonstrate that leveraging human corrections can significantly reduce the number of edits required for subsequent translations (Interactive Demo for integrating manual feedback can be found here: https://huggingface.co/spaces/skoneru/contextual_refinement_ende).
URLs: https://huggingface.co/spaces/skoneru/contextual_refinement_ende).
Authors: Minbeom Kim, Jahyun Koo, Hwanhee Lee, Joonsuk Park, Hwaran Lee, Kyomin Jung
Abstract: As large language models become increasingly integrated into daily life, detecting implicit toxicity across diverse contexts is crucial. To this end, we introduce LifeTox, a dataset designed for identifying implicit toxicity within a broad range of advice-seeking scenarios. Unlike existing safety datasets, LifeTox comprises diverse contexts derived from personal experiences through open-ended questions. Experiments demonstrate that RoBERTa fine-tuned on LifeTox matches or surpasses the zero-shot performance of large language models in toxicity classification tasks. These results underscore the efficacy of LifeTox in addressing the complex challenges inherent in implicit toxicity. We open-sourced the dataset\footnote{\url{https://huggingface.co/datasets/mbkim/LifeTox}} and the LifeTox moderator family; 350M, 7B, and 13B.
Authors: Zonghai Yao, Ahmed Jaafar, Beining Wang, Zhichao Yang, Hong Yu
Abstract: This study examines the effect of prompt engineering on the performance of Large Language Models (LLMs) in clinical note generation. We introduce an Automatic Prompt Optimization (APO) framework to refine initial prompts and compare the outputs of medical experts, non-medical experts, and APO-enhanced GPT3.5 and GPT4. Results highlight GPT4 APO's superior performance in standardizing prompt quality across clinical note sections. A human-in-the-loop approach shows that experts maintain content quality post-APO, with a preference for their own modifications, suggesting the value of expert customization. We recommend a two-phase optimization process, leveraging APO-GPT4 for consistency and expert input for personalization.
Authors: Anshuman Chhabra, Hadi Askari, Prasant Mohapatra
Abstract: We characterize and study zero-shot abstractive summarization in Large Language Models (LLMs) by measuring position bias, which we propose as a general formulation of the more restrictive lead bias phenomenon studied previously in the literature. Position bias captures the tendency of a model unfairly prioritizing information from certain parts of the input text over others, leading to undesirable behavior. Through numerous experiments on four diverse real-world datasets, we study position bias in multiple LLM models such as GPT 3.5-Turbo, Llama-2, and Dolly-v2, as well as state-of-the-art pretrained encoder-decoder abstractive summarization models such as Pegasus and BART. Our findings lead to novel insights and discussion on performance and position bias of models for zero-shot summarization tasks.
Authors: Zhichen Dong, Zhanhui Zhou, Chao Yang, Jing Shao, Yu Qiao
Abstract: Large Language Models (LLMs) are now commonplace in conversation applications. However, their risks of misuse for generating harmful responses have raised serious societal concerns and spurred recent research on LLM conversation safety. Therefore, in this survey, we provide a comprehensive overview of recent studies, covering three critical aspects of LLM conversation safety: attacks, defenses, and evaluations. Our goal is to provide a structured summary that enhances understanding of LLM conversation safety and encourages further investigation into this important subject. For easy reference, we have categorized all the studies mentioned in this survey according to our taxonomy, available at: https://github.com/niconi19/LLM-conversation-safety.
Authors: Ramon Ferrer-i-Cancho
Abstract: The word order of a sentence is shaped by multiple principles. The principle of syntactic dependency distance minimization is in conflict with the principle of surprisal minimization (or predictability maximization) in single head syntactic dependency structures: while the former predicts that the head should be placed at the center of the linear arrangement, the latter predicts that the head should be placed at one of the ends (either first or last). A critical question is when surprisal minimization (or predictability maximization) should surpass syntactic dependency distance minimization. In the context of single head structures, it has been predicted that this is more likely to happen when two conditions are met, i.e. (a) fewer words are involved and (b) words are shorter. Here we test the prediction on the noun phrase when it is composed of a demonstrative, a numeral, an adjective and a noun. We find that, across preferred orders in languages, the noun tends to be placed at one of the ends, confirming the theoretical prediction. We also show evidence of anti locality effects: syntactic dependency distances in preferred orders are longer than expected by chance.
Authors: Zhiyu Yang, Zihan Zhou, Shuo Wang, Xin Cong, Xu Han, Yukun Yan, Zhenghao Liu, Zhixing Tan, Pengyuan Liu, Dong Yu, Zhiyuan Liu, Xiaodong Shi, Maosong Sun
Abstract: Scientific data visualization plays a crucial role in research by enabling the direct display of complex information and assisting researchers in identifying implicit patterns. Despite its importance, the use of Large Language Models (LLMs) for scientific data visualization remains rather unexplored. In this study, we introduce MatPlotAgent, an efficient model-agnostic LLM agent framework designed to automate scientific data visualization tasks. Leveraging the capabilities of both code LLMs and multi-modal LLMs, MatPlotAgent consists of three core modules: query understanding, code generation with iterative debugging, and a visual feedback mechanism for error correction. To address the lack of benchmarks in this field, we present MatPlotBench, a high-quality benchmark consisting of 100 human-verified test cases. Additionally, we introduce a scoring approach that utilizes GPT-4V for automatic evaluation. Experimental results demonstrate that MatPlotAgent can improve the performance of various LLMs, including both commercial and open-source models. Furthermore, the proposed evaluation method shows a strong correlation with human-annotated scores.
Authors: Jiyoung Lee, Minwoo Kim, Seungho Kim, Junghwan Kim, Seunghyun Won, Hwaran Lee, Edward Choi
Abstract: For Large Language Models (LLMs) to be effectively deployed in a specific country, they must possess an understanding of the nation's culture and basic knowledge. To this end, we introduce National Alignment, which measures an alignment between an LLM and a targeted country from two aspects: social value alignment and common knowledge alignment. Social value alignment evaluates how well the model understands nation-specific social values, while common knowledge alignment examines how well the model captures basic knowledge related to the nation. We constructed KorNAT, the first benchmark that measures national alignment with South Korea. For the social value dataset, we obtained ground truth labels from a large-scale survey involving 6,174 unique Korean participants. For the common knowledge dataset, we constructed samples based on Korean textbooks and GED reference materials. KorNAT contains 4K and 6K multiple-choice questions for social value and common knowledge, respectively. Our dataset creation process is meticulously designed and based on statistical sampling theory and was refined through multiple rounds of human review. The experiment results of seven LLMs reveal that only a few models met our reference score, indicating a potential for further enhancement. KorNAT has received government approval after passing an assessment conducted by a government-affiliated organization dedicated to evaluating dataset quality. Samples and detailed evaluation protocols of our dataset can be found in https://selectstar.ai/ko/papers-national-alignment
Authors: Debrup Das, Debopriyo Banerjee, Somak Aditya, Ashish Kulkarni
Abstract: Tool-augmented Large Language Models (TALM) are known to enhance the skillset of large language models (LLM), thereby, leading to their improved reasoning abilities across many tasks. While, TALMs have been successfully employed in different question-answering benchmarks, their efficacy on complex mathematical reasoning benchmarks, and the potential complimentary benefits offered by tools for knowledge retrieval and mathematical equation solving, are open research questions. In this work, we present MATHSENSEI, a tool-augmented large language model for mathematical reasoning. Augmented with tools for knowledge retrieval (Bing Web Search), program execution (Python), and symbolic equation solving (Wolfram-Alpha), we study the complimentary benefits of these tools through evaluations on mathematical reasoning datasets. We perform exhaustive ablations on MATH,a popular dataset for evaluating mathematical reasoning on diverse mathematical disciplines. We also conduct experiments involving well-known tool planners to study the impact of tool sequencing on the model performance. MATHSENSEI achieves 13.5% better accuracy over gpt-3.5-turbo with chain-of-thought on the MATH dataset. We further observe that TALMs are not as effective for simpler math word problems (in GSM-8k), and the benefit increases as the complexity and required knowledge increases (progressively over AQuA, MMLU-Math, and higher level complex questions in MATH). The code and data are available at https://github.com/Debrup-61/MathSensei.
Authors: Leane Jourdan, Florian Boudin, Nicolas Hernandez, Richard Dufour
Abstract: Writing a scientific article is a challenging task as it is a highly codified and specific genre, consequently proficiency in written communication is essential for effectively conveying research findings and ideas. In this article, we propose an original textual resource on the revision step of the writing process of scientific articles. This new dataset, called CASIMIR, contains the multiple revised versions of 15,646 scientific articles from OpenReview, along with their peer reviews. Pairs of consecutive versions of an article are aligned at sentence-level while keeping paragraph location information as metadata for supporting future revision studies at the discourse level. Each pair of revised sentences is enriched with automatically extracted edits and associated revision intention. To assess the initial quality on the dataset, we conducted a qualitative study of several state-of-the-art text revision approaches and compared various evaluation metrics. Our experiments led us to question the relevance of the current evaluation methods for the text revision task.
Authors: Felix K\"unnecke, Anna Filighera, Colin Leong, Tim Steuer
Abstract: Grading short answer questions automatically with interpretable reasoning behind the grading decision is a challenging goal for current transformer approaches. Justification cue detection, in combination with logical reasoners, has shown a promising direction for neuro-symbolic architectures in ASAG. But, one of the main challenges is the requirement of annotated justification cues in the students' responses, which only exist for a few ASAG datasets. To overcome this challenge, we contribute (1) a weakly supervised annotation procedure for justification cues in ASAG datasets, and (2) a neuro-symbolic model for explainable ASAG based on justification cues. Our approach improves upon the RMSE by 0.24 to 0.3 compared to the state-of-the-art on the Short Answer Feedback dataset in a bilingual, multi-domain, and multi-question training setup. This result shows that our approach provides a promising direction for generating high-quality grades and accompanying explanations for future research in ASAG and educational NLP.
Authors: Fiona Anting Tan, Gerard Christopher Yeo, Fanyou Wu, Weijie Xu, Vinija Jain, Aman Chadha, Kokil Jaidka, Yang Liu, See-Kiong Ng
Abstract: Recent advances in large language models (LLMs) demonstrate that their capabilities are comparable, or even superior, to humans in many tasks in natural language processing. Despite this progress, LLMs are still inadequate at social-cognitive reasoning, which humans are naturally good at. Drawing inspiration from psychological research on the links between certain personality traits and Theory-of-Mind (ToM) reasoning, and from prompt engineering research on the hyper-sensitivity of prompts in affecting LLMs capabilities, this study investigates how inducing personalities in LLMs using prompts affects their ToM reasoning capabilities. Our findings show that certain induced personalities can significantly affect the LLMs' reasoning capabilities in three different ToM tasks. In particular, traits from the Dark Triad have a larger variable effect on LLMs like GPT-3.5, Llama 2, and Mistral across the different ToM tasks. We find that LLMs that exhibit a higher variance across personality prompts in ToM also tends to be more controllable in personality tests: personality traits in LLMs like GPT-3.5, Llama 2 and Mistral can be controllably adjusted through our personality prompts. In today's landscape where role-play is a common strategy when using LLMs, our research highlights the need for caution, as models that adopt specific personas with personalities potentially also alter their reasoning abilities in an unexpected manner.
Authors: Yiding Liu, Jingjing Wang, Jiamin Luo, Tao Zeng, Guodong Zhou
Abstract: Aspect Sentiment Understanding (ASU) in interactive scenarios (e.g., Question-Answering and Dialogue) has attracted ever-more interest in recent years and achieved important progresses. However, existing studies on interactive ASU largely ignore the coreference issue for opinion targets (i.e., aspects), while this phenomenon is ubiquitous in interactive scenarios especially dialogues, limiting the ASU performance. Recently, large language models (LLMs) shows the powerful ability to integrate various NLP tasks with the chat paradigm. In this way, this paper proposes a new Chat-based Aspect Sentiment Understanding (ChatASU) task, aiming to explore LLMs' ability in understanding aspect sentiments in dialogue scenarios. Particularly, this ChatASU task introduces a sub-task, i.e., Aspect Chain Reasoning (ACR) task, to address the aspect coreference issue. On this basis, we propose a Trusted Self-reflexion Approach (TSA) with ChatGLM as backbone to ChatASU. Specifically, this TSA treats the ACR task as an auxiliary task to boost the performance of the primary ASU task, and further integrates trusted learning into reflexion mechanisms to alleviate the LLMs-intrinsic factual hallucination problem in TSA. Furthermore, a high-quality ChatASU dataset is annotated to evaluate TSA, and extensive experiments show that our proposed TSA can significantly outperform several state-of-the-art baselines, justifying the effectiveness of TSA to ChatASU and the importance of considering the coreference and hallucination issues in ChatASU.
Authors: Rui Yang, Haoran Liu, Edison Marrese-Taylor, Qingcheng Zeng, Yu He Ke, Wanxin Li, Lechao Cheng, Qingyu Chen, James Caverlee, Yutaka Matsuo, Irene Li
Abstract: Large Language Models (LLMs) have significantly advanced healthcare innovation on generation capabilities. However, their application in real clinical settings is challenging due to potential deviations from medical facts and inherent biases. In this work, we develop an augmented LLM framework, KG-Rank, which leverages a medical knowledge graph (KG) with ranking and re-ranking techniques, aiming to improve free-text question-answering (QA) in the medical domain. Specifically, upon receiving a question, we initially retrieve triplets from a medical KG to gather factual information. Subsequently, we innovatively apply ranking methods to refine the ordering of these triplets, aiming to yield more precise answers. To the best of our knowledge, KG-Rank is the first application of ranking models combined with KG in medical QA specifically for generating long answers. Evaluation of four selected medical QA datasets shows that KG-Rank achieves an improvement of over 18% in the ROUGE-L score. Moreover, we extend KG-Rank to open domains, where it realizes a 14% improvement in ROUGE-L, showing the effectiveness and potential of KG-Rank.
Authors: Yuxuan Yao, Sichun Luo, Haohan Zhao, Guanzhi Deng, Linqi Song
Abstract: We present CNER-UAV, a fine-grained \textbf{C}hinese \textbf{N}ame \textbf{E}ntity \textbf{R}ecognition dataset specifically designed for the task of address resolution in \textbf{U}nmanned \textbf{A}erial \textbf{V}ehicle delivery systems. The dataset encompasses a diverse range of five categories, enabling comprehensive training and evaluation of NER models. To construct this dataset, we sourced the data from a real-world UAV delivery system and conducted a rigorous data cleaning and desensitization process to ensure privacy and data integrity. The resulting dataset, consisting of around 12,000 annotated samples, underwent human experts and \textbf{L}arge \textbf{L}anguage \textbf{M}odel annotation. We evaluated classical NER models on our dataset and provided in-depth analysis. The dataset and models are publicly available at \url{https://github.com/zhhvvv/CNER-UAV}.
Authors: Dong Shu, Tianle Chen, Mingyu Jin, Yiting Zhang, Chong Zhang, Mengnan Du, Yongfeng Zhang
Abstract: The task of predicting multiple links within knowledge graphs (KGs) stands as a challenge in the field of knowledge graph analysis, a challenge increasingly resolvable due to advancements in natural language processing (NLP) and KG embedding techniques. This paper introduces a novel methodology, the Knowledge Graph Large Language Model Framework (KG-LLM), which leverages pivotal NLP paradigms, including chain-of-thought (CoT) prompting and in-context learning (ICL), to enhance multi-hop link prediction in KGs. By converting the KG to a CoT prompt, our framework is designed to discern and learn the latent representations of entities and their interrelations. To show the efficacy of the KG-LLM Framework, we fine-tune three leading Large Language Models (LLMs) within this framework, employing both non-ICL and ICL tasks for a comprehensive evaluation. Further, we explore the framework's potential to provide LLMs with zero-shot capabilities for handling previously unseen prompts. Our experimental findings discover that integrating ICL and CoT not only augments the performance of our approach but also significantly boosts the models' generalization capacity, thereby ensuring more precise predictions in unfamiliar scenarios.
Authors: Louis Owen, Vishesh Tripathi, Abhay Kumar, Biddwan Ahmed
Abstract: The recent breakthroughs in Large Language Models (LLMs) have mostly focused on languages with easily available and sufficient resources, such as English. However, there remains a significant gap for languages that lack sufficient linguistic resources in the public domain. Our work introduces Komodo-7B, 7-billion-parameter Large Language Models designed to address this gap by seamlessly operating across Indonesian, English, and 11 regional languages in Indonesia. Komodo-7B is a family of LLMs that consist of Komodo-7B-Base and Komodo-7B-Instruct. Komodo-7B-Instruct stands out by achieving state-of-the-art performance in various tasks and languages, outperforming the benchmarks set by OpenAI's GPT-3.5, Cohere's Aya-101, Llama-2-Chat-13B, Mixtral-8x7B-Instruct-v0.1, Gemma-7B-it , and many more. This model not only demonstrates superior performance in both language-specific and overall assessments but also highlights its capability to excel in linguistic diversity. Our commitment to advancing language models extends beyond well-resourced languages, aiming to bridge the gap for those with limited linguistic assets. Additionally, Komodo-7B-Instruct's better cross-language understanding contributes to addressing educational disparities in Indonesia, offering direct translations from English to 11 regional languages, a significant improvement compared to existing language translation services. Komodo-7B represents a crucial step towards inclusivity and effectiveness in language models, providing to the linguistic needs of diverse communities.
Authors: Thennal D K, Ganesh Nathan, Suchithra M S
Abstract: Fine-tuning pre-trained models provides significant advantages in downstream performance. The ubiquitous nature of pre-trained models such as BERT and its derivatives in natural language processing has also led to a proliferation of task-specific fine-tuned models. As these models typically only perform one task well, additional training or ensembling is required in multi-task scenarios. The growing field of model merging provides a solution, dealing with the challenge of combining multiple task-specific models into a single multi-task model. In this study, we introduce a novel model merging method for Transformers, combining insights from previous work in Fisher-weighted averaging and the use of Fisher information in model pruning. Utilizing the Fisher information of mask nodes within the Transformer architecture, we devise a computationally efficient weighted-averaging scheme. Our method exhibits a regular and significant performance increase across various models in the BERT family, outperforming full-scale Fisher-weighted averaging in a fraction of the computational cost, with baseline performance improvements of up to +6.5 and a speedup of 57.4x in the biggest model. Our results prove the potential of our method in current multi-task learning environments and suggest its scalability and adaptability to new model architectures and learning scenarios.
Authors: Cenyuan Zhang, Xiaoqing Zheng, Ruicheng Yin, Shujie Geng, Jianhan Xu, Xuan Gao, Changze Lv, Zixuan Ling, Xuanjing Huang, Miao Cao, Jianfeng Feng
Abstract: Deciphering natural language from brain activity through non-invasive devices remains a formidable challenge. Previous non-invasive decoders either require multiple experiments with identical stimuli to pinpoint cortical regions and enhance signal-to-noise ratios in brain activity, or they are limited to discerning basic linguistic elements such as letters and words. We propose a novel approach to decoding continuous language from single-trial non-invasive fMRI recordings, in which a three-dimensional convolutional network augmented with information bottleneck is developed to automatically identify responsive voxels to stimuli, and a character-based decoder is designed for the semantic reconstruction of continuous language characterized by inherent character structures. The resulting decoder can produce intelligible textual sequences that faithfully capture the meaning of perceived speech both within and across subjects, while existing decoders exhibit significantly inferior performance in cross-subject contexts. The ability to decode continuous language from single trials across subjects demonstrates the promising applications of non-invasive language brain-computer interfaces in both healthcare and neuroscience.
Authors: Zhihan Zhou, Yanrong Ji, Weijian Li, Pratik Dutta, Ramana Davuluri, Han Liu
Abstract: Decoding the linguistic intricacies of the genome is a crucial problem in biology, and pre-trained foundational models such as DNABERT and Nucleotide Transformer have made significant strides in this area. Existing works have largely hinged on k-mer, fixed-length permutations of A, T, C, and G, as the token of the genome language due to its simplicity. However, we argue that the computation and sample inefficiencies introduced by k-mer tokenization are primary obstacles in developing large genome foundational models. We provide conceptual and empirical insights into genome tokenization, building on which we propose to replace k-mer tokenization with Byte Pair Encoding (BPE), a statistics-based data compression algorithm that constructs tokens by iteratively merging the most frequent co-occurring genome segment in the corpus. We demonstrate that BPE not only overcomes the limitations of k-mer tokenization but also benefits from the computational efficiency of non-overlapping tokenization. Based on these insights, we introduce DNABERT-2, a refined genome foundation model that adapts an efficient tokenizer and employs multiple strategies to overcome input length constraints, reduce time and memory expenditure, and enhance model capability. Furthermore, we identify the absence of a comprehensive and standardized benchmark for genome understanding as another significant impediment to fair comparative analysis. In response, we propose the Genome Understanding Evaluation (GUE), a comprehensive multi-species genome classification dataset that amalgamates $36$ distinct datasets across $9$ tasks, with input lengths ranging from $70$ to $10000$. Through comprehensive experiments on the GUE benchmark, we demonstrate that DNABERT-2 achieves comparable performance to the state-of-the-art model with $21 \times$ fewer parameters and approximately $92 \times$ less GPU time in pre-training.
Authors: Pratyush Maini, Sachin Goyal, Zachary C. Lipton, J. Zico Kolter, Aditi Raghunathan
Abstract: Large web-sourced multimodal datasets have powered a slew of new methods for learning general-purpose visual representations, advancing the state of the art in computer vision and revolutionizing zero- and few-shot recognition. One crucial decision facing practitioners is how, if at all, to curate these ever-larger datasets. For example, the creators of the LAION-5B dataset chose to retain only image-caption pairs whose CLIP similarity score exceeded a designated threshold. In this paper, we propose a new state-of-the-art data filtering approach motivated by our observation that nearly 40% of LAION's images contain text that overlaps significantly with the caption. Intuitively, such data could be wasteful as it incentivizes models to perform optical character recognition rather than learning visual features. However, naively removing all such data could also be wasteful, as it throws away images that contain visual features (in addition to overlapping text). Our simple and scalable approach, T-MARS (Text Masking and Re-Scoring), filters out only those pairs where the text dominates the remaining visual features -- by first masking out the text and then filtering out those with a low CLIP similarity score of the masked image. Experimentally, T-MARS outperforms the top-ranked method on the "medium scale" of DataComp (a data filtering benchmark) by a margin of 6.5% on ImageNet and 4.7% on VTAB. Additionally, our systematic evaluation on various data pool sizes from 2M to 64M shows that the accuracy gains enjoyed by T-MARS linearly increase as data and compute are scaled exponentially. Code is available at https://github.com/locuslab/T-MARS.
Authors: Jiangrui Zheng, Xueqing Liu, Guanqun Yang, Mirazul Haque, Xing Qian, Ravishka Rathnasuriya, Wei Yang, Girish Budhrani
Abstract: To protect users from massive hateful content, existing works studied automated hate speech detection. Despite the existing efforts, one question remains: do automated hate speech detectors conform to social media content policies? A platform's content policies are a checklist of content moderated by the social media platform. Because content moderation rules are often uniquely defined, existing hate speech datasets cannot directly answer this question. This work seeks to answer this question by creating HateModerate, a dataset for testing the behaviors of automated content moderators against content policies. First, we engage 28 annotators and GPT in a six-step annotation process, resulting in a list of hateful and non-hateful test suites matching each of Facebook's 41 hate speech policies. Second, we test the performance of state-of-the-art hate speech detectors against HateModerate, revealing substantial failures these models have in their conformity to the policies. Third, using HateModerate, we augment the training data of a top-downloaded hate detector on HuggingFace. We observe significant improvement in the models' conformity to content policies while having comparable scores on the original test data. Our dataset and code can be found in the attachment.
Authors: Gr\'egoire Del\'etang, Anian Ruoss, Paul-Ambroise Duquenne, Elliot Catt, Tim Genewein, Christopher Mattern, Jordi Grau-Moya, Li Kevin Wenliang, Matthew Aitchison, Laurent Orseau, Marcus Hutter, Joel Veness
Abstract: It has long been established that predictive models can be transformed into lossless compressors and vice versa. Incidentally, in recent years, the machine learning community has focused on training increasingly large and powerful self-supervised (language) models. Since these large language models exhibit impressive predictive capabilities, they are well-positioned to be strong compressors. In this work, we advocate for viewing the prediction problem through the lens of compression and evaluate the compression capabilities of large (foundation) models. We show that large language models are powerful general-purpose predictors and that the compression viewpoint provides novel insights into scaling laws, tokenization, and in-context learning. For example, Chinchilla 70B, while trained primarily on text, compresses ImageNet patches to 43.4% and LibriSpeech samples to 16.4% of their raw size, beating domain-specific compressors like PNG (58.5%) or FLAC (30.3%), respectively. Finally, we show that the prediction-compression equivalence allows us to use any compressor (like gzip) to build a conditional generative model.
Authors: Yinpei Dai, Run Peng, Sikai Li, Joyce Chai
Abstract: Zero-Shot Object Navigation (ZSON) enables agents to navigate towards open-vocabulary objects in unknown environments. The existing works of ZSON mainly focus on following individual instructions to find generic object classes, neglecting the utilization of natural language interaction and the complexities of identifying user-specific objects. To address these limitations, we introduce Zero-shot Interactive Personalized Object Navigation (ZIPON), where robots need to navigate to personalized goal objects while engaging in conversations with users. To solve ZIPON, we propose a new framework termed Open-woRld Interactive persOnalized Navigation (ORION), which uses Large Language Models (LLMs) to make sequential decisions to manipulate different modules for perception, navigation and communication. Experimental results show that the performance of interactive agents that can leverage user feedback exhibits significant improvement. However, obtaining a good balance between task completion and the efficiency of navigation and interaction remains challenging for all methods. We further provide more findings on the impact of diverse user feedback forms on the agents' performance.
Authors: Saurabh Garg, Mehrdad Farajtabar, Hadi Pouransari, Raviteja Vemulapalli, Sachin Mehta, Oncel Tuzel, Vaishaal Shankar, Fartash Faghri
Abstract: Keeping large foundation models up to date on latest data is inherently expensive. To avoid the prohibitive costs of constantly retraining, it is imperative to \emph{continually} train these models. This problem is exacerbated by the lack of any large scale continual learning benchmarks or baselines. We introduce the first set of web-scale Time-Continual (TiC) benchmarks for training vision-language models: TiC-DataComp, TiC-YFCC, and TiC-Redcaps. TiC-DataComp, our largest dataset, contains over 12.7B timestamped image-text pairs spanning 9 years (2014--2022). We first use our benchmarks to curate various \emph{dynamic} evaluations to measure temporal robustness of existing models. We show OpenAI's CLIP (trained on data up to 2020) loses $\approx 8\%$ zero-shot accuracy on our curated retrieval task from 2021--2022 compared with more recently trained models in OpenCLIP repository. We then study how to efficiently train models on time-continuous data. We demonstrate that a simple rehearsal-based approach that continues training from the last checkpoint and replays old data reduces compute by $2.5\times$ when compared to the standard practice of retraining from scratch. Code is available at https://github.com/apple/ml-tic-clip.
Authors: Yangyi Chen, Karan Sikka, Michael Cogswell, Heng Ji, Ajay Divakaran
Abstract: We present DRESS, a large vision language model (LVLM) that innovatively exploits Natural Language feedback (NLF) from Large Language Models to enhance its alignment and interactions by addressing two key limitations in the state-of-the-art LVLMs. First, prior LVLMs generally rely only on the instruction finetuning stage to enhance alignment with human preferences. Without incorporating extra feedback, they are still prone to generate unhelpful, hallucinated, or harmful responses. Second, while the visual instruction tuning data is generally structured in a multi-turn dialogue format, the connections and dependencies among consecutive conversational turns are weak. This reduces the capacity for effective multi-turn interactions. To tackle these, we propose a novel categorization of the NLF into two key types: critique and refinement. The critique NLF identifies the strengths and weaknesses of the responses and is used to align the LVLMs with human preferences. The refinement NLF offers concrete suggestions for improvement and is adopted to improve the interaction ability of the LVLMs-- which focuses on LVLMs' ability to refine responses by incorporating feedback in multi-turn interactions. To address the non-differentiable nature of NLF, we generalize conditional reinforcement learning for training. Our experimental results demonstrate that DRESS can generate more helpful (9.76%), honest (11.52%), and harmless (21.03%) responses, and more effectively learn from feedback during multi-turn interactions compared to SOTA LVMLs.
Authors: Ravi Patel, Angus Brayne, Rogier Hintzen, Daniel Jaroslawicz, Georgiana Neculae, Dane Corneil
Abstract: Machine learning models, particularly language models, are notoriously difficult to introspect. Black-box models can mask both issues in model training and harmful biases. For human-in-the-loop processes, opaque predictions can drive lack of trust, limiting a model's impact even when it performs effectively. To address these issues, we introduce Retrieve to Explain (R2E). R2E is a retrieval-based language model that prioritizes amongst a pre-defined set of possible answers to a research question based on the evidence in a document corpus, using Shapley values to identify the relative importance of pieces of evidence to the final prediction. R2E can adapt to new evidence without retraining, and incorporate structured data through templating into natural language. We assess on the use case of drug target identification from published scientific literature, where we show that the model outperforms an industry-standard genetics-based approach on predicting clinical trial outcomes.
Authors: Denis Jered McInerney, William Dickinson, Lucy C. Flynn, Andrea C. Young, Geoffrey S. Young, Jan-Willem van de Meent, Byron C. Wallace
Abstract: Many diagnostic errors occur because clinicians cannot easily access relevant information in patient Electronic Health Records (EHRs). In this work we propose a method to use LLMs to identify pieces of evidence in patient EHR data that indicate increased or decreased risk of specific diagnoses; our ultimate aim is to increase access to evidence and reduce diagnostic errors. In particular, we propose a Neural Additive Model to make predictions backed by evidence with individualized risk estimates at time-points where clinicians are still uncertain, aiming to specifically mitigate delays in diagnosis and errors stemming from an incomplete differential. To train such a model, it is necessary to infer temporally fine-grained retrospective labels of eventual "true" diagnoses. We do so with LLMs, to ensure that the input text is from before a confident diagnosis can be made. We use an LLM to retrieve an initial pool of evidence, but then refine this set of evidence according to correlations learned by the model. We conduct an in-depth evaluation of the usefulness of our approach by simulating how it might be used by a clinician to decide between a pre-defined list of differential diagnoses.
Authors: Brandon McKinzie, Zhe Gan, Jean-Philippe Fauconnier, Sam Dodge, Bowen Zhang, Philipp Dufter, Dhruti Shah, Xianzhi Du, Futang Peng, Floris Weers, Anton Belyi, Haotian Zhang, Karanjeet Singh, Doug Kang, Ankur Jain, Hongyu H\`e, Max Schwarzer, Tom Gunter, Xiang Kong, Aonan Zhang, Jianyu Wang, Chong Wang, Nan Du, Tao Lei, Sam Wiseman, Mark Lee, Zirui Wang, Ruoming Pang, Peter Grasch, Alexander Toshev, Yinfei Yang
Abstract: In this work, we discuss building performant Multimodal Large Language Models (MLLMs). In particular, we study the importance of various architecture components and data choices. Through careful and comprehensive ablations of the image encoder, the vision language connector, and various pre-training data choices, we identified several crucial design lessons. For example, we demonstrate that for large-scale multimodal pre-training using a careful mix of image-caption, interleaved image-text, and text-only data is crucial for achieving state-of-the-art (SOTA) few-shot results across multiple benchmarks, compared to other published pre-training results. Further, we show that the image encoder together with image resolution and the image token count has substantial impact, while the vision-language connector design is of comparatively negligible importance. By scaling up the presented recipe, we build MM1, a family of multimodal models up to 30B parameters, including both dense models and mixture-of-experts (MoE) variants, that are SOTA in pre-training metrics and achieve competitive performance after supervised fine-tuning on a range of established multimodal benchmarks. Thanks to large-scale pre-training, MM1 enjoys appealing properties such as enhanced in-context learning, and multi-image reasoning, enabling few-shot chain-of-thought prompting.
Authors: Anique Tahir, Lu Cheng, Huan Liu
Abstract: The scaling of Large Language Models (LLMs) for retrieval-based tasks, particularly in Retrieval Augmented Generation (RAG), faces significant memory constraints, especially when fine-tuning extensive prompt sequences. Current open-source libraries support full-model inference and fine-tuning across multiple GPUs but fall short of accommodating the efficient parameter distribution required for retrieved context. Addressing this gap, we introduce a novel framework for PEFT-compatible fine-tuning of Llama-2 models, leveraging distributed training. Our framework uniquely utilizes JAX's just-in-time (JIT) compilation and tensor-sharding for efficient resource management, thereby enabling accelerated fine-tuning with reduced memory requirements. This advancement significantly improves the scalability and feasibility of fine-tuning LLMs for complex RAG applications, even on systems with limited GPU resources. Our experiments show more than 12x improvement in runtime compared to Hugging Face/DeepSpeed implementation with four GPUs while consuming less than half the VRAM per GPU.