Authors: Abdelhadi Soudi, Violetta Cavalli-Sforza, Abderrahim Jamari
Abstract: In this paper, we show that the multiple-stem approach to nouns with a broken plural pattern allows for greater generalizations to be stated in the morphological system. Such an approach dispenses with truncating/deleting rules and other complex rules that are required to account for the highly allomorphic broken plural system. The generation of inflected sound nouns necessitates a pre-specification of the affixes denoting the sound plural masculine and the sound plural feminine, namely uwna and aAt, in the lexicon. The first subsection of section one provides an evaluation of some of the previous analyses of the Arabic broken plural. We provide both linguistic and statistical evidence against deriving broken plurals from the singular or the root. In subsection two, we propose a multiple stem approach to the Arabic Noun Plural System within the Lexeme-based Morphology framework. In section two, we look at the noun inflection of Arabic. Section three provides an implementation of the Arabic Noun system in MORPHE. In this context, we show how the generalizations discussed in the linguistic analysis section are captured in Morphe using the equivalencing nodes.
Authors: Xinyu Wang, Sai Koneru, Pranav Narayanan Venkit, Brett Frischmann, Sarah Rajtmajer
Abstract: As social media has become a predominant mode of communication globally, the rise of abusive content threatens to undermine civil discourse. Recognizing the critical nature of this issue, a significant body of research has been dedicated to developing language models that can detect various types of online abuse, e.g., hate speech, cyberbullying. However, there exists a notable disconnect between platform policies, which often consider the author's intention as a criterion for content moderation, and the current capabilities of detection models, which typically lack efforts to capture intent. This paper examines the role of intent in content moderation systems. We review state of the art detection models and benchmark training datasets for online abuse to assess their awareness and ability to capture intent. We propose strategic changes to the design and development of automated detection and moderation systems to improve alignment with ethical and policy conceptualizations of abuse.
Authors: Ilya Ilyankou, James Haworth, Stefano Cavazzi
Abstract: The Common Crawl (CC) corpus is the largest open web crawl dataset containing 9.5+ petabytes of data captured since 2008. The dataset is instrumental in training large language models, and as such it has been studied for (un)desirable content, and distilled for smaller, domain-specific datasets. However, to our knowledge, no research has been dedicated to using CC as a source of annotated geospatial data. In this paper, we introduce an efficient pipeline to extract annotated user-generated tracks from GPX files found in CC, and the resulting multimodal dataset with 1,416 pairings of human-written descriptions and MultiLineString vector data. The dataset can be used to study people's outdoor activity patterns, the way people talk about their outdoor experiences, and for developing trajectory generation or track annotation models.
Authors: Jace Grandinetti, Rafe McBeth
Abstract: Large Language Models (LLMs) have achieved remarkable progress, yet their application in specialized fields, such as medical physics, remains challenging due to the need for domain-specific knowledge. This study introduces ARCoT (Adaptable Retrieval-based Chain of Thought), a framework designed to enhance the domain-specific accuracy of LLMs without requiring fine-tuning or extensive retraining. ARCoT integrates a retrieval mechanism to access relevant domain-specific information and employs step-back and chain-of-thought prompting techniques to guide the LLM's reasoning process, ensuring more accurate and context-aware responses. Benchmarking on a medical physics multiple-choice exam, our model outperformed standard LLMs and reported average human performance, demonstrating improvements of up to 68% and achieving a high score of 90%. This method reduces hallucinations and increases domain-specific performance. The versatility and model-agnostic nature of ARCoT make it easily adaptable to various domains, showcasing its significant potential for enhancing the accuracy and reliability of LLMs in specialized fields.
Authors: Virgile Rennard, Guokan Shang, Julie Hunter, Michalis Vazirgiannis
Abstract: We introduce an extractive summarization system for meetings that leverages discourse structure to better identify salient information from complex multi-party discussions. Using discourse graphs to represent semantic relations between the contents of utterances in a meeting, we train a GNN-based node classification model to select the most important utterances, which are then combined to create an extractive summary. Experimental results on AMI and ICSI demonstrate that our approach surpasses existing text-based and graph-based extractive summarization systems, as measured by both classification and summarization metrics. Additionally, we conduct ablation studies on discourse structure and relation type to provide insights for future NLP applications leveraging discourse analysis theory.
Authors: Michael Feffer, Ronald Xu, Yuekai Sun, Mikhail Yurochkin
Abstract: In the advent of democratized usage of large language models (LLMs), there is a growing desire to systematize LLM prompt creation and selection processes beyond iterative trial-and-error. Prior works majorly focus on searching the space of prompts without accounting for relations between prompt variations. Here we propose a framework, Prompt Exploration with Prompt Regression (PEPR), to predict the effect of prompt combinations given results for individual prompt elements as well as a simple method to select an effective prompt for a given use-case. We evaluate our approach with open-source LLMs of different sizes on several different tasks.
Authors: Denis Kokosinskii, Nikolay Arefyev
Abstract: Word Sense Induction (WSI) is the task of discovering senses of an ambiguous word by grouping usages of this word into clusters corresponding to these senses. Many approaches were proposed to solve WSI in English and a few other languages, but these approaches are not easily adaptable to new languages. We present multilingual substitution-based WSI methods that support any of 100 languages covered by the underlying multilingual language model with minimal to no adaptation required. Despite the multilingual capabilities, our methods perform on par with the existing monolingual approaches on popular English WSI datasets. At the same time, they will be most useful for lower-resourced languages which miss lexical resources available for English, thus, have higher demand for unsupervised methods like WSI.
Authors: Allmin Balloccu, Jack Zhang
Abstract: This paper introduces Dynamic Embeddings with Task-Oriented prompting (DETOT), a novel approach aimed at improving the adaptability and efficiency of machine learning models by implementing a flexible embedding layer. Unlike traditional static embeddings [14], DETOT dynamically adjusts embeddings based on task-specific requirements and performance feedback, optimizing input data representation for individual tasks [4]. This method enhances both accuracy and computational performance by tailoring the representation layer to meet the unique needs of each task. The structure of DETOT is detailed, highlighting its task-specific adaptation, continuous feedback loop, and mechanisms for preventing overfitting. Empirical evaluations demonstrate its superiority over existing methods.
Authors: Hasti Toossi, Guo Qing Huai, Jinyu Liu, Eric Khiu, A. Seza Do\u{g}ru\"oz, En-Shiun Annie Lee
Abstract: In the pursuit of supporting more languages around the world, tools that characterize properties of languages play a key role in expanding the existing multilingual NLP research. In this study, we focus on a widely used typological knowledge base, URIEL, which aggregates linguistic information into numeric vectors. Specifically, we delve into the soundness and reproducibility of the approach taken by URIEL in quantifying language similarity. Our analysis reveals URIEL's ambiguity in calculating language distances and in handling missing values. Moreover, we find that URIEL does not provide any information about typological features for 31\% of the languages it represents, undermining the reliabilility of the database, particularly on low-resource languages. Our literature review suggests URIEL and lang2vec are used in papers on diverse NLP tasks, which motivates us to rigorously verify the database as the effectiveness of these works depends on the reliability of the information the tool provides.
Authors: Yongrae Jo, Seongyun Lee, Minju Seo, Sung Ju Hwang, Moontae Lee
Abstract: Text-to-SQL models are pivotal for making Electronic Health Records (EHRs) accessible to healthcare professionals without SQL knowledge. With the advancements in large language models, these systems have become more adept at translating complex questions into SQL queries. Nonetheless, the critical need for reliability in healthcare necessitates these models to accurately identify unanswerable questions or uncertain predictions, preventing misinformation. To address this problem, we present a self-training strategy using pseudo-labeled unanswerable questions to enhance the reliability of text-to-SQL models for EHRs. This approach includes a two-stage training process followed by a filtering method based on the token entropy and query execution. Our methodology's effectiveness is validated by our top performance in the EHRSQL 2024 shared task, showcasing the potential to improve healthcare decision-making through more reliable text-to-SQL systems.
Authors: Sichang Tu, Abigail Powers, Natalie Merrill, Negar Fani, Sierra Carter, Stephen Doogan, Jinho D. Choi
Abstract: The shortage of clinical workforce presents significant challenges in mental healthcare, limiting access to formal diagnostics and services. We aim to tackle this shortage by integrating a customized large language model (LLM) into the workflow, thus promoting equity in mental healthcare for the general population. Although LLMs have showcased their capability in clinical decision-making, their adaptation to severe conditions like Post-traumatic Stress Disorder (PTSD) remains largely unexplored. Therefore, we collect 411 clinician-administered diagnostic interviews and devise a novel approach to obtain high-quality data. Moreover, we build a comprehensive framework to automate PTSD diagnostic assessments based on interview contents by leveraging two state-of-the-art LLMs, GPT-4 and Llama-2, with potential for broader clinical diagnoses. Our results illustrate strong promise for LLMs, tested on our dataset, to aid clinicians in diagnostic validation. To the best of our knowledge, this is the first AI system that fully automates assessments for mental illness based on clinician-administered interviews.
Authors: Manav Chaudhary, Harshit Gupta, Vasudeva Varma
Abstract: The proliferation of LLMs in various NLP tasks has sparked debates regarding their reliability, particularly in annotation tasks where biases and hallucinations may arise. In this shared task, we address the challenge of distinguishing annotations made by LLMs from those made by human domain experts in the context of COVID-19 symptom detection from tweets in Latin American Spanish. This paper presents BrainStorm @ iREL's approach to the SMM4H 2024 Shared Task, leveraging the inherent topical information in tweets, we propose a novel approach to identify and classify annotations, aiming to enhance the trustworthiness of annotated data.
Authors: Claudiu Creanga, Liviu P. Dinu
Abstract: Natural Language Inference (NLI) is foundational for evaluating language understanding in AI. However, progress has plateaued, with models failing on ambiguous examples and exhibiting poor generalization. We argue that this stems from disregarding the subjective nature of meaning, which is intrinsically tied to an individual's \textit{weltanschauung} (which roughly translates to worldview). Existing NLP datasets often obscure this by aggregating labels or filtering out disagreement. We propose a perspectivist approach: building datasets that capture annotator demographics, values, and justifications for their labels. Such datasets would explicitly model diverse worldviews. Our initial experiments with a subset of the SBIC dataset demonstrate that even limited annotator metadata can improve model performance.
Authors: Karthika NJ, Ayush Maheshwari, Atul Kumar Singh, Preethi Jyothi, Ganesh Ramakrishnan, Krishnakant Bhatt
Abstract: Lexicon or dictionary generation across domains is of significant societal importance, as it can potentially enhance information accessibility for a diverse user base while preserving language identity. Prior work in the field primarily focuses on bilingual lexical induction, which deals with word alignments using mapping-based or corpora-based approaches. Though initiated by researchers, the research associated with lexicon generation is limited, even more so with domain-specific lexicons. This task becomes particularly important in atypical medical, engineering, and other technical domains, owing to the highly infrequent usage of the terms and negligibly low data availability of technical terms in many low-resource languages. Owing to the research gap in lexicon generation, especially with a limited focus on the domain-specific area, we propose a new model to generate dictionary words for 6 Indian languages in the multi-domain setting. Our model consists of domain-specific and domain-generic layers that encode information, and these layers are invoked via a learnable routing technique. Further, we propose an approach to explicitly leverage the relatedness between these Indian languages toward coherent translation. We also release a new benchmark dataset across 6 Indian languages that span 8 diverse domains that can propel further research in domain-specific lexicon induction. We conduct both zero-shot and few-shot experiments across multiple domains to show the efficacy of our proposed model in generalizing to unseen domains and unseen languages.
Authors: Claudiu Creanga, Liviu Petrisor Dinu
Abstract: We used Data Maps to model and characterize the AuTexTification dataset. This provides insights about the behaviour of individual samples during training across epochs (training dynamics). We characterized the samples across 3 dimensions: confidence, variability and correctness. This shows the presence of 3 regions: easy-to-learn, ambiguous and hard-to-learn examples. We used a classic CNN architecture and found out that training the model only on a subset of ambiguous examples improves the model's out-of-distribution generalization.
Authors: Siddhant Agarwal, Shivam Sharma, Preslav Nakov, Tanmoy Chakraborty
Abstract: Memes have evolved as a prevalent medium for diverse communication, ranging from humour to propaganda. With the rising popularity of image-focused content, there is a growing need to explore its potential harm from different aspects. Previous studies have analyzed memes in closed settings - detecting harm, applying semantic labels, and offering natural language explanations. To extend this research, we introduce MemeMQA, a multimodal question-answering framework aiming to solicit accurate responses to structured questions while providing coherent explanations. We curate MemeMQACorpus, a new dataset featuring 1,880 questions related to 1,122 memes with corresponding answer-explanation pairs. We further propose ARSENAL, a novel two-stage multimodal framework that leverages the reasoning capabilities of LLMs to address MemeMQA. We benchmark MemeMQA using competitive baselines and demonstrate its superiority - ~18% enhanced answer prediction accuracy and distinct text generation lead across various metrics measuring lexical and semantic alignment over the best baseline. We analyze ARSENAL's robustness through diversification of question-set, confounder-based evaluation regarding MemeMQA's generalizability, and modality-specific assessment, enhancing our understanding of meme interpretation in the multimodal communication landscape.
Authors: Anthony Hughes, Xingyi Song
Abstract: Evidence-based medicine is the practice of making medical decisions that adhere to the latest, and best known evidence at that time. Currently, the best evidence is often found in the form of documents, such as randomized control trials, meta-analyses and systematic reviews. This research focuses on aligning medical claims made on social media platforms with this medical evidence. By doing so, individuals without medical expertise can more effectively assess the veracity of such medical claims. We study three core tasks: identifying medical claims, extracting medical vocabulary from these claims, and retrieving evidence relevant to those identified medical claims. We propose a novel system that can generate synthetic medical claims to aid each of these core tasks. We additionally introduce a novel dataset produced by our synthetic generator that, when applied to these tasks, demonstrates not only a more flexible and holistic approach, but also an improvement in all comparable metrics. We make our dataset, the Expansive Medical Claim Corpus (EMCC), available at https://zenodo.org/records/8321460
Authors: Claudiu Creanga, Liviu P. Dinu
Abstract: This paper outlines the approach of the ISDS-NLP team in the SemEval 2024 Task 10: Emotion Discovery and Reasoning its Flip in Conversation (EDiReF). For Subtask 1 we obtained a weighted F1 score of 0.43 and placed 12 in the leaderboard. We investigate two distinct approaches: Masked Language Modeling (MLM) and Causal Language Modeling (CLM). For MLM, we employ pre-trained BERT-like models in a multilingual setting, fine-tuning them with a classifier to predict emotions. Experiments with varying input lengths, classifier architectures, and fine-tuning strategies demonstrate the effectiveness of this approach. Additionally, we utilize Mistral 7B Instruct V0.2, a state-of-the-art model, applying zero-shot and few-shot prompting techniques. Our findings indicate that while Mistral shows promise, MLMs currently outperform them in sentence-level emotion classification.
Authors: Hendrik Damm, Tabea M. G. Pakull, Bahad{\i}r Ery{\i}lmaz, Helmut Becker, Ahmad Idrissi-Yaghir, Henning Sch\"afer, Sergej Schultenk\"amper, Christoph M. Friedrich
Abstract: This study aims to leverage state of the art language models to automate generating the "Brief Hospital Course" and "Discharge Instructions" sections of Discharge Summaries from the MIMIC-IV dataset, reducing clinicians' administrative workload. We investigate how automation can improve documentation accuracy, alleviate clinician burnout, and enhance operational efficacy in healthcare facilities. This research was conducted within our participation in the Shared Task Discharge Me! at BioNLP @ ACL 2024. Various strategies were employed, including few-shot learning, instruction tuning, and Dynamic Expert Selection (DES), to develop models capable of generating the required text sections. Notably, utilizing an additional clinical domain-specific dataset demonstrated substantial potential to enhance clinical language processing. The DES method, which optimizes the selection of text outputs from multiple predictions, proved to be especially effective. It achieved the highest overall score of 0.332 in the competition, surpassing single-model outputs. This finding suggests that advanced deep learning methods in combination with DES can effectively automate parts of electronic health record documentation. These advancements could enhance patient care by freeing clinician time for patient interactions. The integration of text selection strategies represents a promising avenue for further research.
Authors: Gargi Sathe, Aneesh Shamraj, Aditya Surve, Nahush Patil, Kumkum Saxena
Abstract: This paper presents an evaluation of the mathematical capability of ChatGPT across diverse languages like Hindi, Gujarati, and Marathi. ChatGPT, based on GPT-3.5 by OpenAI, has garnered significant attention for its natural language understanding and generation abilities. However, its performance in solving mathematical problems across multiple natural languages remains a comparatively unexplored area, especially in regional Indian languages. In this paper, we explore those capabilities as well as using chain-of-thought prompting to figure out if it increases the accuracy of responses as much as it does in the English language and provide insights into the current limitations.
Authors: Yu Huang, Liang Guo, Wanqian Guo, Zhe Tao, Yang Lv, Zhihao Sun, Dongfang Zhao
Abstract: In the field of environmental science, it is crucial to have robust evaluation metrics for large language models to ensure their efficacy and accuracy. We propose EnviroExam, a comprehensive evaluation method designed to assess the knowledge of large language models in the field of environmental science. EnviroExam is based on the curricula of top international universities, covering undergraduate, master's, and doctoral courses, and includes 936 questions across 42 core courses. By conducting 0-shot and 5-shot tests on 31 open-source large language models, EnviroExam reveals the performance differences among these models in the domain of environmental science and provides detailed evaluation standards. The results show that 61.3% of the models passed the 5-shot tests, while 48.39% passed the 0-shot tests. By introducing the coefficient of variation as an indicator, we evaluate the performance of mainstream open-source large language models in environmental science from multiple perspectives, providing effective criteria for selecting and fine-tuning language models in this field. Future research will involve constructing more domain-specific test sets using specialized environmental science textbooks to further enhance the accuracy and specificity of the evaluation.
Authors: Ning Shi, Zijun Wu, Lili Mou
Abstract: Recent studies have demonstrated the potential to control paraphrase generation, such as through syntax, which has broad applications in various downstream tasks. However, these methods often require detailed parse trees or syntactic exemplars, which are not user-friendly. Furthermore, an inference gap exists, as control specifications are only available during training but not inference. In this work, we propose a new setup for controlled paraphrasing. Specifically, we represent user-intended actions as action tokens, allowing embedding and concatenating them with text embeddings, thus flowing together to a self-attention encoder for representation fusion. To address the inference gap, we introduce an optional action token as a placeholder that encourages the model to determine the appropriate action when control specifications are inaccessible. Experimental results show that our method successfully enables specific action-controlled paraphrasing and preserves the same or even better performance compared to conventional uncontrolled methods when actions are not given. Our findings thus promote the concept of optional action control for a more user-centered design via representation learning.
Authors: Amr Keleg, Walid Magdy, Sharon Goldwater
Abstract: On annotating multi-dialect Arabic datasets, it is common to randomly assign the samples across a pool of native Arabic speakers. Recent analyses recommended routing dialectal samples to native speakers of their respective dialects to build higher-quality datasets. However, automatically identifying the dialect of samples is hard. Moreover, the pool of annotators who are native speakers of specific Arabic dialects might be scarce. Arabic Level of Dialectness (ALDi) was recently introduced as a quantitative variable that measures how sentences diverge from Standard Arabic. On randomly assigning samples to annotators, we hypothesize that samples of higher ALDi scores are harder to label especially if they are written in dialects that the annotators do not speak. We test this by analyzing the relation between ALDi scores and the annotators' agreement, on 15 public datasets having raw individual sample annotations for various sentence-classification tasks. We find strong evidence supporting our hypothesis for 11 of them. Consequently, we recommend prioritizing routing samples of high ALDi scores to native speakers of each sample's dialect, for which the dialect could be automatically identified at higher accuracies.
Authors: Shaina Raza, Ananya Raval, Veronica Chatrath
Abstract: In addressing the critical need for safety in Large Language Models (LLMs), it is crucial to ensure that the outputs are not only safe but also retain their contextual accuracy. Many existing LLMs are safe fine-tuned either with safety demonstrations, or rely only on adversarial testing. While able to get safe outputs, they often risk losing contextual meaning as they mitigate bias and toxicity. In response, we present MBIAS, a LLM framework instruction fine-tuned on a custom dataset specifically designed for safety interventions. MBIAS aims to address the significant issues of bias and toxicity in LLMs generations that typically manifest as underrepresentation or negative portrayals across various demographics, including inappropriate linguistic mentions and biased content in social media. We experiment on MBIAS for safety interventions using various configurations, and demonstrate more than a 30\% reduction in overall bias and toxicity while successfully retaining key information. Additionally, a demographic analysis on an out-of-distribution test set confirms the robustness of our approach, with reductions in bias and toxicity exceeding 90\% across various demographics. The dataset and instruction fine-tuned MBIAS are made available to the research community at https://huggingface.co/newsmediabias/MBIAS.
Authors: Euna Jung, Jaeill Kim, Jungmin Ko, Jinwoo Park, Wonjong Rhee
Abstract: The latest advancements in unsupervised learning of sentence embeddings predominantly involve employing contrastive learning-based (CL-based) fine-tuning over pre-trained language models. In this study, we analyze the latest sentence embedding methods by adopting representation rank as the primary tool of analysis. We first define Phase 1 and Phase 2 of fine-tuning based on when representation rank peaks. Utilizing these phases, we conduct a thorough analysis and obtain essential findings across key aspects, including alignment and uniformity, linguistic abilities, and correlation between performance and rank. For instance, we find that the dynamics of the key aspects can undergo significant changes as fine-tuning transitions from Phase 1 to Phase 2. Based on these findings, we experiment with a rank reduction (RR) strategy that facilitates rapid and stable fine-tuning of the latest CL-based methods. Through empirical investigations, we showcase the efficacy of RR in enhancing the performance and stability of five state-of-the-art sentence embedding methods.
Authors: Canshi Wei
Abstract: Fine-grained image classification, particularly in zero/few-shot scenarios, presents a significant challenge for vision-language models (VLMs), such as CLIP. These models often struggle with the nuanced task of distinguishing between semantically similar classes due to limitations in their pre-trained recipe, which lacks supervision signals for fine-grained categorization. This paper introduces CascadeVLM, an innovative framework that overcomes the constraints of previous CLIP-based methods by effectively leveraging the granular knowledge encapsulated within large vision-language models (LVLMs). Experiments across various fine-grained image datasets demonstrate that CascadeVLM significantly outperforms existing models, specifically on the Stanford Cars dataset, achieving an impressive 85.6% zero-shot accuracy. Performance gain analysis validates that LVLMs produce more accurate predictions for challenging images that CLIPs are uncertain about, bringing the overall accuracy boost. Our framework sheds light on a holistic integration of VLMs and LVLMs for effective and efficient fine-grained image classification.
Authors: Andrew Shin, Kunitake Kaneko
Abstract: Large language models (LLMs) have demonstrated remarkable performances on a wide range of natural language tasks. Yet, LLMs' successes have been largely restricted to tasks concerning words, sentences, or documents, and it remains questionable how much they understand the minimal units of text, namely characters. In this paper, we examine contemporary LLMs regarding their ability to understand character composition of words, and show that most of them fail to reliably carry out even the simple tasks that can be handled by humans with perfection. We analyze their behaviors with comparison to token level performances, and discuss the potential directions for future research.
Authors: Md. Ashraful Islam, Mohammed Eunus Ali, Md Rizwan Parvez
Abstract: Code synthesis, which requires a deep understanding of complex natural language problem descriptions, generation of code instructions for complex algorithms and data structures, and the successful execution of comprehensive unit tests, presents a significant challenge. While large language models (LLMs) demonstrate impressive proficiency in natural language processing, their performance in code generation tasks remains limited. In this paper, we introduce a new approach to code generation tasks leveraging multi-agent prompting that uniquely replicates the full cycle of program synthesis as observed in human developers. Our framework, MapCoder, consists of four LLM agents specifically designed to emulate the stages of this cycle: recalling relevant examples, planning, code generation, and debugging. After conducting thorough experiments, with multiple LLM ablations and analyses across eight challenging competitive problem-solving and program synthesis benchmarks, MapCoder showcases remarkable code generation capabilities, achieving new state-of-the-art results (pass@1) on HumanEval (93.9%), MBPP (83.1%), APPS (22.0%), CodeContests (28.5%), and xCodeEval (45.3%). Moreover, our method consistently delivers superior performance across various programming languages and varying problem difficulties. We open-source our framework at https://github.com/Md-Ashraful-Pramanik/MapCoder.
Authors: Nikil Sharan Prabahar Balasubramanian, Sagnik Dakshit
Abstract: The advancements in the development of Large Language Models have evolved as a transformative paradigm in conversational tasks which has led to its integration in the critical domain of healthcare. With LLMs becoming widely popular and their public access through open-source models, there is a need to investigate their potential and limitations. One such critical task where LLMs are applied but require a deeper understanding is that of self-diagnosis of medical conditions in the interest of public health. The widespread integration of Gemini with Google search, GPT-4.0 with Bing search, has led to shift in trend of self-diagnosis from search engine LLMs. In this paper, we prepare a prompt engineered dataset of 10000 samples and test the performance on the general task of self-diagnosis. We compare the performance of GPT-4.0 and Gemini model on the task of self-diagnosis and record accuracies of 63.07% and 6.01% respectively. We also discuss the challenges, limitations, and potential of both Gemini and GPT-4.0 for the task of self-diagnosis to facilitate future research and towards the broader impact of general public knowledge. Furthermore, we demonstrate the potential and improvement in performance for the task of self-diagnosis using Retrieval Augmented Generation.
Authors: William M. Hayes, Nicolas Yax, Stefano Palminteri
Abstract: In-context learning enables large language models (LLMs) to perform a variety of tasks, including learning to make reward-maximizing choices in simple bandit tasks. Given their potential use as (autonomous) decision-making agents, it is important to understand how these models perform such reinforcement learning (RL) tasks and the extent to which they are susceptible to biases. Motivated by the fact that, in humans, it has been widely documented that the value of an outcome depends on how it compares to other local outcomes, the present study focuses on whether similar value encoding biases apply to how LLMs encode rewarding outcomes. Results from experiments with multiple bandit tasks and models show that LLMs exhibit behavioral signatures of a relative value bias. Adding explicit outcome comparisons to the prompt produces opposing effects on performance, enhancing maximization in trained choice sets but impairing generalization to new choice sets. Computational cognitive modeling reveals that LLM behavior is well-described by a simple RL algorithm that incorporates relative values at the outcome encoding stage. Lastly, we present preliminary evidence that the observed biases are not limited to fine-tuned LLMs, and that relative value processing is detectable in the final hidden layer activations of a raw, pretrained model. These findings have important implications for the use of LLMs in decision-making applications.
Authors: Jianbo Dai, Jianqiao Lu, Yunlong Feng, Rongju Ruan, Ming Cheng, Haochen Tan, Zhijiang Guo
Abstract: Recent advancements in large language models (LLMs) have greatly improved code generation, specifically at the function level. For instance, GPT-4 has achieved an 88.4% pass rate on HumanEval. However, this draws into question the adequacy of existing benchmarks in thoroughly assessing function-level code generation capabilities. Our study analyzed two common benchmarks, HumanEval and MBPP, and found that these might not thoroughly evaluate LLMs' code generation capacities due to limitations in quality, difficulty, and granularity. To resolve this, we introduce the Mostly Hard Python Problems (MHPP) dataset, consisting of 140 unique human-curated problems. By focusing on the combination of natural language and code reasoning, MHPP gauges LLMs' abilities to comprehend specifications and restrictions, engage in multi-step reasoning, and apply coding knowledge effectively. Initial evaluations of 22 LLMs using MHPP showed many high-performing models on HumanEval failed to achieve similar success on MHPP. Moreover, MHPP highlighted various previously undiscovered limitations within various LLMs, leading us to believe that it could pave the way for a better understanding of LLMs' capabilities and limitations. Dataset and code are available at https://github.com/SparksofAGI/MHPP.
Authors: Sanchit Sinha, Yuguang Yue, Victor Soto, Mayank Kulkarni, Jianhua Lu, Aidong Zhang
Abstract: Adapting large language models (LLMs) to unseen tasks with in-context training samples without fine-tuning remains an important research problem. To learn a robust LLM that adapts well to unseen tasks, multiple meta-training approaches have been proposed such as MetaICL and MetaICT, which involve meta-training pre-trained LLMs on a wide variety of diverse tasks. These meta-training approaches essentially perform in-context multi-task fine-tuning and evaluate on a disjointed test set of tasks. Even though they achieve impressive performance, their goal is never to compute a truly general set of parameters. In this paper, we propose MAML-en-LLM, a novel method for meta-training LLMs, which can learn truly generalizable parameters that not only perform well on disjointed tasks but also adapts to unseen tasks. We see an average increase of 2% on unseen domains in the performance while a massive 4% improvement on adaptation performance. Furthermore, we demonstrate that MAML-en-LLM outperforms baselines in settings with limited amount of training data on both seen and unseen domains by an average of 2%. Finally, we discuss the effects of type of tasks, optimizers and task complexity, an avenue barely explored in meta-training literature. Exhaustive experiments across 7 task settings along with two data settings demonstrate that models trained with MAML-en-LLM outperform SOTA meta-training approaches.
Authors: Pengxiang Lan, Enneng Yang, Yuting Liu, Guibing Guo, Linying Jiang, Jianzhe Zhao, Xingwei Wang
Abstract: Prompt tuning is a promising method to fine-tune a pre-trained language model without retraining its large-scale parameters. Instead, it attaches a soft prompt to the input text, whereby downstream tasks can be well adapted by merely learning the embeddings of prompt tokens. Nevertheless, existing methods still suffer from two challenges: (i) they are hard to balance accuracy and efficiency. A longer (shorter) soft prompt generally leads to a better (worse) accuracy but at the cost of more (less) training time. (ii) The performance may not be consistent when adapting to different downstream tasks. We attribute it to the same embedding space but responsible for different requirements of downstream tasks. To address these issues, we propose an Efficient Prompt Tuning method (EPT) by multi-space projection and prompt fusion. Specifically, it decomposes a given soft prompt into a shorter prompt and two low-rank matrices, whereby the number of parameters is greatly reduced as well as the training time. The accuracy is also enhanced by leveraging low-rank matrices and the short prompt as additional knowledge sources to enrich the semantics of the original short prompt. In addition, we project the soft prompt into multiple subspaces to improve the performance consistency, and then adaptively learn the combination weights of different spaces through a gating network. Experimental experiments on 13 natural language processing downstream tasks show that our method significantly and consistently outperforms 11 comparison methods with the relative percentage of improvements up to 28.8%, and training time decreased by 14%.
Authors: Zhongxiang Sun, Kepu Zhang, Haoyu Wang, Xiao Zhang, Jun Xu
Abstract: In-context learning has been extensively validated in large language models. However, the mechanism and selection strategy for in-context example selection, which is a crucial ingredient in this approach, lacks systematic and in-depth research. In this paper, we propose a data compression approach to the selection of in-context examples. We introduce a two-stage method that can effectively choose relevant examples and retain sufficient information about the training dataset within the in-context examples. Our method shows a significant improvement of an average of 5.90% across five different real-world datasets using four language models.
Authors: Quentin Meeus, Marie-Francine Moens, Hugo Van hamme
Abstract: While extensively explored in text-based tasks, Named Entity Recognition (NER) remains largely neglected in spoken language understanding. Existing resources are limited to a single, English-only dataset. This paper addresses this gap by introducing MSNER, a freely available, multilingual speech corpus annotated with named entities. It provides annotations to the VoxPopuli dataset in four languages (Dutch, French, German, and Spanish). We have also releasing an efficient annotation tool that leverages automatic pre-annotations for faster manual refinement. This results in 590 and 15 hours of silver-annotated speech for training and validation, alongside a 17-hour, manually-annotated evaluation set. We further provide an analysis comparing silver and gold annotations. Finally, we present baseline NER models to stimulate further research on this newly available dataset.
Authors: Mengyu Li, Yonghao Liu, Fausto Giunchiglia, Xiaoyue Feng, Renchu Guan
Abstract: Text classification is a crucial and fundamental task in natural language processing. Compared with the previous learning paradigm of pre-training and fine-tuning by cross entropy loss, the recently proposed supervised contrastive learning approach has received tremendous attention due to its powerful feature learning capability and robustness. Although several studies have incorporated this technique for text classification, some limitations remain. First, many text datasets are imbalanced, and the learning mechanism of supervised contrastive learning is sensitive to data imbalance, which may harm the model performance. Moreover, these models leverage separate classification branch with cross entropy and supervised contrastive learning branch without explicit mutual guidance. To this end, we propose a novel model named SharpReCL for imbalanced text classification tasks. First, we obtain the prototype vector of each class in the balanced classification branch to act as a representation of each class. Then, by further explicitly leveraging the prototype vectors, we construct a proper and sufficient target sample set with the same size for each class to perform the supervised contrastive learning procedure. The empirical results show the effectiveness of our model, which even outperforms popular large language models across several datasets.
Authors: Suyash Vardhan Mathur, Akshett Rai Jindal, Manish Shrivastava
Abstract: While significant work has been done in the field of NLP on vertical thinking, which involves primarily logical thinking, little work has been done towards lateral thinking, which involves looking at problems from an unconventional perspective and defying existing conceptions and notions. Towards this direction, SemEval 2024 introduces the task of BRAINTEASER, which involves two types of questions -- Sentence Puzzles and Word Puzzles that defy conventional common-sense reasoning and constraints. In this paper, we tackle both types of questions using few-shot prompting on GPT-3.5 and gain insights regarding the difference in the nature of the two types. Our prompting strategy placed us 26th on the leaderboard for the Sentence Puzzle and 15th on the Word Puzzle task.
Authors: Xuanli He, Qiongkai Xu, Jun Wang, Benjamin I. P. Rubinstein, Trevor Cohn
Abstract: Modern NLP models are often trained on public datasets drawn from diverse sources, rendering them vulnerable to data poisoning attacks. These attacks can manipulate the model's behavior in ways engineered by the attacker. One such tactic involves the implantation of backdoors, achieved by poisoning specific training instances with a textual trigger and a target class label. Several strategies have been proposed to mitigate the risks associated with backdoor attacks by identifying and removing suspected poisoned examples. However, we observe that these strategies fail to offer effective protection against several advanced backdoor attacks. To remedy this deficiency, we propose a novel defensive mechanism that first exploits training dynamics to identify poisoned samples with high precision, followed by a label propagation step to improve recall and thus remove the majority of poisoned instances. Compared with recent advanced defense methods, our method considerably reduces the success rates of several backdoor attacks while maintaining high classification accuracy on clean test sets.
Authors: Bowen Chen, Namgi Han, Yusuke Miyao
Abstract: Large Language Models (LLMs), trained on massive corpora with billions of parameters, show unprecedented performance in various fields. Though surprised by their excellent performances, researchers also noticed some special behaviors of those LLMs. One of those behaviors is memorization, in which LLMs can generate the same content used to train them. Though previous research has discussed memorization, the memorization of LLMs still lacks explanation, especially the cause of memorization and the dynamics of generating them. In this research, we comprehensively discussed memorization from various perspectives and extended the discussion scope to not only just the memorized content but also less and unmemorized content. Through various studies, we found that: (1) Through experiments, we revealed the relation of memorization between model size, continuation size, and context size. Further, we showed how unmemorized sentences transition to memorized sentences. (2) Through embedding analysis, we showed the distribution and decoding dynamics across model size in embedding space for sentences with different memorization scores. The n-gram statistics analysis presents d (3) An analysis over n-gram and entropy decoding dynamics discovered a boundary effect when the model starts to generate memorized sentences or unmemorized sentences. (4)We trained a Transformer model to predict the memorization of different models, showing that it is possible to predict memorizations by context.
Authors: Subhankar Maity, Aniket Deroy, Sudeshna Sarkar
Abstract: In the era of generative artificial intelligence (AI), the fusion of large language models (LLMs) offers unprecedented opportunities for innovation in the field of modern education. We embark on an exploration of prompted LLMs within the context of educational and assessment applications to uncover their potential. Through a series of carefully crafted research questions, we investigate the effectiveness of prompt-based techniques in generating open-ended questions from school-level textbooks, assess their efficiency in generating open-ended questions from undergraduate-level technical textbooks, and explore the feasibility of employing a chain-of-thought inspired multi-stage prompting approach for language-agnostic multiple-choice question (MCQ) generation. Additionally, we evaluate the ability of prompted LLMs for language learning, exemplified through a case study in the low-resource Indian language Bengali, to explain Bengali grammatical errors. We also evaluate the potential of prompted LLMs to assess human resource (HR) spoken interview transcripts. By juxtaposing the capabilities of LLMs with those of human experts across various educational tasks and domains, our aim is to shed light on the potential and limitations of LLMs in reshaping educational practices.
Authors: Congchi Yin, Ziyi Ye, Piji Li
Abstract: Many recent studies have shown that the perception of speech can be decoded from brain signals and subsequently reconstructed as continuous language. However, there is a lack of neurological basis for how the semantic information embedded within brain signals can be used more effectively to guide language reconstruction. The theory of predictive coding suggests that human brain naturally engages in continuously predicting future word representations that span multiple timescales. This implies that the decoding of brain signals could potentially be associated with a predictable future. To explore the predictive coding theory within the context of language reconstruction, this paper proposes a novel model \textsc{PredFT} for jointly modeling neural decoding and brain prediction. It consists of a main decoding network for language reconstruction and a side network for predictive coding. The side network obtains brain predictive coding representation from related brain regions of interest with a multi-head self-attention module. This representation is fused into the main decoding network with cross-attention to facilitate the language models' generation process. Experiments are conducted on the largest naturalistic language comprehension fMRI dataset Narratives. \textsc{PredFT} achieves current state-of-the-art decoding performance with a maximum BLEU-1 score of $27.8\%$.
Authors: Baolong Bi, Shenghua Liu, Lingrui Mei, Yiwei Wang, Pengliang Ji, Xueqi Cheng
Abstract: The knowledge within large language models (LLMs) may become outdated quickly. While in-context editing (ICE) is currently the most effective method for knowledge editing (KE), it is constrained by the black-box modeling of LLMs and thus lacks interpretability. Our work aims to elucidate the superior performance of ICE on the KE by analyzing the impacts of in-context new knowledge on token-wise distributions. We observe that despite a significant boost in logits of the new knowledge, the performance of is still hindered by stubborn knowledge. Stubborn knowledge refers to as facts that have gained excessive confidence during pretraining, making it hard to edit effectively. To address this issue and further enhance the performance of ICE, we propose a novel approach termed $\textbf{De}$coding by $\textbf{C}$ontrasting $\textbf{K}$nowledge (DeCK). DeCK derives the distribution of the next token by contrasting the logits obtained from the newly edited knowledge guided by ICE with those from the unedited parametric knowledge. Our experiments consistently demonstrate that DeCK enhances the confidence of LLMs in edited facts. For instance, it improves the performance of LLaMA3-8B-instruct on MQuAKE by up to 219%, demonstrating its capability to strengthen ICE in the editing of stubborn knowledge. Our work paves the way to develop the both effective and accountable KE methods for LLMs. (The source code is available at: $\href{https://github.com/byronBBL/DeCK}{\text{this https URL.}}$ )
Authors: Mireia Hernandez Caralt, Clarence Boon Liang Ng, Marek Rei
Abstract: Electronic Health Records (EHR) serve as a valuable source of patient information, offering insights into medical histories, treatments, and outcomes. Previous research has developed systems for detecting applicable ICD codes that should be assigned while writing a given EHR document, mainly focusing on discharge summaries written at the end of a hospital stay. In this work, we investigate the potential of predicting these codes for the whole patient stay at different time points during their stay, even before they are officially assigned by clinicians. The development of methods to predict diagnoses and treatments earlier in advance could open opportunities for predictive medicine, such as identifying disease risks sooner, suggesting treatments, and optimizing resource allocation. Our experiments show that predictions regarding final ICD codes can be made already two days after admission and we propose a custom model that improves performance on this early prediction task.
Authors: Ghazaleh Mahmoudi, Babak Behkamkia, Sauleh Eetemadi
Abstract: Stance detection, the classification of attitudes expressed in a text towards a specific topic, is vital for applications like fake news detection and opinion mining. However, the scarcity of labeled data remains a challenge for this task. To address this problem, we propose Dynamic Model Adaptation with Contextual Data Generation (DyMoAdapt) that combines Few-Shot Learning and Large Language Models. In this approach, we aim to fine-tune an existing model at test time. We achieve this by generating new topic-specific data using GPT-3. This method could enhance performance by allowing the adaptation of the model to new topics. However, the results did not increase as we expected. Furthermore, we introduce the Multi Generated Topic VAST (MGT-VAST) dataset, which extends VAST using GPT-3. In this dataset, each context is associated with multiple topics, allowing the model to understand the relationship between contexts and various potential topics
Authors: Hadeel Saadany, Ashraf Tantawy, Constantin Orasan
Abstract: With the advent of Neural Machine Translation (NMT) systems, the MT output has reached unprecedented accuracy levels which resulted in the ubiquity of MT tools on almost all online platforms with multilingual content. However, NMT systems, like other state-of-the-art AI generative systems, are prone to errors that are deemed machine hallucinations. The problem with NMT hallucinations is that they are remarkably \textit{fluent} hallucinations. Since they are trained to produce grammatically correct utterances, NMT systems are capable of producing mistranslations that are too fluent to be recognised by both users of the MT tool, as well as by automatic quality metrics that are used to gauge their performance. In this paper, we introduce an authentic dataset of machine translation critical errors to point to the ethical and safety issues involved in the common use of MT. The dataset comprises mistranslations of Arabic mental health postings manually annotated with critical error types. We also show how the commonly used quality metrics do not penalise critical errors and highlight this as a critical issue that merits further attention from researchers.
Authors: Huawei Lin, Jikai Long, Zhaozhuo Xu, Weijie Zhao
Abstract: Given a Large Language Model (LLM) generation, how can we identify which training data led to this generation? In this paper, we proposed RapidIn, a scalable framework adapting to LLMs for estimating the influence of each training data. The proposed framework consists of two stages: caching and retrieval. First, we compress the gradient vectors by over 200,000x, allowing them to be cached on disk or in GPU/CPU memory. Then, given a generation, RapidIn efficiently traverses the cached gradients to estimate the influence within minutes, achieving over a 6,326x speedup. Moreover, RapidIn supports multi-GPU parallelization to substantially accelerate caching and retrieval. Our empirical result confirms the efficiency and effectiveness of RapidIn.
Authors: Siva Rajesh Kasa, Aniket Goel, Karan Gupta, Sumegh Roychowdhury, Anish Bhanushali, Nikhil Pattisapu, Prasanna Srinivasa Murthy
Abstract: Ordinal Classification (OC) is a widely encountered challenge in Natural Language Processing (NLP), with applications in various domains such as sentiment analysis, rating prediction, and more. Previous approaches to tackle OC have primarily focused on modifying existing or creating novel loss functions that \textbf{explicitly} account for the ordinal nature of labels. However, with the advent of Pretrained Language Models (PLMs), it became possible to tackle ordinality through the \textbf{implicit} semantics of the labels as well. This paper provides a comprehensive theoretical and empirical examination of both these approaches. Furthermore, we also offer strategic recommendations regarding the most effective approach to adopt based on specific settings.
Authors: Minghao Wu, Yulin Yuan, Gholamreza Haffari, Longyue Wang
Abstract: Recent advancements in machine translation (MT) have significantly enhanced translation quality across various domains. However, the translation of literary texts remains a formidable challenge due to their complex language, figurative expressions, and cultural nuances. In this work, we introduce a novel multi-agent framework based on large language models (LLMs) for literary translation, implemented as a company called TransAgents, which mirrors traditional translation publication process by leveraging the collective capabilities of multiple agents, to address the intricate demands of translating literary works. To evaluate the effectiveness of our system, we propose two innovative evaluation strategies: Monolingual Human Preference (MHP) and Bilingual LLM Preference (BLP). MHP assesses translations from the perspective of monolingual readers of the target language, while BLP uses advanced LLMs to compare translations directly with the original texts. Empirical findings indicate that despite lower d-BLEU scores, translations from TransAgents are preferred by both human evaluators and LLMs over human-written references, particularly in genres requiring domain-specific knowledge. We also highlight the strengths and limitations of TransAgents through case studies and suggests directions for future research.
Authors: Chris Emezue
Abstract: Structured prediction tasks, like machine translation, involve learning functions that map structured inputs to structured outputs. Recurrent Neural Networks (RNNs) have historically been a popular choice for such tasks, including in natural language processing (NLP) applications. However, training RNNs using Maximum Likelihood Estimation (MLE) has its limitations, including exposure bias and a mismatch between training and testing metrics. SEARNN, based on the learning to search (L2S) framework, has been proposed as an alternative to MLE for RNN training. This project explored the potential of SEARNN to improve machine translation for low-resourced African languages -- a challenging task characterized by limited training data availability and the morphological complexity of the languages. Through experiments conducted on translation for English to Igbo, French to \ewe, and French to \ghomala directions, this project evaluated the efficacy of SEARNN over MLE in addressing the unique challenges posed by these languages. With an average BLEU score improvement of $5.4$\% over the MLE objective, we proved that SEARNN is indeed a viable algorithm to effectively train RNNs on machine translation for low-resourced languages.
Authors: Andrew Rueda, Elena \'Alvarez Mellado, Constantine Lignos
Abstract: Modern named entity recognition systems have steadily improved performance in the age of larger and more powerful neural models. However, over the past several years, the state-of-the-art has seemingly hit another plateau on the benchmark CoNLL-03 English dataset. In this paper, we perform a deep dive into the test outputs of the highest-performing NER models, conducting a fine-grained evaluation of their performance by introducing new document-level annotations on the test set. We go beyond F1 scores by categorizing errors in order to interpret the true state of the art for NER and guide future work. We review previous attempts at correcting the various flaws of the test set and introduce CoNLL#, a new corrected version of the test set that addresses its systematic and most prevalent errors, allowing for low-noise, interpretable error analysis.
Authors: Ermo Hua, Biqing Qi, Kaiyan Zhang, Yue Yu, Ning Ding, Xingtai Lv, Kai Tian, Bowen Zhou
Abstract: Supervised Fine-Tuning (SFT) and Reinforcement Learning from Human Feedback (RLHF) are two fundamental processes for enhancing the capabilities of Language Models (LMs) post pre-training, aligning them better with human preferences. Although SFT advances in training efficiency, RLHF delivers better alignment, thus they are often combined. However, common practices simply apply them sequentially without unifying their optimization targets, resulting in a trade-off between fitting different objectives, and ignoring the opportunities to bridge the paradigm gap and take the strength from both. To obtain a unified understanding, we interpret SFT and RLHF using two sub-processes -- Preference Estimation and Transition Optimization -- defined at token level within the Markov Decision Process (MDP) framework. This modeling shows that SFT is only a specialized case of RLHF with inferior estimation and optimization. RLHF evaluates the quality of model's entire generated answer, whereas SFT only scores predicted tokens based on preceding tokens from target answers. Therefore, SFT overestimates the ability of model, leading to inferior optimization. Building on this view, we introduce Intuitive Fine-tuning (IFT) to integrate SFT and RLHF into a single process. IFT captures LMs' intuitive sense of the entire answers through a temporal residual connection, while using a single policy and the same volume of non-preference-labeled data as SFT. Our experiments show that IFT performs comparably or even superiorly to sequential recipes of SFT and some typical alignment methods across several tasks, particularly those requires generation, reasoning, and fact-following abilities. An explainable Frozen Lake game further validates the effectiveness of IFT.
Authors: Qingchen Yu, Zifan Zheng, Shichao Song, Zhiyu Li, Feiyu Xiong, Bo Tang, Ding Chen
Abstract: The continuous advancement of large language models (LLMs) has brought increasing attention to the critical issue of developing fair and reliable methods for evaluating their performance. Particularly, the emergence of subjective or non-subjective cheating phenomena, such as test set leakage and prompt format overfitting, poses significant challenges to the reliable evaluation of LLMs. Since evaluation frameworks often utilize Regular Expression (RegEx) for answer extraction, some models may adjust their responses to comply with specific formats that are easily extractable by RegEx. Nevertheless, the key answer extraction module based on RegEx frequently suffers from extraction errors. This paper conducts a comprehensive analysis of the entire LLM evaluation chain, demonstrating that optimizing the key answer extraction module can improve extraction accuracy, reduce LLMs' reliance on specific answer formats, and enhance the reliability of LLM evaluation. To address these issues, we propose xFinder, a model specifically designed for key answer extraction. As part of this process, we create a specialized dataset, the Key Answer Finder (KAF) dataset, to ensure effective model training and evaluation. Through generalization testing and evaluation in real-world scenarios, the results demonstrate that the smallest xFinder model with only 500 million parameters achieves an average answer extraction accuracy of 93.42%. In contrast, RegEx accuracy in the best evaluation framework is 74.38%. xFinder exhibits stronger robustness and higher accuracy compared to existing evaluation frameworks. All resources for xFinder are available at \url{https://github.com/IAAR-Shanghai/xFinder}.
Authors: Eduard Poesina, Cornelia Caragea, Radu Tudor Ionescu
Abstract: Natural language inference (NLI), the task of recognizing the entailment relationship in sentence pairs, is an actively studied topic serving as a proxy for natural language understanding. Despite the relevance of the task in building conversational agents and improving text classification, machine translation and other NLP tasks, to the best of our knowledge, there is no publicly available NLI corpus for the Romanian language. To this end, we introduce the first Romanian NLI corpus (RoNLI) comprising 58K training sentence pairs, which are obtained via distant supervision, and 6K validation and test sentence pairs, which are manually annotated with the correct labels. We conduct experiments with multiple machine learning methods based on distant learning, ranging from shallow models based on word embeddings to transformer-based neural networks, to establish a set of competitive baselines. Furthermore, we improve on the best model by employing a new curriculum learning strategy based on data cartography. Our dataset and code to reproduce the baselines are available https://github.com/Eduard6421/RONLI.
Authors: Zijian Feng, Hanzhang Zhou, Zixiao Zhu, Junlang Qian, Kezhi Mao
Abstract: Prompts play a crucial role in guiding the responses of Large Language Models (LLMs). However, the intricate role of individual tokens in prompts, known as input saliency, in shaping the responses remains largely underexplored. Existing saliency methods either misalign with LLM generation objectives or rely heavily on linearity assumptions, leading to potential inaccuracies. To address this, we propose Token Distribution Dynamics (TDD), a \textcolor{black}{simple yet effective} approach to unveil and manipulate the role of prompts in generating LLM outputs. TDD leverages the robust interpreting capabilities of the language model head (LM head) to assess input saliency. It projects input tokens into the embedding space and then estimates their significance based on distribution dynamics over the vocabulary. We introduce three TDD variants: forward, backward, and bidirectional, each offering unique insights into token relevance. Extensive experiments reveal that the TDD surpasses state-of-the-art baselines with a big margin in elucidating the causal relationships between prompts and LLM outputs. Beyond mere interpretation, we apply TDD to two prompt manipulation tasks for controlled text generation: zero-shot toxic language suppression and sentiment steering. Empirical results underscore TDD's proficiency in identifying both toxic and sentimental cues in prompts, subsequently mitigating toxicity or modulating sentiment in the generated content.
Authors: Rabindra Lamsal, Maria Rodriguez Read, Shanika Karunasekera, Muhammad Imran
Abstract: During times of crisis, social media platforms play a vital role in facilitating communication and coordinating resources. Amidst chaos and uncertainty, communities often rely on these platforms to share urgent pleas for help, extend support, and organize relief efforts. However, the sheer volume of conversations during such periods, which can escalate to unprecedented levels, necessitates the automated identification and matching of requests and offers to streamline relief operations. This study addresses the challenge of efficiently identifying and matching assistance requests and offers on social media platforms during emergencies. We propose CReMa (Crisis Response Matcher), a systematic approach that integrates textual, temporal, and spatial features for multi-lingual request-offer matching. By leveraging CrisisTransformers, a set of pre-trained models specific to crises, and a cross-lingual embedding space, our methodology enhances the identification and matching tasks while outperforming strong baselines such as RoBERTa, MPNet, and BERTweet, in classification tasks, and Universal Sentence Encoder, Sentence Transformers in crisis embeddings generation tasks. We introduce a novel multi-lingual dataset that simulates scenarios of help-seeking and offering assistance on social media across the 16 most commonly used languages in Australia. We conduct comprehensive cross-lingual experiments across these 16 languages, also while examining trade-offs between multiple vector search strategies and accuracy. Additionally, we analyze a million-scale geotagged global dataset to comprehend patterns in relation to seeking help and offering assistance on social media. Overall, these contributions advance the field of crisis informatics and provide benchmarks for future research in the area.
Authors: Tom Roth, Inigo Jauregi Unanue, Alsharif Abuadbba, Massimo Piccardi
Abstract: Text classifiers are vulnerable to adversarial examples -- correctly-classified examples that are deliberately transformed to be misclassified while satisfying acceptability constraints. The conventional approach to finding adversarial examples is to define and solve a combinatorial optimisation problem over a space of allowable transformations. While effective, this approach is slow and limited by the choice of transformations. An alternate approach is to directly generate adversarial examples by fine-tuning a pre-trained language model, as is commonly done for other text-to-text tasks. This approach promises to be much quicker and more expressive, but is relatively unexplored. For this reason, in this work we train an encoder-decoder paraphrase model to generate a diverse range of adversarial examples. For training, we adopt a reinforcement learning algorithm and propose a constraint-enforcing reward that promotes the generation of valid adversarial examples. Experimental results over two text classification datasets show that our model has achieved a higher success rate than the original paraphrase model, and overall has proved more effective than other competitive attacks. Finally, we show how key design choices impact the generated examples and discuss the strengths and weaknesses of the proposed approach.
Authors: Chen Huang, Yiping Jin, Ilija Ilievski, Wenqiang Lei, Jiancheng Lv
Abstract: Human annotation is a time-consuming task that requires a significant amount of effort. To address this issue, interactive data annotation utilizes an annotation model to provide suggestions for humans to approve or correct. However, annotation models trained with limited labeled data are prone to generating incorrect suggestions, leading to extra human correction effort. To tackle this challenge, we propose Araida, an analogical reasoning-based approach that enhances automatic annotation accuracy in the interactive data annotation setting and reduces the need for human corrections. Araida involves an error-aware integration strategy that dynamically coordinates an annotation model and a k-nearest neighbors (KNN) model, giving more importance to KNN's predictions when predictions from the annotation model are deemed inaccurate. Empirical studies demonstrate that Araida is adaptable to different annotation tasks and models. On average, it reduces human correction labor by 11.02% compared to vanilla interactive data annotation methods.
Authors: Kamil Guttmann, Miko{\l}aj Pokrywka, Adrian Charkiewicz, Artur Nowakowski
Abstract: This paper explores Minimum Bayes Risk (MBR) decoding for self-improvement in machine translation (MT), particularly for domain adaptation and low-resource languages. We implement the self-improvement process by fine-tuning the model on its MBR-decoded forward translations. By employing COMET as the MBR utility metric, we aim to achieve the reranking of translations that better aligns with human preferences. The paper explores the iterative application of this approach and the potential need for language-specific MBR utility metrics. The results demonstrate significant enhancements in translation quality for all examined language pairs, including successful application to domain-adapted models and generalisation to low-resource settings. This highlights the potential of COMET-guided MBR for efficient MT self-improvement in various scenarios.
Authors: Fons Hartendorp, Tom Seinen, Erik van Mulligen, Suzan Verberne
Abstract: Biomedical entity linking, a main component in automatic information extraction from health-related texts, plays a pivotal role in connecting textual entities (such as diseases, drugs and body parts mentioned by patients) to their corresponding concepts in a structured biomedical knowledge base. The task remains challenging despite recent developments in natural language processing. This paper presents the first evaluated biomedical entity linking model for the Dutch language. We use MedRoBERTa.nl as base model and perform second-phase pretraining through self-alignment on a Dutch biomedical ontology extracted from the UMLS and Dutch SNOMED. We derive a corpus from Wikipedia of ontology-linked Dutch biomedical entities in context and fine-tune our model on this dataset. We evaluate our model on the Dutch portion of the Mantra GSC-corpus and achieve 54.7% classification accuracy and 69.8% 1-distance accuracy. We then perform a case study on a collection of unlabeled, patient-support forum data and show that our model is hampered by the limited quality of the preceding entity recognition step. Manual evaluation of small sample indicates that of the correctly extracted entities, around 65% is linked to the correct concept in the ontology. Our results indicate that biomedical entity linking in a language other than English remains challenging, but our Dutch model can be used to for high-level analysis of patient-generated text.
Authors: Dawid Wi\'sniewski, Zofia Rostek, Artur Nowakowski
Abstract: People use language for various purposes. Apart from sharing information, individuals may use it to express emotions or to show respect for another person. In this paper, we focus on the formality level of machine-generated translations and present FAME-MT -- a dataset consisting of 11.2 million translations between 15 European source languages and 8 European target languages classified to formal and informal classes according to target sentence formality. This dataset can be used to fine-tune machine translation models to ensure a given formality level for each European target language considered. We describe the dataset creation procedure, the analysis of the dataset's quality showing that FAME-MT is a reliable source of language register information, and we present a publicly available proof-of-concept machine translation model that uses the dataset to steer the formality level of the translation. Currently, it is the largest dataset of formality annotations, with examples expressed in 112 European language pairs. The dataset is published online: https://github.com/laniqo-public/fame-mt/ .
Authors: Tabea M. G. Pakull, Hendrik Damm, Ahmad Idrissi-Yaghir, Henning Sch\"afer, Peter A. Horn, Christoph M. Friedrich
Abstract: This paper details the efforts of the WisPerMed team in the BioLaySumm2024 Shared Task on automatic lay summarization in the biomedical domain, aimed at making scientific publications accessible to non-specialists. Large language models (LLMs), specifically the BioMistral and Llama3 models, were fine-tuned and employed to create lay summaries from complex scientific texts. The summarization performance was enhanced through various approaches, including instruction tuning, few-shot learning, and prompt variations tailored to incorporate specific context information. The experiments demonstrated that fine-tuning generally led to the best performance across most evaluated metrics. Few-shot learning notably improved the models' ability to generate relevant and factually accurate texts, particularly when using a well-crafted prompt. Additionally, a Dynamic Expert Selection (DES) mechanism to optimize the selection of text outputs based on readability and factuality metrics was developed. Out of 54 participants, the WisPerMed team reached the 4th place, measured by readability, factuality, and relevance. Determined by the overall score, our approach improved upon the baseline by approx. 5.5 percentage points and was only approx 1.5 percentage points behind the first place.
Authors: Ziyin Zhang, Lizhen Xu, Zhaokun Jiang, Hongkun Hao, Rui Wang
Abstract: We present GSM-MC and MATH-MC, two multiple-choice (MC) datasets constructed by collecting answers and incorrect predictions on GSM8K and MATH from over 50 open-source models. Through extensive experiments, we show that LLMs' performance on the MC versions of these two popular benchmarks is strongly correlated with their performance on the original versions, and is quite robust to distractor choices and option orders, while the evaluation time is reduced by a factor of up to 30. Following a similar procedure, we also introduce PythonIO, a new program output prediction MC dataset constructed from two other popular LLM evaluation benchmarks HumanEval and MBPP. Our data and code are available at https://github.com/Geralt-Targaryen/MC-Evaluation.
Authors: Silvia Garc\'ia-M\'endez, Francisco de Arriba-P\'erez, Mar\'ia del Carmen Somoza-L\'opez
Abstract: Transformer architectures contribute to managing long-term dependencies for Natural Language Processing, representing one of the most recent changes in the field. These architectures are the basis of the innovative, cutting-edge Large Language Models (LLMs) that have produced a huge buzz in several fields and industrial sectors, among the ones education stands out. Accordingly, these generative Artificial Intelligence-based solutions have directed the change in techniques and the evolution in educational methods and contents, along with network infrastructure, towards high-quality learning. Given the popularity of LLMs, this review seeks to provide a comprehensive overview of those solutions designed specifically to generate and evaluate educational materials and which involve students and teachers in their design or experimental plan. To the best of our knowledge, this is the first review of educational applications (e.g., student assessment) of LLMs. As expected, the most common role of these systems is as virtual tutors for automatic question generation. Moreover, the most popular models are GTP-3 and BERT. However, due to the continuous launch of new generative models, new works are expected to be published shortly.
Authors: Saadia Gabriel, Isha Puri, Xuhai Xu, Matteo Malgaroli, Marzyeh Ghassemi
Abstract: Large language models (LLMs) are already being piloted for clinical use in hospital systems like NYU Langone, Dana-Farber and the NHS. A proposed deployment use case is psychotherapy, where a LLM-powered chatbot can treat a patient undergoing a mental health crisis. Deployment of LLMs for mental health response could hypothetically broaden access to psychotherapy and provide new possibilities for personalizing care. However, recent high-profile failures, like damaging dieting advice offered by the Tessa chatbot to patients with eating disorders, have led to doubt about their reliability in high-stakes and safety-critical settings. In this work, we develop an evaluation framework for determining whether LLM response is a viable and ethical path forward for the automation of mental health treatment. Using human evaluation with trained clinicians and automatic quality-of-care metrics grounded in psychology research, we compare the responses provided by peer-to-peer responders to those provided by a state-of-the-art LLM. We show that LLMs like GPT-4 use implicit and explicit cues to infer patient demographics like race. We then show that there are statistically significant discrepancies between patient subgroups: Responses to Black posters consistently have lower empathy than for any other demographic group (2%-13% lower than the control group). Promisingly, we do find that the manner in which responses are generated significantly impacts the quality of the response. We conclude by proposing safety guidelines for the potential deployment of LLMs for mental health response.
Authors: Olga Kellert, Carlos G\'omez-Rodr\'iguez, Mahmud Uz Zaman
Abstract: TripAdvisor reviews and comparable data sources play an important role in many tasks in Natural Language Processing (NLP), providing a data basis for the identification and classification of subjective judgments, such as hotel or restaurant reviews, into positive or negative polarities. This study explores three important factors influencing variation in crowdsourced polarity judgments, focusing on TripAdvisor reviews in Spanish. Three hypotheses are tested: the role of Part Of Speech (POS), the impact of sentiment words such as "tasty", and the influence of neutral words like "ok" on judgment variation. The study's methodology employs one-word titles, demonstrating their efficacy in studying polarity variation of words. Statistical tests on mean equality are performed on word groups of our interest. The results of this study reveal that adjectives in one-word titles tend to result in lower judgment variation compared to other word types or POS. Sentiment words contribute to lower judgment variation as well, emphasizing the significance of sentiment words in research on polarity judgments, and neutral words are associated with higher judgment variation as expected. However, these effects cannot be always reproduced in longer titles, which suggests that longer titles do not represent the best data source for testing the ambiguity of single words due to the influence on word polarity by other words like negation in longer titles. This empirical investigation contributes valuable insights into the factors influencing polarity variation of words, providing a foundation for NLP practitioners that aim to capture and predict polarity judgments in Spanish and for researchers that aim to understand factors influencing judgment variation.
Authors: Yue Chen, Chen Huang, Yang Deng, Wenqiang Lei, Dingnan Jin, Jia Liu, Tat-Seng Chua
Abstract: Equipping a conversational search engine with strategies regarding when to ask clarification questions is becoming increasingly important across various domains. Attributing to the context understanding capability of LLMs and their access to domain-specific sources of knowledge, LLM-based clarification strategies feature rapid transfer to various domains in a post-hoc manner. However, they still struggle to deliver promising performance on unseen domains, struggling to achieve effective domain transferability. We take the first step to investigate this issue and existing methods tend to produce one-size-fits-all strategies across diverse domains, limiting their search effectiveness. In response, we introduce a novel method, called Style, to achieve effective domain transferability. Our experimental results indicate that Style bears strong domain transferability, resulting in an average search performance improvement of ~10% on four unseen domains.
Authors: Tong Zhang, Peixin Qin, Yang Deng, Chen Huang, Wenqiang Lei, Junhong Liu, Dingnan Jin, Hongru Liang, Tat-Seng Chua
Abstract: Large language models (LLMs) are increasingly used to meet user information needs, but their effectiveness in dealing with user queries that contain various types of ambiguity remains unknown, ultimately risking user trust and satisfaction. To this end, we introduce CLAMBER, a benchmark for evaluating LLMs using a well-organized taxonomy. Building upon the taxonomy, we construct ~12K high-quality data to assess the strengths, weaknesses, and potential risks of various off-the-shelf LLMs. Our findings indicate the limited practical utility of current LLMs in identifying and clarifying ambiguous user queries, even enhanced by chain-of-thought (CoT) and few-shot prompting. These techniques may result in overconfidence in LLMs and yield only marginal enhancements in identifying ambiguity. Furthermore, current LLMs fall short in generating high-quality clarifying questions due to a lack of conflict resolution and inaccurate utilization of inherent knowledge. In this paper, CLAMBER presents a guidance and promotes further research on proactive and trustworthy LLMs. Our dataset is available at https://github.com/zt991211/CLAMBER
Authors: Chen Huang, Yang Deng, Wenqiang Lei, Jiancheng Lv, Ido Dagan
Abstract: To obtain high-quality annotations under limited budget, semi-automatic annotation methods are commonly used, where a portion of the data is annotated by experts and a model is then trained to complete the annotations for the remaining data. However, these methods mainly focus on selecting informative data for expert annotations to improve the model predictive ability (i.e., triage-to-human data), while the rest of the data is indiscriminately assigned to model annotation (i.e., triage-to-model data). This may lead to inefficiencies in budget allocation for annotations, as easy data that the model could accurately annotate may be unnecessarily assigned to the expert, and hard data may be misclassified by the model. As a result, the overall annotation quality may be compromised. To address this issue, we propose a selective annotation framework called SANT. It effectively takes advantage of both the triage-to-human and triage-to-model data through the proposed error-aware triage and bi-weighting mechanisms. As such, informative or hard data is assigned to the expert for annotation, while easy data is handled by the model. Experimental results show that SANT consistently outperforms other baselines, leading to higher-quality annotation through its proper allocation of data to both expert and model workers. We provide pioneering work on data annotation within budget constraints, establishing a landmark for future triage-based annotation studies.
Authors: Jumbly Grindrod, J. D. Porter, Nat Hansen
Abstract: Current language models are built on the so-called distributional semantic approach to linguistic meaning that has the distributional hypothesis at its core. The distributional hypothesis involves a holistic conception of word meaning: the meaning of a word depends upon its relations to other words in the model. A standard objection to meaning holism is the charge of instability: any change in the meaning properties of a linguistic system (a human speaker, for example) would lead to many changes or possibly a complete change in the entire system. When the systems in question are trying to communicate with each other, it has been argued that instability of this kind makes communication impossible (Fodor and Lepore 1992, 1996, 1999). In this article, we examine whether the instability objection poses a problem for distributional models of meaning. First, we distinguish between distinct forms of instability that these models could exhibit, and we argue that only one such form is relevant for understanding the relation between instability and communication: what we call differential instability. Differential instability is variation in the relative distances between points in a space, rather than variation in the absolute position of those points. We distinguish differential and absolute instability by constructing two of our own models, a toy model constructed from the text of two novels, and a more sophisticated model constructed using the Word2vec algorithm from a combination of Wikipedia and SEP articles. We demonstrate the two forms of instability by showing how these models change as the corpora they are constructed from increase in size.
Authors: Hao Chen, Biaojie Zeng, Xin Lin, Liang He, Aimin Zhou
Abstract: Math world problems correction(MWPC) is a novel task dedicated to rectifying reasoning errors in the process of solving mathematical problems. In this paper, leveraging the advancements in large language models (LLMs), we address two key objectives:(1) Distinguishing between mathematical reasoning and error correction; (2) Exploring strategies to enhance the error correction capabilities of LLMs in mathematics to solve MWPC task. We noticed that, in real-time education,assisting students in recognizing their mistakes is more crucial than simply providing correct answers. However, current research tends to prioritize obtaining accurate solutions to math problems rather than correcting potentially incorrect ones. Therefore, we modify the research paradigm, demonstrating that improving mathematical reasoning abilities does not equate to mastery in error correction. Meanwhile, we propose a novel method called diagnostic-oriented promping(DOP) aimed at facilitating LLMs to excel in error correction. In experiments, DOP has shown outstanding performance, highlighting its significant impact. We argue that in mathematical education, the demand for outstanding correctors surpasses that for proficient reasoners. Codes and data are available on https://github.com/ChenhaoEcnuCS/Reason-Correct.
Authors: Richard Futrell, Michael Hahn
Abstract: Human language is a unique form of communication in the natural world, distinguished by its structured nature. Most fundamentally, it is systematic, meaning that signals can be broken down into component parts that are individually meaningful -- roughly, words -- which are combined in a regular way to form sentences. Furthermore, the way in which these parts are combined maintains a kind of locality: words are usually concatenated together, and they form contiguous phrases, keeping related parts of sentences close to each other. We address the challenge of understanding how these basic properties of language arise from broader principles of efficient communication under information processing constraints. Here we show that natural-language-like systematicity arises from minimization of excess entropy, a measure of statistical complexity that represents the minimum amount of information necessary for predicting the future of a sequence based on its past. In simulations, we show that codes that minimize excess entropy factorize their source distributions into approximately independent components, and then express those components systematically and locally. Next, in a series of massively cross-linguistic corpus studies, we show that human languages are structured to have low excess entropy at the level of phonology, morphology, syntax, and semantics. Our result suggests that human language performs a sequential generalization of Independent Components Analysis on the statistical distribution over meanings that need to be expressed. It establishes a link between the statistical and algebraic structure of human language, and reinforces the idea that the structure of human language may have evolved to minimize cognitive load while maximizing communicative expressiveness.
Authors: Ting Jiang, Shaohan Huang, Shengyue Luo, Zihan Zhang, Haizhen Huang, Furu Wei, Weiwei Deng, Feng Sun, Qi Zhang, Deqing Wang, Fuzhen Zhuang
Abstract: Low-rank adaptation is a popular parameter-efficient fine-tuning method for large language models. In this paper, we analyze the impact of low-rank updating, as implemented in LoRA. Our findings suggest that the low-rank updating mechanism may limit the ability of LLMs to effectively learn and memorize new knowledge. Inspired by this observation, we propose a new method called MoRA, which employs a square matrix to achieve high-rank updating while maintaining the same number of trainable parameters. To achieve it, we introduce the corresponding non-parameter operators to reduce the input dimension and increase the output dimension for the square matrix. Furthermore, these operators ensure that the weight can be merged back into LLMs, which makes our method can be deployed like LoRA. We perform a comprehensive evaluation of our method across five tasks: instruction tuning, mathematical reasoning, continual pretraining, memory and pretraining. Our method outperforms LoRA on memory-intensive tasks and achieves comparable performance on other tasks.
Authors: Xiaobo Liang, Haoke Zhang, Helan hu, Juntao Li, Jun Xu, Min Zhang
Abstract: The rapid advancement of large language models has given rise to a plethora of applications across a myriad of real-world tasks, mainly centered on aligning with human intent. However, the complexities inherent in human intent necessitate a dependence on labor-intensive and time-consuming human evaluation. To alleviate this constraint, we delve into the paradigm of employing open-source large language models as evaluators, aligning with the prevailing trend of utilizing GPT-4. Particularly, we present a step-by-step evaluation framework: \textbf{Fennec}, capable of \textbf{F}ine-grained \textbf{E}valuatio\textbf{N} and correctio\textbf{N} \textbf{E}xtended through bran\textbf{C}hing and bridging. Specifically, the branching operation dissects the evaluation task into various dimensions and granularities, thereby alleviating the challenges associated with evaluation. Concurrently, the bridging operation amalgamates diverse training datasets, augmenting the variety of evaluation tasks. In experimental trials, our 7B model consistently outperforms open-source larger-scale evaluation models across various widely adopted benchmarks in terms of both \textit{Agreement} and \textit{Consistency}, closely approaching the capabilities of GPT-4. We employ the fine-grained correction capabilities induced by the evaluation model to refine multiple model responses, and the results show that the refinement elevates the quality of responses, leading to an improvement of 1-2 points on the MT-Bench. Our code is available at Github\footnote{\url{https://github.com/dropreg/Fennec}}.
Authors: Haoxiang Shi, Jiaan Wang, Jiarong Xu, Cen Wang, Tetsuya Sakai
Abstract: Text-to-Table aims to generate structured tables to convey the key information from unstructured documents. Existing text-to-table datasets are typically oriented English, limiting the research in non-English languages. Meanwhile, the emergence of large language models (LLMs) has shown great success as general task solvers in multi-lingual settings (e.g., ChatGPT), theoretically enabling text-to-table in other languages. In this paper, we propose a Chinese text-to-table dataset, CT-Eval, to benchmark LLMs on this task. Our preliminary analysis of English text-to-table datasets highlights two key factors for dataset construction: data diversity and data hallucination. Inspired by this, the CT-Eval dataset selects a popular Chinese multidisciplinary online encyclopedia as the source and covers 28 domains to ensure data diversity. To minimize data hallucination, we first train an LLM to judge and filter out the task samples with hallucination, then employ human annotators to clean the hallucinations in the validation and testing sets. After this process, CT-Eval contains 88.6K task samples. Using CT-Eval, we evaluate the performance of open-source and closed-source LLMs. Our results reveal that zero-shot LLMs (including GPT-4) still have a significant performance gap compared with human judgment. Furthermore, after fine-tuning, open-source LLMs can significantly improve their text-to-table ability, outperforming GPT-4 by a large margin. In short, CT-Eval not only helps researchers evaluate and quickly understand the Chinese text-to-table ability of existing LLMs but also serves as a valuable resource to significantly improve the text-to-table performance of LLMs.
Authors: Tong Zeng, Daniel E. Acuna
Abstract: Scientist learn early on how to cite scientific sources to support their claims. Sometimes, however, scientists have challenges determining where a citation should be situated -- or, even worse, fail to cite a source altogether. Automatically detecting sentences that need a citation (i.e., citation worthiness) could solve both of these issues, leading to more robust and well-constructed scientific arguments. Previous researchers have applied machine learning to this task but have used small datasets and models that do not take advantage of recent algorithmic developments such as attention mechanisms in deep learning. We hypothesize that we can develop significantly accurate deep learning architectures that learn from large supervised datasets constructed from open access publications. In this work, we propose a Bidirectional Long Short-Term Memory (BiLSTM) network with attention mechanism and contextual information to detect sentences that need citations. We also produce a new, large dataset (PMOA-CITE) based on PubMed Open Access Subset, which is orders of magnitude larger than previous datasets. Our experiments show that our architecture achieves state of the art performance on the standard ACL-ARC dataset ($F_{1}=0.507$) and exhibits high performance ($F_{1}=0.856$) on the new PMOA-CITE. Moreover, we show that it can transfer learning across these datasets. We further use interpretable models to illuminate how specific language is used to promote and inhibit citations. We discover that sections and surrounding sentences are crucial for our improved predictions. We further examined purported mispredictions of the model, and uncovered systematic human mistakes in citation behavior and source data. This opens the door for our model to check documents during pre-submission and pre-archival procedures. We make this new dataset, the code, and a web-based tool available to the community.
Authors: Hongwei Liu, Zilong Zheng, Yuxuan Qiao, Haodong Duan, Zhiwei Fei, Fengzhe Zhou, Wenwei Zhang, Songyang Zhang, Dahua Lin, Kai Chen
Abstract: Recent advancements in large language models (LLMs) have showcased significant improvements in mathematics. However, traditional math benchmarks like GSM8k offer a unidimensional perspective, falling short in providing a holistic assessment of the LLMs' math capabilities. To address this gap, we introduce MathBench, a new benchmark that rigorously assesses the mathematical capabilities of large language models. MathBench spans a wide range of mathematical disciplines, offering a detailed evaluation of both theoretical understanding and practical problem-solving skills. The benchmark progresses through five distinct stages, from basic arithmetic to college mathematics, and is structured to evaluate models at various depths of knowledge. Each stage includes theoretical questions and application problems, allowing us to measure a model's mathematical proficiency and its ability to apply concepts in practical scenarios. MathBench aims to enhance the evaluation of LLMs' mathematical abilities, providing a nuanced view of their knowledge understanding levels and problem solving skills in a bilingual context. The project is released at https://github.com/open-compass/MathBench .
Authors: Xin Du, Lixin Xiu, Kumiko Tanaka-Ishii
Abstract: We apply an information-theoretic perspective to reconsider generative document retrieval (GDR), in which a document $x \in X$ is indexed by $t \in T$, and a neural autoregressive model is trained to map queries $Q$ to $T$. GDR can be considered to involve information transmission from documents $X$ to queries $Q$, with the requirement to transmit more bits via the indexes $T$. By applying Shannon's rate-distortion theory, the optimality of indexing can be analyzed in terms of the mutual information, and the design of the indexes $T$ can then be regarded as a {\em bottleneck} in GDR. After reformulating GDR from this perspective, we empirically quantify the bottleneck underlying GDR. Finally, using the NQ320K and MARCO datasets, we evaluate our proposed bottleneck-minimal indexing method in comparison with various previous indexing methods, and we show that it outperforms those methods.
Authors: Ruizhe Chen, Tianxiang Hu, Yang Feng, Zuozhu Liu
Abstract: Concerns regarding Large Language Models (LLMs) to memorize and disclose private information, particularly Personally Identifiable Information (PII), become prominent within the community. Many efforts have been made to mitigate the privacy risks. However, the mechanism through which LLMs memorize PII remains poorly understood. To bridge this gap, we introduce a pioneering method for pinpointing PII-sensitive neurons (privacy neurons) within LLMs. Our method employs learnable binary weight masks to localize specific neurons that account for the memorization of PII in LLMs through adversarial training. Our investigations discover that PII is memorized by a small subset of neurons across all layers, which shows the property of PII specificity. Furthermore, we propose to validate the potential in PII risk mitigation by deactivating the localized privacy neurons. Both quantitative and qualitative experiments demonstrate the effectiveness of our neuron localization algorithm.
Authors: Oliver Kramer
Abstract: Large Language Models (LLMs) exhibit world knowledge and inference capabilities, making them powerful tools for various applications. This paper proposes a feedback loop mechanism that leverages these capabilities to tune Evolution Strategies (ES) parameters effectively. The mechanism involves a structured process of providing programming instructions, executing the corresponding code, and conducting thorough analysis. This process is specifically designed for the optimization of ES parameters. The method operates through an iterative cycle, ensuring continuous refinement of the ES parameters. First, LLMs process the instructions to generate or modify the code. The code is then executed, and the results are meticulously logged. Subsequent analysis of these results provides insights that drive further improvements. An experiment on tuning the learning rates of ES using the LLaMA3 model demonstrate the feasibility of this approach. This research illustrates how LLMs can be harnessed to improve ES algorithms' performance and suggests broader applications for similar feedback loop mechanisms in various domains.
Authors: Kamila Barylska, Anna Gogoli\'nska
Abstract: Diabetes is a chronic condition, considered one of the civilization diseases, that is characterized by sustained high blood sugar levels. There is no doubt that more and more people is going to suffer from diabetes, hence it is crucial to understand better its biological foundations. The essential processes related to the control of glucose levels in the blood are: glycolysis (process of breaking down of glucose) and glucose synthesis, both taking place in the liver. The glycolysis occurs during feeding and it is stimulated by insulin. On the other hand, the glucose synthesis arises during fasting and it is stimulated by glucagon. In the paper we present a Petri net model of glycolysis and glucose synthesis in the liver. The model is created based on medical literature. Standard Petri nets techniques are used to analyse the properties of the model: traps, reachability graphs, tokens dynamics, deadlocks analysis. The results are described in the paper. Our analysis shows that the model captures the interactions between different enzymes and substances, which is consistent with the biological processes occurring during fasting and feeding. The model constitutes the first element of our long-time goal to create the whole body model of the glucose regulation in a healthy human and a person with diabetes.
Authors: Sandeep Singh Sengar, Affan Bin Hasan, Sanjay Kumar, Fiona Carroll
Abstract: In recent years, the study of artificial intelligence (AI) has undergone a paradigm shift. This has been propelled by the groundbreaking capabilities of generative models both in supervised and unsupervised learning scenarios. Generative AI has shown state-of-the-art performance in solving perplexing real-world conundrums in fields such as image translation, medical diagnostics, textual imagery fusion, natural language processing, and beyond. This paper documents the systematic review and analysis of recent advancements and techniques in Generative AI with a detailed discussion of their applications including application-specific models. Indeed, the major impact that generative AI has made to date, has been in language generation with the development of large language models, in the field of image translation and several other interdisciplinary applications of generative AI. Moreover, the primary contribution of this paper lies in its coherent synthesis of the latest advancements in these areas, seamlessly weaving together contemporary breakthroughs in the field. Particularly, how it shares an exploration of the future trajectory for generative AI. In conclusion, the paper ends with a discussion of Responsible AI principles, and the necessary ethical considerations for the sustainability and growth of these generative models.
Authors: Karan Taneja, Pratyusha Maiti, Sandeep Kakar, Pranav Guruprasad, Sanjeev Rao, Ashok K. Goel
Abstract: Conversational AI agents often require extensive datasets for training that are not publicly released, are limited to social chit-chat or handling a specific domain, and may not be easily extended to accommodate the latest advances in AI technologies. This paper introduces Jill Watson, a conversational Virtual Teaching Assistant (VTA) leveraging the capabilities of ChatGPT. Jill Watson based on ChatGPT requires no prior training and uses a modular design to allow the integration of new APIs using a skill-based architecture inspired by XiaoIce. Jill Watson is also well-suited for intelligent textbooks as it can process and converse using multiple large documents. We exclusively utilize publicly available resources for reproducibility and extensibility. Comparative analysis shows that our system outperforms the legacy knowledge-based Jill Watson as well as the OpenAI Assistants service. We employ many safety measures that reduce instances of hallucinations and toxicity. The paper also includes real-world examples from a classroom setting that demonstrate different features of Jill Watson and its effectiveness.
Authors: David Xu
Abstract: Multi-modal learning in the audio-language domain has seen significant advancements in recent years. However, audio-language learning faces challenges due to limited and lower-quality data compared to image-language tasks. Existing audio-language datasets are notably smaller, and manual labeling is hindered by the need to listen to entire audio clips for accurate labeling. Our method systematically generates audio-caption pairs by augmenting audio clips with natural language labels and corresponding audio signal processing operations. Leveraging a Large Language Model, we generate descriptions of augmented audio clips with a prompt template. This scalable method produces AudioSetMix, a high-quality training dataset for text-and-audio related models. Integration of our dataset improves models performance on benchmarks by providing diversified and better-aligned examples. Notably, our dataset addresses the absence of modifiers (adjectives and adverbs) in existing datasets. By enabling models to learn these concepts, and generating hard negative examples during training, we achieve state-of-the-art performance on multiple benchmarks.
Authors: Jos\'e Luiz Nunes, Guilherme F. C. F. Almeida, Marcelo de Araujo, Simone D. J. Barbosa
Abstract: Large language models (LLMs) have taken centre stage in debates on Artificial Intelligence. Yet there remains a gap in how to assess LLMs' conformity to important human values. In this paper, we investigate whether state-of-the-art LLMs, GPT-4 and Claude 2.1 (Gemini Pro and LLAMA 2 did not generate valid results) are moral hypocrites. We employ two research instruments based on the Moral Foundations Theory: (i) the Moral Foundations Questionnaire (MFQ), which investigates which values are considered morally relevant in abstract moral judgements; and (ii) the Moral Foundations Vignettes (MFVs), which evaluate moral cognition in concrete scenarios related to each moral foundation. We characterise conflicts in values between these different abstractions of moral evaluation as hypocrisy. We found that both models displayed reasonable consistency within each instrument compared to humans, but they displayed contradictory and hypocritical behaviour when we compared the abstract values present in the MFQ to the evaluation of concrete moral violations of the MFV.
Authors: Chuanneng Sun, Songjun Huang, Dario Pompili
Abstract: In recent years, Large Language Models (LLMs) have shown great abilities in various tasks, including question answering, arithmetic problem solving, and poem writing, among others. Although research on LLM-as-an-agent has shown that LLM can be applied to Reinforcement Learning (RL) and achieve decent results, the extension of LLM-based RL to Multi-Agent System (MAS) is not trivial, as many aspects, such as coordination and communication between agents, are not considered in the RL frameworks of a single agent. To inspire more research on LLM-based MARL, in this letter, we survey the existing LLM-based single-agent and multi-agent RL frameworks and provide potential research directions for future research. In particular, we focus on the cooperative tasks of multiple agents with a common goal and communication among them. We also consider human-in/on-the-loop scenarios enabled by the language component in the framework.
Authors: Aloni Cohen, Alexander Hoover, Gabe Schoenbach
Abstract: A zero-bit watermarked language model produces text that is indistinguishable from that of the underlying model, but which can be detected as machine-generated using a secret key. But merely detecting AI-generated spam, say, as watermarked may not prevent future abuses. If we could additionally trace the text to a spammer's API token, we could then cut off their access to the model. We introduce multi-user watermarks, which allow tracing model-generated text to individuals or to groups of colluding users. We construct multi-user watermarking schemes from undetectable zero-bit watermarking schemes. Importantly, our schemes provide both zero-bit and multi-user assurances at the same time: detecting shorter snippets as well as the original scheme and tracing longer excerpts to individuals. Along the way, we give a generic construction of a watermarking scheme that embeds long messages into generated text. Ours are the first black-box reductions between watermarking schemes for language models. A major challenge for black-box reductions is the lack of a unified abstraction for robustness -- that marked text is detectable after edits. Existing works give incomparable robustness guarantees, based on bespoke requirements on the language model's outputs and the users' edits. We introduce a new abstraction -- AEB-robustness -- to overcome this challenge. AEB-robustness provides that the watermark is detectable whenever the edited text "approximates enough blocks" of model-generated output. Specifying the robustness condition amounts to defining approximates, enough, and blocks. Using our new abstraction, we relate the robustness properties of our constructions to that of the underlying zero-bit scheme. Whereas prior works only guarantee robustness for a single text generated in response to a single prompt, our schemes are robust against adaptive prompting, a stronger adversarial model.
Authors: Jian Hu, Xibin Wu, Weixun Wang, Xianyu, Dehao Zhang, Yu Cao
Abstract: As large language models (LLMs) continue to grow by scaling laws, reinforcement learning from human feedback (RLHF) has gained significant attention due to its outstanding performance. However, unlike pretraining or fine-tuning a single model, scaling reinforcement learning from human feedback (RLHF) for training large language models poses coordination challenges across four models. We present OpenRLHF, an open-source framework enabling efficient RLHF scaling. Unlike existing RLHF frameworks that co-locate four models on the same GPUs, OpenRLHF re-designs scheduling for the models beyond 70B parameters using Ray, vLLM, and DeepSpeed, leveraging improved resource utilization and diverse training approaches. Integrating seamlessly with Hugging Face, OpenRLHF provides an out-of-the-box solution with optimized algorithms and launch scripts, which ensures user-friendliness. OpenRLHF implements RLHF, DPO, rejection sampling, and other alignment techniques. Empowering state-of-the-art LLM development, OpenRLHF's code is available at https://github.com/OpenLLMAI/OpenRLHF.
Authors: Oleksiy Ostapenko, Zhan Su, Edoardo Maria Ponti, Laurent Charlin, Nicolas Le Roux, Matheus Pereira, Lucas Caccia, Alessandro Sordoni
Abstract: The growing number of parameter-efficient adaptations of a base large language model (LLM) calls for studying whether we can reuse such trained adapters to improve performance for new tasks. We study how to best build a library of adapters given multi-task data and devise techniques for both zero-shot and supervised task generalization through routing in such library. We benchmark existing approaches to build this library and introduce model-based clustering, MBC, a method that groups tasks based on the similarity of their adapter parameters, indirectly optimizing for transfer across the multi-task dataset. To re-use the library, we present a novel zero-shot routing mechanism, Arrow, which enables dynamic selection of the most relevant adapters for new inputs without the need for retraining. We experiment with several LLMs, such as Phi-2 and Mistral, on a wide array of held-out tasks, verifying that MBC-based adapters and Arrow routing lead to superior generalization to new tasks. We make steps towards creating modular, adaptable LLMs that can match or outperform traditional joint training.
Authors: Mohit Tomar, Abhisek Tiwari, Sriparna Saha
Abstract: With the advancement of internet communication and telemedicine, people are increasingly turning to the web for various healthcare activities. With an ever-increasing number of diseases and symptoms, diagnosing patients becomes challenging. In this work, we build a diagnosis assistant to assist doctors, which identifies diseases based on patient-doctor interaction. During diagnosis, doctors utilize both symptomatology knowledge and diagnostic experience to identify diseases accurately and efficiently. Inspired by this, we investigate the role of medical knowledge in disease diagnosis through doctor-patient interaction. We propose a two-channel, knowledge-infused, discourse-aware disease diagnosis model (KI-DDI), where the first channel encodes patient-doctor communication using a transformer-based encoder, while the other creates an embedding of symptom-disease using a graph attention network (GAT). In the next stage, the conversation and knowledge graph embeddings are infused together and fed to a deep neural network for disease identification. Furthermore, we first develop an empathetic conversational medical corpus comprising conversations between patients and doctors, annotated with intent and symptoms information. The proposed model demonstrates a significant improvement over the existing state-of-the-art models, establishing the crucial roles of (a) a doctor's effort for additional symptom extraction (in addition to patient self-report) and (b) infusing medical knowledge in identifying diseases effectively. Many times, patients also show their medical conditions, which acts as crucial evidence in diagnosis. Therefore, integrating visual sensory information would represent an effective avenue for enhancing the capabilities of diagnostic assistants.
Authors: Biao Yi, Sishuo Chen, Yiming Li, Tong Li, Baolei Zhang, Zheli Liu
Abstract: Backdoor attacks pose an increasingly severe security threat to Deep Neural Networks (DNNs) during their development stage. In response, backdoor sample purification has emerged as a promising defense mechanism, aiming to eliminate backdoor triggers while preserving the integrity of the clean content in the samples. However, existing approaches have been predominantly focused on the word space, which are ineffective against feature-space triggers and significantly impair performance on clean data. To address this, we introduce a universal backdoor defense that purifies backdoor samples in the activation space by drawing abnormal activations towards optimized minimum clean activation distribution intervals. The advantages of our approach are twofold: (1) By operating in the activation space, our method captures from surface-level information like words to higher-level semantic concepts such as syntax, thus counteracting diverse triggers; (2) the fine-grained continuous nature of the activation space allows for more precise preservation of clean content while removing triggers. Furthermore, we propose a detection module based on statistical information of abnormal activations, to achieve a better trade-off between clean accuracy and defending performance.
Authors: Yunxin Li, Shenyuan Jiang, Baotian Hu, Longyue Wang, Wanqi Zhong, Wenhan Luo, Lin Ma, Min Zhang
Abstract: Recent advancements in Multimodal Large Language Models (MLLMs) underscore the significance of scalable models and data to boost performance, yet this often incurs substantial computational costs. Although the Mixture of Experts (MoE) architecture has been employed to efficiently scale large language and image-text models, these efforts typically involve fewer experts and limited modalities. To address this, our work presents the pioneering attempt to develop a unified MLLM with the MoE architecture, named Uni-MoE that can handle a wide array of modalities. Specifically, it features modality-specific encoders with connectors for a unified multimodal representation. We also implement a sparse MoE architecture within the LLMs to enable efficient training and inference through modality-level data parallelism and expert-level model parallelism. To enhance the multi-expert collaboration and generalization, we present a progressive training strategy: 1) Cross-modality alignment using various connectors with different cross-modality data, 2) Training modality-specific experts with cross-modality instruction data to activate experts' preferences, and 3) Tuning the Uni-MoE framework utilizing Low-Rank Adaptation (LoRA) on mixed multimodal instruction data. We evaluate the instruction-tuned Uni-MoE on a comprehensive set of multimodal datasets. The extensive experimental results demonstrate Uni-MoE's principal advantage of significantly reducing performance bias in handling mixed multimodal datasets, alongside improved multi-expert collaboration and generalization. Our findings highlight the substantial potential of MoE frameworks in advancing MLLMs and the code is available at https://github.com/HITsz-TMG/UMOE-Scaling-Unified-Multimodal-LLMs.
URLs: https://github.com/HITsz-TMG/UMOE-Scaling-Unified-Multimodal-LLMs.
Authors: Manuel E. Lladser, Alexander J. Paradise
Abstract: A subset of points in a metric space is said to resolve it if each point in the space is uniquely characterized by its distance to each point in the subset. In particular, resolving sets can be used to represent points in abstract metric spaces as Euclidean vectors. Importantly, due to the triangle inequality, points close by in the space are represented as vectors with similar coordinates, which may find applications in classification problems of symbolic objects under suitably chosen metrics. In this manuscript, we address the resolvability of Jaccard spaces, i.e., metric spaces of the form $(2^X,\text{Jac})$, where $2^X$ is the power set of a finite set $X$, and $\text{Jac}$ is the Jaccard distance between subsets of $X$. Specifically, for different $a,b\in 2^X$, $\text{Jac}(a,b)=\frac{|a\Delta b|}{|a\cup b|}$, where $|\cdot|$ denotes size (i.e., cardinality) and $\Delta$ denotes the symmetric difference of sets. We combine probabilistic and linear algebra arguments to construct highly likely but nearly optimal (i.e., of minimal size) resolving sets of $(2^X,\text{Jac})$. In particular, we show that the metric dimension of $(2^X,\text{Jac})$, i.e., the minimum size of a resolving set of this space, is $\Theta(|X|/\ln|X|)$.
Authors: Chiyu Zhang, Yifei Sun, Minghao Wu, Jun Chen, Jie Lei, Muhammad Abdul-Mageed, Rong Jin, Angli Liu, Ji Zhu, Sem Park, Ning Yao, Bo Long
Abstract: Content-based recommendation systems play a crucial role in delivering personalized content to users in the digital world. In this work, we introduce EmbSum, a novel framework that enables offline pre-computations of users and candidate items while capturing the interactions within the user engagement history. By utilizing the pretrained encoder-decoder model and poly-attention layers, EmbSum derives User Poly-Embedding (UPE) and Content Poly-Embedding (CPE) to calculate relevance scores between users and candidate items. EmbSum actively learns the long user engagement histories by generating user-interest summary with supervision from large language model (LLM). The effectiveness of EmbSum is validated on two datasets from different domains, surpassing state-of-the-art (SoTA) methods with higher accuracy and fewer parameters. Additionally, the model's ability to generate summaries of user interests serves as a valuable by-product, enhancing its usefulness for personalized content recommendations.
Authors: Hui Zheng, Hai-Teng Wang, Wei-Bang Jiang, Zhong-Tao Chen, Li He, Pei-Yang Lin, Peng-Hu Wei, Guo-Guang Zhao, Yun-Zhe Liu
Abstract: Invasive brain-computer interfaces have garnered significant attention due to their high performance. The current intracranial stereoElectroEncephaloGraphy (sEEG) foundation models typically build univariate representations based on a single channel. Some of them further use Transformer to model the relationship among channels. However, due to the locality and specificity of brain computation, their performance on more difficult tasks, e.g., speech decoding, which demands intricate processing in specific brain regions, is yet to be fully investigated. We hypothesize that building multi-variate representations within certain brain regions can better capture the specific neural processing. To explore this hypothesis, we collect a well-annotated Chinese word-reading sEEG dataset, targeting language-related brain networks, over 12 subjects. Leveraging this benchmark dataset, we developed the Du-IN model that can extract contextual embeddings from specific brain regions through discrete codebook-guided mask modeling. Our model achieves SOTA performance on the downstream 61-word classification task, surpassing all baseline models. Model comparison and ablation analysis reveal that our design choices, including (i) multi-variate representation by fusing channels in vSMC and STG regions and (ii) self-supervision by discrete codebook-guided mask modeling, significantly contribute to these performances. Collectively, our approach, inspired by neuroscience findings, capitalizing on multi-variate neural representation from specific brain regions, is suitable for invasive brain modeling. It marks a promising neuro-inspired AI approach in BCI.
Authors: Gengchen Wei, Xinle Pang, Tianning Zhang, Yu Sun, Xun Qian, Chen Lin, Han-Sen Zhong, Wanli Ouyang
Abstract: With over 200 million published academic documents and millions of new documents being written each year, academic researchers face the challenge of searching for information within this vast corpus. However, existing retrieval systems struggle to understand the semantics and domain knowledge present in academic papers. In this work, we demonstrate that by utilizing large language models, a document retrieval system can achieve advanced semantic understanding capabilities, significantly outperforming existing systems. Our approach involves training the retriever and reranker using domain-specific data generated by large language models. Additionally, we utilize large language models to identify candidates from the references of retrieved papers to further enhance the performance. We use a test set annotated by academic researchers in the fields of quantum physics and computer vision to evaluate our system's performance. The results show that DocReLM achieves a Top 10 accuracy of 44.12% in computer vision, compared to Google Scholar's 15.69%, and an increase to 36.21% in quantum physics, while that of Google Scholar is 12.96%.
Authors: Jialong Guo, Xinghao Chen, Yehui Tang, Yunhe Wang
Abstract: Transformers have become foundational architectures for both natural language and computer vision tasks. However, the high computational cost makes it quite challenging to deploy on resource-constraint devices. This paper investigates the computational bottleneck modules of efficient transformer, i.e., normalization layers and attention modules. LayerNorm is commonly used in transformer architectures but is not computational friendly due to statistic calculation during inference. However, replacing LayerNorm with more efficient BatchNorm in transformer often leads to inferior performance and collapse in training. To address this problem, we propose a novel method named PRepBN to progressively replace LayerNorm with re-parameterized BatchNorm in training. Moreover, we propose a simplified linear attention (SLA) module that is simple yet effective to achieve strong performance. Extensive experiments on image classification as well as object detection demonstrate the effectiveness of our proposed method. For example, our SLAB-Swin obtains $83.6\%$ top-1 accuracy on ImageNet-1K with $16.2$ms latency, which is $2.4$ms less than that of Flatten-Swin with $0.1\%$ higher accuracy. We also evaluated our method for language modeling task and obtain comparable performance and lower latency.Codes are publicly available at https://github.com/xinghaochen/SLAB and https://github.com/mindspore-lab/models/tree/master/research/huawei-noah/SLAB.
URLs: https://github.com/xinghaochen/SLAB, https://github.com/mindspore-lab/models/tree/master/research/huawei-noah/SLAB.
Authors: Zishan Gu, Fenglin Liu, Changchang Yin, Ping Zhang
Abstract: The adoption of large language models (LLMs) in healthcare has attracted significant research interest. However, their performance in healthcare remains under-investigated and potentially limited, due to i) they lack rich domain-specific knowledge and medical reasoning skills; and ii) most state-of-the-art LLMs are unimodal, text-only models that cannot directly process multimodal inputs. To this end, we propose a multimodal medical collaborative reasoning framework \textbf{MultiMedRes}, which incorporates a learner agent to proactively gain essential information from domain-specific expert models, to solve medical multimodal reasoning problems. Our method includes three steps: i) \textbf{Inquire}: The learner agent first decomposes given complex medical reasoning problems into multiple domain-specific sub-problems; ii) \textbf{Interact}: The agent then interacts with domain-specific expert models by repeating the ``ask-answer'' process to progressively obtain different domain-specific knowledge; iii) \textbf{Integrate}: The agent finally integrates all the acquired domain-specific knowledge to accurately address the medical reasoning problem. We validate the effectiveness of our method on the task of difference visual question answering for X-ray images. The experiments demonstrate that our zero-shot prediction achieves state-of-the-art performance, and even outperforms the fully supervised methods. Besides, our approach can be incorporated into various LLMs and multimodal LLMs to significantly boost their performance.
Authors: Ahnaf Mozib Samin, M. Firoz Ahmed, Md. Mushtaq Shahriyar Rafee
Abstract: With the utilization of Transformer architecture, large Vision and Language (V&L) models have shown promising performance in even zero-shot settings. Several studies, however, indicate a lack of robustness of the models when dealing with complex linguistics and visual attributes. In this work, we introduce a novel V&L benchmark - ColorFoil, by creating color-related foils to assess the models' perception ability to detect colors like red, white, green, etc. We evaluate seven state-of-the-art V&L models including CLIP, ViLT, GroupViT, and BridgeTower, etc. in a zero-shot setting and present intriguing findings from the V&L models. The experimental evaluation indicates that ViLT and BridgeTower demonstrate much better color perception capabilities compared to CLIP and its variants and GroupViT. Moreover, CLIP-based models and GroupViT struggle to distinguish colors that are visually distinct to humans with normal color perception ability.
Authors: Shinyoung Kang, Jihan Kim
Abstract: In this study, we explore the potential of using quantum natural language processing (QNLP) to inverse design metal-organic frameworks (MOFs) with targeted properties. Specifically, by analyzing 150 hypothetical MOF structures consisting of 10 metal nodes and 15 organic ligands, we categorize these structures into four distinct classes for pore volume and $H_{2}$ uptake values. We then compare various QNLP models (i.e. the bag-of-words, DisCoCat (Distributional Compositional Categorical), and sequence-based models) to identify the most effective approach to process the MOF dataset. Using a classical simulator provided by the IBM Qiskit, the bag-of-words model is identified to be the optimum model, achieving validation accuracies of 85.7% and 86.7% for binary classification tasks on pore volume and $H_{2}$ uptake, respectively. Further, we developed multi-class classification models tailored to the probabilistic nature of quantum circuits, with average test accuracies of 88.4% and 80.7% across different classes for pore volume and $H_{2}$ uptake datasets. Finally, the performance of generating MOF with target properties showed accuracies of 93.5% for pore volume and 89% for $H_{2}$ uptake, respectively. Although our investigation covers only a fraction of the vast MOF search space, it marks a promising first step towards using quantum computing for materials design, offering a new perspective through which to explore the complex landscape of MOFs.
Authors: Jungwoo Kim, Ji-Su Lee, Huijae Kim, Taesik Lee
Abstract: This paper presents an analytical framework for conducting academic reviews in the field of Healthcare Systems Engineering, employing ChatGPT, a state-of-the-art tool among recent language models. We utilized 9,809 abstract paragraphs from conference presentations to systematically review the field. The framework comprises distinct analytical processes, each employing tailored prompts and the systematic use of the ChatGPT API. Through this framework, we organized the target field into 11 topic categories and conducted a comprehensive analysis covering quantitative yearly trends and detailed sub-categories. This effort explores the potential for leveraging ChatGPT to alleviate the burden of academic reviews. Furthermore, it provides valuable insights into the dynamic landscape of Healthcare Systems Engineering research.
Authors: Siyu Lou, Yuntian Chen, Xiaodan Liang, Liang Lin, Quanshi Zhang
Abstract: In this study, we propose an axiomatic system to define and quantify the precise memorization and in-context reasoning effects used by the large language model (LLM) for language generation. These effects are formulated as non-linear interactions between tokens/words encoded by the LLM. Specifically, the axiomatic system enables us to categorize the memorization effects into foundational memorization effects and chaotic memorization effects, and further classify in-context reasoning effects into enhanced inference patterns, eliminated inference patterns, and reversed inference patterns. Besides, the decomposed effects satisfy the sparsity property and the universal matching property, which mathematically guarantee that the LLM's confidence score can be faithfully decomposed into the memorization effects and in-context reasoning effects. Experiments show that the clear disentanglement of memorization effects and in-context reasoning effects enables a straightforward examination of detailed inference patterns encoded by LLMs.
Authors: Jan-Christoph Klie, Rahul Nair, Juan Haladjian, Marc Kirchner
Abstract: Annotated datasets are an essential ingredient to train, evaluate, compare and productionalize supervised machine learning models. It is therefore imperative that annotations are of high quality. For their creation, good quality management and thereby reliable quality estimates are needed. Then, if quality is insufficient during the annotation process, rectifying measures can be taken to improve it. Quality estimation is often performed by having experts manually label instances as correct or incorrect. But checking all annotated instances tends to be expensive. Therefore, in practice, usually only subsets are inspected; sizes are chosen mostly without justification or regard to statistical power and more often than not, are relatively small. Basing estimates on small sample sizes, however, can lead to imprecise values for the error rate. Using unnecessarily large sample sizes costs money that could be better spent, for instance on more annotations. Therefore, we first describe in detail how to use confidence intervals for finding the minimal sample size needed to estimate the annotation error rate. Then, we propose applying acceptance sampling as an alternative to error rate estimation We show that acceptance sampling can reduce the required sample sizes up to 50% while providing the same statistical guarantees.
Authors: Diego Sanmartin
Abstract: Ensuring factual accuracy while maintaining the creative capabilities of Large Language Model Agents (LMAs) poses significant challenges in the development of intelligent agent systems. LMAs face prevalent issues such as information hallucinations, catastrophic forgetting, and limitations in processing long contexts when dealing with knowledge-intensive tasks. This paper introduces a KG-RAG (Knowledge Graph-Retrieval Augmented Generation) pipeline, a novel framework designed to enhance the knowledge capabilities of LMAs by integrating structured Knowledge Graphs (KGs) with the functionalities of LLMs, thereby significantly reducing the reliance on the latent knowledge of LLMs. The KG-RAG pipeline constructs a KG from unstructured text and then performs information retrieval over the newly created graph to perform KGQA (Knowledge Graph Question Answering). The retrieval methodology leverages a novel algorithm called Chain of Explorations (CoE) which benefits from LLMs reasoning to explore nodes and relationships within the KG sequentially. Preliminary experiments on the ComplexWebQuestions dataset demonstrate notable improvements in the reduction of hallucinated content and suggest a promising path toward developing intelligent systems adept at handling knowledge-intensive tasks.
Authors: Zhenwei Shao, Zhou Yu, Jun Yu, Xuecheng Ouyang, Lihao Zheng, Zhenbiao Gai, Mingyang Wang, Jiajun Ding
Abstract: By harnessing the capabilities of large language models (LLMs), recent large multimodal models (LMMs) have shown remarkable versatility in open-world multimodal understanding. Nevertheless, they are usually parameter-heavy and computation-intensive, thus hindering their applicability in resource-constrained scenarios. To this end, several lightweight LMMs have been proposed successively to maximize the capabilities under constrained scale (e.g., 3B). Despite the encouraging results achieved by these methods, most of them only focus on one or two aspects of the design space, and the key design choices that influence model capability have not yet been thoroughly investigated. In this paper, we conduct a systematic study for lightweight LMMs from the aspects of model architecture, training strategy, and training data. Based on our findings, we obtain Imp -- a family of highly capable LMMs at the 2B-4B scales. Notably, our Imp-3B model steadily outperforms all the existing lightweight LMMs of similar size, and even surpasses the state-of-the-art LMMs at the 13B scale. With low-bit quantization and resolution reduction techniques, our Imp model can be deployed on a Qualcomm Snapdragon 8Gen3 mobile chip with a high inference speed of about 13 tokens/s.
Authors: Zhankui He, Zhouhang Xie, Harald Steck, Dawen Liang, Rahul Jha, Nathan Kallus, Julian McAuley
Abstract: Large language models (LLMs) are revolutionizing conversational recommender systems by adeptly indexing item content, understanding complex conversational contexts, and generating relevant item titles. However, controlling the distribution of recommended items remains a challenge. This leads to suboptimal performance due to the failure to capture rapidly changing data distributions, such as item popularity, on targeted conversational recommendation platforms. In conversational recommendation, LLMs recommend items by generating the titles (as multiple tokens) autoregressively, making it difficult to obtain and control the recommendations over all items. Thus, we propose a Reindex-Then-Adapt (RTA) framework, which converts multi-token item titles into single tokens within LLMs, and then adjusts the probability distributions over these single-token item titles accordingly. The RTA framework marries the benefits of both LLMs and traditional recommender systems (RecSys): understanding complex queries as LLMs do; while efficiently controlling the recommended item distributions in conversational recommendations as traditional RecSys do. Our framework demonstrates improved accuracy metrics across three different conversational recommendation datasets and two adaptation settings
Authors: Robert E. Wray, James R. Kirk, John E. Laird
Abstract: Cognitive systems generally require a human to translate a problem definition into some specification that the cognitive system can use to attempt to solve the problem or perform the task. In this paper, we illustrate that large language models (LLMs) can be utilized to map a problem class, defined in natural language, into a semi-formal specification that can then be utilized by an existing reasoning and learning system to solve instances from the problem class. We present the design of LLM-enabled cognitive task analyst agent(s). Implemented with LLM agents, this system produces a definition of problem spaces for tasks specified in natural language. LLM prompts are derived from the definition of problem spaces in the AI literature and general problem-solving strategies (Polya's How to Solve It). A cognitive system can then use the problem-space specification, applying domain-general problem solving strategies ("weak methods" such as search), to solve multiple instances of problems from the problem class. This result, while preliminary, suggests the potential for speeding cognitive systems research via disintermediation of problem formulation while also retaining core capabilities of cognitive systems, such as robust inference and online learning.
Authors: Gabriele Sarti, Malvina Nissim
Abstract: We introduce IT5, the first family of encoder-decoder transformer models pretrained specifically on Italian. We document and perform a thorough cleaning procedure for a large Italian corpus and use it to pretrain four IT5 model sizes. We then introduce the ItaGen benchmark, which includes a broad range of natural language understanding and generation tasks for Italian, and use it to evaluate the performance of IT5 models and multilingual baselines. We find monolingual IT5 models to provide the best scale-to-performance ratio across tested models, consistently outperforming their multilingual counterparts and setting a new state-of-the-art for Italian language generation.
Authors: Mutian He, Tianqing Fang, Weiqi Wang, Yangqiu Song
Abstract: Conceptualization, or viewing entities and situations as instances of abstract concepts in mind and making inferences based on that, is a vital component in human intelligence for commonsense reasoning. Despite recent progress in artificial intelligence to acquire and model commonsense attributed to neural language models and commonsense knowledge graphs (CKGs), conceptualization is yet to be introduced thoroughly, making current approaches ineffective to cover knowledge about countless diverse entities and situations in the real world. To address the problem, we thoroughly study the role of conceptualization in commonsense reasoning, and formulate a framework to replicate human conceptual induction by acquiring abstract knowledge about events regarding abstract concepts, as well as higher-level triples or inferences upon them. We then apply the framework to ATOMIC, a large-scale human-annotated CKG, aided by the taxonomy Probase. We annotate a dataset on the validity of contextualized conceptualizations from ATOMIC on both event and triple levels, develop a series of heuristic rules based on linguistic features, and train a set of neural models to generate and verify abstract knowledge. Based on these components, a pipeline to acquire abstract knowledge is built. A large abstract CKG upon ATOMIC is then induced, ready to be instantiated to infer about unseen entities or situations. Finally, we empirically show the benefits of augmenting CKGs with abstract knowledge in downstream tasks like commonsense inference and zero-shot commonsense QA.
Authors: Shaoxiong Ji, Ya Gao, Pekka Marttinen
Abstract: Adverse drug events (ADEs) are an important aspect of drug safety. Various texts such as biomedical literature, drug reviews, and user posts on social media and medical forums contain a wealth of information about ADEs. Recent studies have applied word embedding and deep learning -based natural language processing to automate ADE detection from text. However, they did not explore incorporating explicit medical knowledge about drugs and adverse reactions or the corresponding feature learning. This paper adopts the heterogenous text graph which describes relationships between documents, words and concepts, augments it with medical knowledge from the Unified Medical Language System, and proposes a concept-aware attention mechanism which learns features differently for the different types of nodes in the graph. We further utilize contextualized embeddings from pretrained language models and convolutional graph neural networks for effective feature representation and relational learning. Experiments on four public datasets show that our model achieves performance competitive to the recent advances and the concept-aware attention consistently outperforms other attention mechanisms.
Authors: Zhuosheng Zhang, Aston Zhang, Mu Li, Hai Zhao, George Karypis, Alex Smola
Abstract: Large language models (LLMs) have shown impressive performance on complex reasoning by leveraging chain-of-thought (CoT) prompting to generate intermediate reasoning chains as the rationale to infer the answer. However, existing CoT studies have primarily focused on the language modality. We propose Multimodal-CoT that incorporates language (text) and vision (images) modalities into a two-stage framework that separates rationale generation and answer inference. In this way, answer inference can leverage better generated rationales that are based on multimodal information. Experimental results on ScienceQA and A-OKVQA benchmark datasets show the effectiveness of our proposed approach. With Multimodal-CoT, our model under 1 billion parameters achieves state-of-the-art performance on the ScienceQA benchmark. Our analysis indicates that Multimodal-CoT offers the advantages of mitigating hallucination and enhancing convergence speed. Code is publicly available at https://github.com/amazon-science/mm-cot.
Authors: Jincenzi Wu, Zhuang Chen, Jiawen Deng, Sahand Sabour, Helen Meng, Minlie Huang
Abstract: Theory of mind (ToM) refers to humans' ability to understand and infer the desires, beliefs, and intentions of others. The acquisition of ToM plays a key role in humans' social cognition and interpersonal relations. Though indispensable for social intelligence, ToM is still lacking for modern AI and NLP systems since they cannot access the human mental state and cognitive process beneath the training corpus. To empower AI systems with the ToM ability and narrow the gap between them and humans, in this paper, we propose COKE: the first cognitive knowledge graph for machine theory of mind. Specifically, COKE formalizes ToM as a collection of 45k+ manually verified cognitive chains that characterize human mental activities and subsequent behavioral/affective responses when facing specific social circumstances. In addition, we further generalize COKE using LLMs and build a powerful generation model COLM tailored for cognitive reasoning. Experimental results in both automatic and human evaluation demonstrate the high quality of COKE, the superior ToM ability of COLM, and its potential to significantly enhance social applications.
Authors: Yafu Li, Qintong Li, Leyang Cui, Wei Bi, Zhilin Wang, Longyue Wang, Linyi Yang, Shuming Shi, Yue Zhang
Abstract: Large language models (LLMs) have achieved human-level text generation, emphasizing the need for effective AI-generated text detection to mitigate risks like the spread of fake news and plagiarism. Existing research has been constrained by evaluating detection methods on specific domains or particular language models. In practical scenarios, however, the detector faces texts from various domains or LLMs without knowing their sources. To this end, we build a comprehensive testbed by gathering texts from diverse human writings and texts generated by different LLMs. Empirical results show challenges in distinguishing machine-generated texts from human-authored ones across various scenarios, especially out-of-distribution. These challenges are due to the decreasing linguistic distinctions between the two sources. Despite challenges, the top-performing detector can identify 86.54% out-of-domain texts generated by a new LLM, indicating the feasibility for application scenarios. We release our resources at https://github.com/yafuly/MAGE.
Authors: Zhen Lin, Shubhendu Trivedi, Jimeng Sun
Abstract: Large language models (LLMs) specializing in natural language generation (NLG) have recently started exhibiting promising capabilities across a variety of domains. However, gauging the trustworthiness of responses generated by LLMs remains an open challenge, with limited research on uncertainty quantification (UQ) for NLG. Furthermore, existing literature typically assumes white-box access to language models, which is becoming unrealistic either due to the closed-source nature of the latest LLMs or computational constraints. In this work, we investigate UQ in NLG for *black-box* LLMs. We first differentiate *uncertainty* vs *confidence*: the former refers to the ``dispersion'' of the potential predictions for a fixed input, and the latter refers to the confidence on a particular prediction/generation. We then propose and compare several confidence/uncertainty measures, applying them to *selective NLG* where unreliable results could either be ignored or yielded for further assessment. Experiments were carried out with several popular LLMs on question-answering datasets (for evaluation purposes). Results reveal that a simple measure for the semantic dispersion can be a reliable predictor of the quality of LLM responses, providing valuable insights for practitioners on uncertainty management when adopting LLMs. The code to replicate our experiments is available at https://github.com/zlin7/UQ-NLG.
Authors: Joel Niklaus, Veton Matoshi, Matthias St\"urmer, Ilias Chalkidis, Daniel E. Ho
Abstract: Large, high-quality datasets are crucial for training Large Language Models (LLMs). However, so far, there are few datasets available for specialized critical domains such as law and the available ones are often only for the English language. We curate and release MultiLegalPile, a 689GB corpus in 24 languages from 17 jurisdictions. The MultiLegalPile corpus, which includes diverse legal data sources with varying licenses, allows for pretraining NLP models under fair use, with more permissive licenses for the Eurlex Resources and Legal mC4 subsets. We pretrain two RoBERTa models and one Longformer multilingually, and 24 monolingual models on each of the language-specific subsets and evaluate them on LEXTREME. Additionally, we evaluate the English and multilingual models on LexGLUE. Our multilingual models set a new SotA on LEXTREME and our English models on LexGLUE. We release the dataset, the trained models, and all of the code under the most open possible licenses.
Authors: Letian Peng, Yuwei Zhang, Jingbo Shang
Abstract: Prompting large language models (LLMs) for data augmentation has recently become a common practice in few-shot NLP tasks. In this paper, we propose Chain-of-Thought Attribute Manipulation (CoTAM), a novel approach that generates new data from existing examples by only tweaking in the user-provided, task-specific attribute, e.g., sentiment polarity or topic in movie reviews. Instead of conventional latent representation controlling, we leverage the chain-of-thought prompting to directly edit the text in three steps, (1) attribute decomposition, (2) manipulation proposal, and (3) sentence reconstruction. Extensive results on various tasks, such as text (pair) classification, aspect-based sentiment analysis, and conditional text generation, verify the superiority of CoTAM over other LLM-based augmentation methods with the same number of training examples for both fine-tuning and in-context learning. Remarkably, the 2D visualization of the augmented dataset using principal component analysis revealed a human-recognizable decision boundary that is likely hinted by the attribute manipulation, demonstrating the potential of our proposed approach.
Authors: Anuoluwapo Aremu, Jesujoba O. Alabi, Daud Abolade, Nkechinyere F. Aguobi, Shamsuddeen Hassan Muhammad, David Ifeoluwa Adelani
Abstract: In this paper, we create NaijaRC: a new multi-choice Reading Comprehension dataset for three native Nigeria languages that is based on high-school reading comprehension examination. We provide baseline results by performing cross-lingual transfer using existing English RACE and Belebele training dataset based on a pre-trained encoder-only model. Additionally, we provide results by prompting large language models (LLMs) like GPT-4.
Authors: Alex Nyffenegger, Matthias St\"urmer, Joel Niklaus
Abstract: Anonymity of both natural and legal persons in court rulings is a critical aspect of privacy protection in the European Union and Switzerland. With the advent of LLMs, concerns about large-scale re-identification of anonymized persons are growing. In accordance with the Federal Supreme Court of Switzerland, we explore the potential of LLMs to re-identify individuals in court rulings by constructing a proof-of-concept using actual legal data from the Swiss federal supreme court. Following the initial experiment, we constructed an anonymized Wikipedia dataset as a more rigorous testing ground to further investigate the findings. With the introduction and application of the new task of re-identifying people in texts, we also introduce new metrics to measure performance. We systematically analyze the factors that influence successful re-identifications, identifying model size, input length, and instruction tuning among the most critical determinants. Despite high re-identification rates on Wikipedia, even the best LLMs struggled with court decisions. The complexity is attributed to the lack of test datasets, the necessity for substantial training resources, and data sparsity in the information used for re-identification. In conclusion, this study demonstrates that re-identification using LLMs may not be feasible for now, but as the proof-of-concept on Wikipedia showed, it might become possible in the future. We hope that our system can help enhance the confidence in the security of anonymized decisions, thus leading to the courts being more confident to publish decisions.
Authors: Jun Zhang, Jue Wang, Huan Li, Lidan Shou, Ke Chen, Gang Chen, Sharad Mehrotra
Abstract: We present a novel inference scheme, self-speculative decoding, for accelerating Large Language Models (LLMs) without the need for an auxiliary model. This approach is characterized by a two-stage process: drafting and verification. The drafting stage generates draft tokens at a slightly lower quality but more quickly, which is achieved by selectively skipping certain intermediate layers during drafting. Subsequently, the verification stage employs the original LLM to validate those draft output tokens in one forward pass. This process ensures the final output remains identical to that produced by the unaltered LLM. Moreover, the proposed method requires no additional neural network training and no extra memory footprint, making it a plug-and-play and cost-effective solution for inference acceleration. Benchmarks with LLaMA-2 and its variants demonstrated a speedup up to 1.99$\times$.
Authors: Panatchakorn Anantaprayoon, Masahiro Kaneko, Naoaki Okazaki
Abstract: Discriminatory gender biases have been found in Pre-trained Language Models (PLMs) for multiple languages. In Natural Language Inference (NLI), existing bias evaluation methods have focused on the prediction results of one specific label out of three labels, such as neutral. However, such evaluation methods can be inaccurate since unique biased inferences are associated with unique prediction labels. Addressing this limitation, we propose a bias evaluation method for PLMs, called NLI-CoAL, which considers all the three labels of NLI task. First, we create three evaluation data groups that represent different types of biases. Then, we define a bias measure based on the corresponding label output of each data group. In the experiments, we introduce a meta-evaluation technique for NLI bias measures and use it to confirm that our bias measure can distinguish biased, incorrect inferences from non-biased incorrect inferences better than the baseline, resulting in a more accurate bias evaluation. We create the datasets in English, Japanese, and Chinese, and successfully validate the compatibility of our bias measure across multiple languages. Lastly, we observe the bias tendencies in PLMs of different languages. To our knowledge, we are the first to construct evaluation datasets and measure PLMs' bias from NLI in Japanese and Chinese.
Authors: Haven Kim, Jongmin Jung, Dasaem Jeong, Juhan Nam
Abstract: Lyric translation, a field studied for over a century, is now attracting computational linguistics researchers. We identified two limitations in previous studies. Firstly, lyric translation studies have predominantly focused on Western genres and languages, with no previous study centering on K-pop despite its popularity. Second, the field of lyric translation suffers from a lack of publicly available datasets; to the best of our knowledge, no such dataset exists. To broaden the scope of genres and languages in lyric translation studies, we introduce a novel singable lyric translation dataset, approximately 89\% of which consists of K-pop song lyrics. This dataset aligns Korean and English lyrics line-by-line and section-by-section. We leveraged this dataset to unveil unique characteristics of K-pop lyric translation, distinguishing it from other extensively studied genres, and to construct a neural lyric translation model, thereby underscoring the importance of a dedicated dataset for singable lyric translations.
Authors: Zhuosheng Zhang, Aston Zhang
Abstract: Autonomous graphical user interface (GUI) agents aim to facilitate task automation by interacting with the user interface without manual intervention. Recent studies have investigated eliciting the capabilities of large language models (LLMs) for effective engagement in diverse environments. To align with the input-output requirement of LLMs, most existing approaches are developed under a sandbox setting where they rely on external tools and application-specific APIs to parse the environment into textual elements and interpret the predicted actions. Consequently, those approaches often grapple with inference inefficiency and error propagation risks. To mitigate the challenges, we introduce Auto-GUI, a multimodal solution that directly interacts with the interface, bypassing the need for environment parsing or reliance on application-dependent APIs. Moreover, we propose a chain-of-action technique -- leveraging a series of intermediate previous action histories and future action plans -- to help the agent decide what action to execute. We evaluate our approach on a new device-control benchmark AITW with 30$K$ unique instructions, spanning multi-step tasks such as application operation, web searching, and web shopping. Experimental results show that Auto-GUI achieves state-of-the-art performance with an action type prediction accuracy of 90\% and an overall action success rate of 74\%. Code is publicly available at https://github.com/cooelf/Auto-GUI.
Authors: Devrim Cavusoglu, Secil Sen, Ulas Sert, Sinan Altinuc
Abstract: Evaluation plays a critical role in deep learning as a fundamental block of any prediction-based system. However, the vast number of Natural Language Processing (NLP) tasks and the development of various metrics have led to challenges in evaluating different systems with different metrics. To address these challenges, we introduce jury, a toolkit that provides a unified evaluation framework with standardized structures for performing evaluation across different tasks and metrics. The objective of jury is to standardize and improve metric evaluation for all systems and aid the community in overcoming the challenges in evaluation. Since its open-source release, jury has reached a wide audience and is available at https://github.com/obss/jury.
Authors: Isidora Chara Tourni, Subhajit Naskar
Abstract: Direct neural machine translation (direct NMT) is a type of NMT system that translates text between two non-English languages. Direct NMT systems often face limitations due to the scarcity of parallel data between non-English language pairs. Several approaches have been proposed to address this limitation, such as multilingual NMT and pivot NMT (translation between two languages via English). Task-level Mixture of expert models (Task-level MoE), an inference-efficient variation of Transformer-based models, has shown promising NMT performance for a large number of language pairs. In Task-level MoE, different language groups can use different routing strategies to optimize cross-lingual learning and inference speed. In this work, we examine Task-level MoE's applicability in direct NMT and propose a series of high-performing training and evaluation configurations, through which Task-level MoE-based direct NMT systems outperform bilingual and pivot-based models for a large number of low and high-resource direct pairs, and translation directions. Our Task-level MoE with 16 experts outperforms bilingual NMT, Pivot NMT models for 7 language pairs, while pivot-based models still performed better in 9 pairs and directions.
Authors: Weixu Zhang, Yifei Wang, Yuanfeng Song, Victor Junqiu Wei, Yuxing Tian, Yiyan Qi, Jonathan H. Chan, Raymond Chi-Wing Wong, Haiqin Yang
Abstract: The emergence of natural language processing has revolutionized the way users interact with tabular data, enabling a shift from traditional query languages and manual plotting to more intuitive, language-based interfaces. The rise of large language models (LLMs) such as ChatGPT and its successors has further advanced this field, opening new avenues for natural language processing techniques. This survey presents a comprehensive overview of natural language interfaces for tabular data querying and visualization, which allow users to interact with data using natural language queries. We introduce the fundamental concepts and techniques underlying these interfaces with a particular emphasis on semantic parsing, the key technology facilitating the translation from natural language to SQL queries or data visualization commands. We then delve into the recent advancements in Text-to-SQL and Text-to-Vis problems from the perspectives of datasets, methodologies, metrics, and system designs. This includes a deep dive into the influence of LLMs, highlighting their strengths, limitations, and potential for future improvements. Through this survey, we aim to provide a roadmap for researchers and practitioners interested in developing and applying natural language interfaces for data interaction in the era of large language models.
Authors: Xuanli He, Yuxiang Wu, Oana-Maria Camburu, Pasquale Minervini, Pontus Stenetorp
Abstract: Recent studies demonstrated that large language models (LLMs) can excel in many tasks via in-context learning (ICL). However, recent works show that ICL-prompted models tend to produce inaccurate results when presented with adversarial inputs. In this work, we investigate whether augmenting ICL with natural language explanations (NLEs) improves the robustness of LLMs on adversarial datasets covering natural language inference and paraphrasing identification. We prompt LLMs with a small set of human-generated NLEs to produce further NLEs, yielding more accurate results than both a zero-shot-ICL setting and using only human-generated NLEs. Our results on five popular LLMs (GPT3.5-turbo, Llama2, Vicuna, Zephyr, and Mistral) show that our approach yields over 6% improvement over baseline approaches for eight adversarial datasets: HANS, ISCS, NaN, ST, PICD, PISP, ANLI, and PAWS. Furthermore, previous studies have demonstrated that prompt selection strategies significantly enhance ICL on in-distribution test sets. However, our findings reveal that these strategies do not match the efficacy of our approach for robustness evaluations, resulting in an accuracy drop of 8% compared to the proposed approach.
Authors: Panfeng Li, Mohamed Abouelenien, Rada Mihalcea, Zhicheng Ding, Qikai Yang, Yiming Zhou
Abstract: Deception detection is gaining increasing interest due to ethical and security concerns. This paper explores the application of convolutional neural networks for the purpose of multimodal deception detection. We use a dataset built by interviewing 104 subjects about two topics, with one truthful and one falsified response from each subject about each topic. In particular, we make three main contributions. First, we extract linguistic and physiological features from this data to train and construct the neural network models. Second, we propose a fused convolutional neural network model using both modalities in order to achieve an improved overall performance. Third, we compare our new approach with earlier methods designed for multimodal deception detection. We find that our system outperforms regular classification methods; our results indicate the feasibility of using neural networks for deception detection even in the presence of limited amounts of data.
Authors: Junhao Zheng, Shengjie Qiu, Qianli Ma
Abstract: Incremental Learning (IL) has been a long-standing problem in both vision and Natural Language Processing (NLP) communities. In recent years, as Pre-trained Language Models (PLMs) have achieved remarkable progress in various NLP downstream tasks, utilizing PLMs as backbones has become a common practice in recent research of IL in NLP. Most assume that catastrophic forgetting is the biggest obstacle to achieving superior IL performance and propose various techniques to overcome this issue. However, we find that this assumption is problematic. Specifically, we revisit more than 20 methods on four classification tasks (Text Classification, Intent Classification, Relation Extraction, and Named Entity Recognition) under the two most popular IL settings (Class-Incremental and Task-Incremental) and reveal that most of them severely underestimate the inherent anti-forgetting ability of PLMs. Based on the observation, we propose a frustratingly easy method called SEQ* for IL with PLMs. The results show that SEQ* has competitive or superior performance compared to state-of-the-art (SOTA) IL methods and requires considerably less trainable parameters and training time. These findings urge us to revisit the IL with PLMs and encourage future studies to have a fundamental understanding of the catastrophic forgetting in PLMs. The data, code and scripts are publicly available at https://github.com/zzz47zzz/pretrained-lm-for-incremental-learning.
URLs: https://github.com/zzz47zzz/pretrained-lm-for-incremental-learning.
Authors: Haotian Ye, Yihong Liu, Chunlan Ma, Hinrich Sch\"utze
Abstract: Transformer-based pre-trained language models (PLMs) have achieved remarkable performance in various natural language processing (NLP) tasks. However, pre-training such models can take considerable resources that are almost only available to high-resource languages. On the contrary, static word embeddings are easier to train in terms of computing resources and the amount of data required. In this paper, we introduce MoSECroT Model Stitching with Static Word Embeddings for Crosslingual Zero-shot Transfer), a novel and challenging task that is especially relevant to low-resource languages for which static word embeddings are available. To tackle the task, we present the first framework that leverages relative representations to construct a common space for the embeddings of a source language PLM and the static word embeddings of a target language. In this way, we can train the PLM on source-language training data and perform zero-shot transfer to the target language by simply swapping the embedding layer. However, through extensive experiments on two classification datasets, we show that although our proposed framework is competitive with weak baselines when addressing MoSECroT, it fails to achieve competitive results compared with some strong baselines. In this paper, we attempt to explain this negative result and provide several thoughts on possible improvement.
Authors: Georgi Karadzhov, Andreas Vlachos, Tom Stafford
Abstract: We explore different aspects of cognitive diversity and its effect on the success of group deliberation. To evaluate this, we use 500 dialogues from small, online groups discussing the Wason Card Selection task - the DeliData corpus. Leveraging the corpus, we perform quantitative analysis evaluating three different measures of cognitive diversity. First, we analyse the effect of group size as a proxy measure for diversity. Second, we evaluate the effect of the size of the initial idea pool. Finally, we look into the content of the discussion by analysing discussed solutions, discussion patterns, and how conversational probing can improve those characteristics. Despite the reputation of groups for compounding bias, we show that small groups can, through dialogue, overcome intuitive biases and improve individual decision-making. Across a large sample and different operationalisations, we consistently find that greater cognitive diversity is associated with more successful group deliberation. Code and data used for the analysis are available in the repository: https://github.com/gkaradzhov/cognitive-diversity-groups-cogsci24.
URLs: https://github.com/gkaradzhov/cognitive-diversity-groups-cogsci24.
Authors: Cheng-Han Chiang, Hung-yi Lee
Abstract: Long-form generations from large language models (LLMs) contains a mix of factual and non-factual claims, making evaluating factuality difficult. To evaluate factual precision of long-form generations in a more fine-grained way, prior works propose to decompose long-form generations into multiple verifiable facts and verify those facts independently. The factuality of the generation is the proportion of verifiable facts among all the facts. Such methods assume that combining factual claims forms a factual paragraph. This paper shows that the assumption can be violated due to entity ambiguity. We show that LLMs can generate paragraphs that contain verifiable facts, but the facts are combined to form a non-factual paragraph due to entity ambiguity. We further reveal that existing factual precision metrics, including FActScore and citation recall, cannot properly evaluate the factuality of these non-factual paragraphs. To address this, we introduce an enhanced metric, D-FActScore, specifically designed for content with ambiguous entities. We evaluate the D-FActScores of people biographies generated with retrieval-augmented generation (RAG). We show that D-FActScore can better assess the factuality of paragraphs with entity ambiguity than FActScore. We also find that four widely used open-source LLMs tend to mix information of distinct entities to form non-factual paragraphs.
Authors: Hanling Yi, Feng Lin, Hongbin Li, Peiyang Ning, Xiaotian Yu, Rong Xiao
Abstract: This research aims to accelerate the inference speed of large language models (LLMs) with billions of parameters. We propose \textbf{S}mart \textbf{P}arallel \textbf{A}uto-\textbf{C}orrect d\textbf{E}coding (SPACE), an innovative approach designed for achieving lossless acceleration of LLMs. By integrating semi-autoregressive inference and speculative decoding capabilities, SPACE uniquely enables autoregressive LLMs to parallelize token generation and verification. This is realized through a specialized semi-autoregressive supervised fine-tuning process that equips existing LLMs with the ability to simultaneously predict multiple tokens. Additionally, an auto-correct decoding algorithm facilitates the simultaneous generation and verification of token sequences within a single model invocation. Through extensive experiments on a range of LLMs, SPACE has demonstrated inference speedup ranging from 2.7x-4.0x on HumanEval-X while maintaining output quality.
Authors: Kinjal Basu, Ibrahim Abdelaziz, Subhajit Chaudhury, Soham Dan, Maxwell Crouse, Asim Munawar, Sadhana Kumaravel, Vinod Muthusamy, Pavan Kapanipathi, Luis A. Lastras
Abstract: There is a growing need for Large Language Models (LLMs) to effectively use tools and external Application Programming Interfaces (APIs) to plan and complete tasks. As such, there is tremendous interest in methods that can acquire sufficient quantities of train and test data that involve calls to tools / APIs. Two lines of research have emerged as the predominant strategies for addressing this challenge. The first has focused on synthetic data generation techniques, while the second has involved curating task-adjacent datasets which can be transformed into API / Tool-based tasks. In this paper, we focus on the task of identifying, curating, and transforming existing datasets and, in turn, introduce API-BLEND, a large corpora for training and systematic testing of tool-augmented LLMs. The datasets mimic real-world scenarios involving API-tasks such as API / tool detection, slot filling, and sequencing of the detected APIs. We demonstrate the utility of the API-BLEND dataset for both training and benchmarking purposes.
Authors: Jiabin Tang, Yuhao Yang, Wei Wei, Lei Shi, Long Xia, Dawei Yin, Chao Huang
Abstract: Heterogeneous graph learning aims to capture complex relationships and diverse relational semantics among entities in a heterogeneous graph to obtain meaningful representations for nodes and edges. Recent advancements in heterogeneous graph neural networks (HGNNs) have achieved state-of-the-art performance by considering relation heterogeneity and using specialized message functions and aggregation rules. However, existing frameworks for heterogeneous graph learning have limitations in generalizing across diverse heterogeneous graph datasets. Most of these frameworks follow the "pre-train" and "fine-tune" paradigm on the same dataset, which restricts their capacity to adapt to new and unseen data. This raises the question: "Can we generalize heterogeneous graph models to be well-adapted to diverse downstream learning tasks with distribution shifts in both node token sets and relation type heterogeneity?'' To tackle those challenges, we propose HiGPT, a general large graph model with Heterogeneous graph instruction-tuning paradigm. Our framework enables learning from arbitrary heterogeneous graphs without the need for any fine-tuning process from downstream datasets. To handle distribution shifts in heterogeneity, we introduce an in-context heterogeneous graph tokenizer that captures semantic relationships in different heterogeneous graphs, facilitating model adaptation. We incorporate a large corpus of heterogeneity-aware graph instructions into our HiGPT, enabling the model to effectively comprehend complex relation heterogeneity and distinguish between various types of graph tokens. Furthermore, we introduce the Mixture-of-Thought (MoT) instruction augmentation paradigm to mitigate data scarcity by generating diverse and informative instructions. Through comprehensive evaluations, our proposed framework demonstrates exceptional performance in terms of generalization performance.
Authors: Yihao Ding, Lorenzo Vaiani, Caren Han, Jean Lee, Paolo Garza, Josiah Poon, Luca Cagliero
Abstract: This paper presents a groundbreaking multimodal, multi-task, multi-teacher joint-grained knowledge distillation model for visually-rich form document understanding. The model is designed to leverage insights from both fine-grained and coarse-grained levels by facilitating a nuanced correlation between token and entity representations, addressing the complexities inherent in form documents. Additionally, we introduce new inter-grained and cross-grained loss functions to further refine diverse multi-teacher knowledge distillation transfer process, presenting distribution gaps and a harmonised understanding of form documents. Through a comprehensive evaluation across publicly available form document understanding datasets, our proposed model consistently outperforms existing baselines, showcasing its efficacy in handling the intricate structures and content of visually complex form documents.
Authors: Zhonghang Li, Lianghao Xia, Jiabin Tang, Yong Xu, Lei Shi, Long Xia, Dawei Yin, Chao Huang
Abstract: Spatio-temporal prediction aims to forecast and gain insights into the ever-changing dynamics of urban environments across both time and space. Its purpose is to anticipate future patterns, trends, and events in diverse facets of urban life, including transportation, population movement, and crime rates. Although numerous efforts have been dedicated to developing neural network techniques for accurate predictions on spatio-temporal data, it is important to note that many of these methods heavily depend on having sufficient labeled data to generate precise spatio-temporal representations. Unfortunately, the issue of data scarcity is pervasive in practical urban sensing scenarios. Consequently, it becomes necessary to build a spatio-temporal model with strong generalization capabilities across diverse spatio-temporal learning scenarios. Taking inspiration from the remarkable achievements of large language models (LLMs), our objective is to create a spatio-temporal LLM that can exhibit exceptional generalization capabilities across a wide range of downstream urban tasks. To achieve this objective, we present the UrbanGPT, which seamlessly integrates a spatio-temporal dependency encoder with the instruction-tuning paradigm. This integration enables LLMs to comprehend the complex inter-dependencies across time and space, facilitating more comprehensive and accurate predictions under data scarcity. To validate the effectiveness of our approach, we conduct extensive experiments on various public datasets, covering different spatio-temporal prediction tasks. The results consistently demonstrate that our UrbanGPT, with its carefully designed architecture, consistently outperforms state-of-the-art baselines. These findings highlight the potential of building large language models for spatio-temporal learning, particularly in zero-shot scenarios where labeled data is scarce.
Authors: Zijie Zeng, Shiqi Liu, Lele Sha, Zhuang Li, Kaixun Yang, Sannyuya Liu, Dragan Ga\v{s}evi\'c, Guanliang Chen
Abstract: This study explores the challenge of sentence-level AI-generated text detection within human-AI collaborative hybrid texts. Existing studies of AI-generated text detection for hybrid texts often rely on synthetic datasets. These typically involve hybrid texts with a limited number of boundaries. We contend that studies of detecting AI-generated content within hybrid texts should cover different types of hybrid texts generated in realistic settings to better inform real-world applications. Therefore, our study utilizes the CoAuthor dataset, which includes diverse, realistic hybrid texts generated through the collaboration between human writers and an intelligent writing system in multi-turn interactions. We adopt a two-step, segmentation-based pipeline: (i) detect segments within a given hybrid text where each segment contains sentences of consistent authorship, and (ii) classify the authorship of each identified segment. Our empirical findings highlight (1) detecting AI-generated sentences in hybrid texts is overall a challenging task because (1.1) human writers' selecting and even editing AI-generated sentences based on personal preferences adds difficulty in identifying the authorship of segments; (1.2) the frequent change of authorship between neighboring sentences within the hybrid text creates difficulties for segment detectors in identifying authorship-consistent segments; (1.3) the short length of text segments within hybrid texts provides limited stylistic cues for reliable authorship determination; (2) before embarking on the detection process, it is beneficial to assess the average length of segments within the hybrid text. This assessment aids in deciding whether (2.1) to employ a text segmentation-based strategy for hybrid texts with longer segments, or (2.2) to adopt a direct sentence-by-sentence classification strategy for those with shorter segments.
Authors: Mengru Wang, Ningyu Zhang, Ziwen Xu, Zekun Xi, Shumin Deng, Yunzhi Yao, Qishen Zhang, Linyi Yang, Jindong Wang, Huajun Chen
Abstract: This paper investigates using knowledge editing techniques to detoxify Large Language Models (LLMs). We construct a benchmark, SafeEdit, which covers nine unsafe categories with various powerful attack prompts and equips comprehensive metrics for systematic evaluation. We conduct experiments with several knowledge editing approaches, indicating that knowledge editing has the potential to detoxify LLMs with a limited impact on general performance efficiently. Then, we propose a simple yet effective baseline, dubbed Detoxifying with Intraoperative Neural Monitoring (DINM), to diminish the toxicity of LLMs within a few tuning steps via only one instance. We further provide an in-depth analysis of the internal mechanism for various detoxifying approaches, demonstrating that previous methods like SFT and DPO may merely suppress the activations of toxic parameters, while DINM mitigates the toxicity of the toxic parameters to a certain extent, making permanent adjustments. We hope that these insights could shed light on future work of developing detoxifying approaches and the underlying knowledge mechanisms of LLMs. Code and benchmark are available at https://github.com/zjunlp/EasyEdit.
Authors: Samuel Cahyawijaya, Holy Lovenia, Pascale Fung
Abstract: In-context learning (ICL) empowers large language models (LLMs) to perform diverse tasks in underrepresented languages using only short in-context information, offering a crucial avenue for narrowing the gap between high-resource and low-resource languages. Nonetheless, there is only a handful of works explored ICL for low-resource languages with most of them focusing on relatively high-resource languages, such as French and Spanish. In this work, we extensively study ICL and its cross-lingual variation (X-ICL) on 25 low-resource and 7 relatively higher-resource languages. Our study not only assesses the effectiveness of ICL with LLMs in low-resource languages but also identifies the shortcomings of in-context label alignment, and introduces a more effective alternative: query alignment. Moreover, we provide valuable insights into various facets of ICL for low-resource languages. Our study concludes the significance of few-shot in-context information on enhancing the low-resource understanding quality of LLMs through semantically relevant information by closing the language gap in the target language and aligning the semantics between the targeted low-resource and the high-resource language that the model is proficient in. Our work highlights the importance of advancing ICL research, particularly for low-resource languages.
Authors: Nhat M. Hoang, Xuan Long Do, Duc Anh Do, Duc Anh Vu, Luu Anh Tuan
Abstract: The proliferation of online toxic speech is a pertinent problem posing threats to demographic groups. While explicit toxic speech contains offensive lexical signals, implicit one consists of coded or indirect language. Therefore, it is crucial for models not only to detect implicit toxic speech but also to explain its toxicity. This draws a unique need for unified frameworks that can effectively detect and explain implicit toxic speech. Prior works mainly formulated the task of toxic speech detection and explanation as a text generation problem. Nonetheless, models trained using this strategy can be prone to suffer from the consequent error propagation problem. Moreover, our experiments reveal that the detection results of such models are much lower than those that focus only on the detection task. To bridge these gaps, we introduce ToXCL, a unified framework for the detection and explanation of implicit toxic speech. Our model consists of three modules: a (i) Target Group Generator to generate the targeted demographic group(s) of a given post; an (ii) Encoder-Decoder Model in which the encoder focuses on detecting implicit toxic speech and is boosted by a (iii) Teacher Classifier via knowledge distillation, and the decoder generates the necessary explanation. ToXCL achieves new state-of-the-art effectiveness, and outperforms baselines significantly.
Authors: Neema Kotonya, Francesca Toni
Abstract: As deep neural models in NLP become more complex, and as a consequence opaque, the necessity to interpret them becomes greater. A burgeoning interest has emerged in rationalizing explanations to provide short and coherent justifications for predictions. In this position paper, we advocate for a formal framework for key concepts and properties about rationalizing explanations to support their evaluation systematically. We also outline one such formal framework, tailored to rationalizing explanations of increasingly complex structures, from free-form explanations to deductive explanations, to argumentative explanations (with the richest structure). Focusing on the automated fact verification task, we provide illustrations of the use and usefulness of our formalization for evaluating explanations, tailored to their varying structures.
Authors: Simran Khanuja, Sathyanarayanan Ramamoorthy, Yueqi Song, Graham Neubig
Abstract: Given the rise of multimedia content, human translators increasingly focus on culturally adapting not only words but also other modalities such as images to convey the same meaning. While several applications stand to benefit from this, machine translation systems remain confined to dealing with language in speech and text. In this work, we take a first step towards translating images to make them culturally relevant. First, we build three pipelines comprising state-of-the-art generative models to do the task. Next, we build a two-part evaluation dataset: i) concept: comprising 600 images that are cross-culturally coherent, focusing on a single concept per image, and ii) application: comprising 100 images curated from real-world applications. We conduct a multi-faceted human evaluation of translated images to assess for cultural relevance and meaning preservation. We find that as of today, image-editing models fail at this task, but can be improved by leveraging LLMs and retrievers in the loop. Best pipelines can only translate 5% of images for some countries in the easier concept dataset and no translation is successful for some countries in the application dataset, highlighting the challenging nature of the task. Our code and data is released here: https://github.com/simran-khanuja/image-transcreation.
URLs: https://github.com/simran-khanuja/image-transcreation.
Authors: Zhoujian Sun, Cheng Luo, Ziyi Liu, Zhengxing Huang
Abstract: The development of large language models (LLMs) has brought unprecedented possibilities for artificial intelligence (AI) based medical diagnosis. However, the application perspective of LLMs in real diagnostic scenarios is still unclear because they are not adept at collecting patient data proactively. This study presents a LLM-based diagnostic system that enhances planning capabilities by emulating doctors. Our system involves two external planners to handle planning tasks. The first planner employs a reinforcement learning approach to formulate disease screening questions and conduct initial diagnoses. The second planner uses LLMs to parse medical guidelines and conduct differential diagnoses. By utilizing real patient electronic medical record data, we constructed simulated dialogues between virtual patients and doctors and evaluated the diagnostic abilities of our system. We demonstrated that our system obtained impressive performance in both disease screening and differential diagnoses tasks. This research represents a step towards more seamlessly integrating AI into clinical settings, potentially enhancing the accuracy and accessibility of medical diagnostics.
Authors: Letian Peng, Jingbo Shang
Abstract: In this paper, we aim to generate text classification data given arbitrary class definitions (i.e., user instruction), so one can train a small text classifier without any human annotation or raw corpus. Compared with pioneer attempts, our proposed Incubator is the first framework that can handle complicated and even mutually dependent classes (e.g., "TED Talk given by Educator" and "Other"). Specifically, Incubator is an LLM firstly tuned on the instruction-to-data mappings that we obtained from classification datasets and descriptions on HuggingFace together with in-context augmentation by GPT-4. We then refine Incubator by learning on the cluster centers of semantic textual embeddings to emphasize the uniformity and semantic diversity in generations. We compare Incubator on various classification tasks with strong baselines such as direct LLM-based inference and training data generation by prompt engineering. Experiments show Incubator is able to (1) perform well on traditional benchmarks, (2) take label dependency and user preference into consideration, and (3) enable logical text mining by incubating multiple classifiers.
Authors: Hongzhan Lin, Zixin Chen, Ziyang Luo, Mingfei Cheng, Jing Ma, Guang Chen
Abstract: Social media abounds with multimodal sarcasm, and identifying sarcasm targets is particularly challenging due to the implicit incongruity not directly evident in the text and image modalities. Current methods for Multimodal Sarcasm Target Identification (MSTI) predominantly focus on superficial indicators in an end-to-end manner, overlooking the nuanced understanding of multimodal sarcasm conveyed through both the text and image. This paper proposes a versatile MSTI framework with a coarse-to-fine paradigm, by augmenting sarcasm explainability with reasoning and pre-training knowledge. Inspired by the powerful capacity of Large Multimodal Models (LMMs) on multimodal reasoning, we first engage LMMs to generate competing rationales for coarser-grained pre-training of a small language model on multimodal sarcasm detection. We then propose fine-tuning the model for finer-grained sarcasm target identification. Our framework is thus empowered to adeptly unveil the intricate targets within multimodal sarcasm and mitigate the negative impact posed by potential noise inherently in LMMs. Experimental results demonstrate that our model far outperforms state-of-the-art MSTI methods, and markedly exhibits explainability in deciphering sarcasm as well.
Authors: Li-Chun Lu, Shou-Jen Chen, Tsung-Min Pai, Chan-Hung Yu, Hung-yi Lee, Shao-Hua Sun
Abstract: Large language models (LLMs) have shown exceptional proficiency in natural language processing but often fall short of generating creative and original responses to open-ended questions. To enhance LLM creativity, our key insight is to emulate the human process of inducing collective creativity through engaging discussions with participants from diverse backgrounds and perspectives. To this end, we propose LLM Discussion, a three-phase discussion framework that facilitates vigorous and diverging idea exchanges and ensures convergence to creative answers. Moreover, we adopt a role-playing technique by assigning distinct roles to LLMs to combat the homogeneity of LLMs. We evaluate the efficacy of the proposed framework with the Alternative Uses Test, Similarities Test, Instances Test, and Scientific Creativity Test through both LLM evaluation and human study. Our proposed framework outperforms single-LLM approaches and existing multi-LLM frameworks across various creativity metrics.
Authors: JoonHo Lee, Jae Oh Woo, Juree Seok, Parisa Hassanzadeh, Wooseok Jang, JuYoun Son, Sima Didari, Baruch Gutow, Heng Hao, Hankyu Moon, Wenjun Hu, Yeong-Dae Kwon, Taehee Lee, Seungjai Min
Abstract: Assessing response quality to instructions in language models is vital but challenging due to the complexity of human language across different contexts. This complexity often results in ambiguous or inconsistent interpretations, making accurate assessment difficult. To address this issue, we propose a novel Uncertainty-aware Reward Model (URM) that introduces a robust uncertainty estimation for the quality of paired responses based on Bayesian approximation. Trained with preference datasets, our uncertainty-enabled proxy not only scores rewards for responses but also evaluates their inherent uncertainty. Empirical results demonstrate significant benefits of incorporating the proposed proxy into language model training. Our method boosts the instruction following capability of language models by refining data curation for training and improving policy optimization objectives, thereby surpassing existing methods by a large margin on benchmarks such as Vicuna and MT-bench. These findings highlight that our proposed approach substantially advances language model training and paves a new way of harnessing uncertainty within language models.
Authors: Devansh Jain, Priyanshu Kumar, Samuel Gehman, Xuhui Zhou, Thomas Hartvigsen, Maarten Sap
Abstract: Recent advances in large language models (LLMs) have led to their extensive global deployment, and ensuring their safety calls for comprehensive and multilingual toxicity evaluations. However, existing toxicity benchmarks are overwhelmingly focused on English, posing serious risks to deploying LLMs in other languages. We address this by introducing PolygloToxicityPrompts (PTP), the first large-scale multilingual toxicity evaluation benchmark of 425K naturally occurring prompts spanning 17 languages. We overcome the scarcity of naturally occurring toxicity in web-text and ensure coverage across languages with varying resources by automatically scraping over 100M web-text documents. Using PTP, we investigate research questions to study the impact of model size, prompt language, and instruction and preference-tuning methods on toxicity by benchmarking over 60 LLMs. Notably, we find that toxicity increases as language resources decrease or model size increases. Although instruction- and preference-tuning reduce toxicity, the choice of preference-tuning method does not have any significant impact. Our findings shed light on crucial shortcomings of LLM safeguarding and highlight areas for future research.
Authors: Jiahao Zhao, Jingwei Zhu, Minghuan Tan, Min Yang, Di Yang, Chenhao Zhang, Guancheng Ye, Chengming Li, Xiping Hu
Abstract: In this paper, we introduce a novel psychological benchmark, CPsyExam, constructed from questions sourced from Chinese language examinations. CPsyExam is designed to prioritize psychological knowledge and case analysis separately, recognizing the significance of applying psychological knowledge to real-world scenarios. From the pool of 22k questions, we utilize 4k to create the benchmark that offers balanced coverage of subjects and incorporates a diverse range of case analysis techniques.Furthermore, we evaluate a range of existing large language models~(LLMs), spanning from open-sourced to API-based models. Our experiments and analysis demonstrate that CPsyExam serves as an effective benchmark for enhancing the understanding of psychology within LLMs and enables the comparison of LLMs across various granularities.
Authors: Mina Ghashami, Soumya Smruti Mishra
Abstract: The SemEval 2024 BRAINTEASER task represents a pioneering venture in Natural Language Processing (NLP) by focusing on lateral thinking, a dimension of cognitive reasoning that is often overlooked in traditional linguistic analyses. This challenge comprises of Sentence Puzzle and Word Puzzle subtasks and aims to test language models' capacity for divergent thinking. In this paper, we present our approach to the BRAINTEASER task. We employ a holistic strategy by leveraging cutting-edge pre-trained models in multiple choice architecture, and diversify the training data with Sentence and Word Puzzle datasets. To gain further improvement, we fine-tuned the model with synthetic humor or jokes dataset and the RiddleSense dataset which helped augmenting the model's lateral thinking abilities. Empirical results show that our approach achieve 92.5% accuracy in Sentence Puzzle subtask and 80.2% accuracy in Word Puzzle subtask.
Authors: Anwoy Chatterjee, Eshaan Tanwar, Subhabrata Dutta, Tanmoy Chakraborty
Abstract: Large Language Models (LLMs) have transformed NLP with their remarkable In-context Learning (ICL) capabilities. Automated assistants based on LLMs are gaining popularity; however, adapting them to novel tasks is still challenging. While colossal models excel in zero-shot performance, their computational demands limit widespread use, and smaller language models struggle without context. This paper investigates whether LLMs can generalize from labeled examples of predefined tasks to novel tasks. Drawing inspiration from biological neurons and the mechanistic interpretation of the Transformer architecture, we explore the potential for information sharing across tasks. We design a cross-task prompting setup with three LLMs and show that LLMs achieve significant performance improvements despite no examples from the target task in the context. Cross-task prompting leads to a remarkable performance boost of 107% for LLaMA-2 7B, 18.6% for LLaMA-2 13B, and 3.2% for GPT 3.5 on average over zero-shot prompting, and performs comparable to standard in-context learning. The effectiveness of generating pseudo-labels for in-task examples is demonstrated, and our analyses reveal a strong correlation between the effect of cross-task examples and model activation similarities in source and target input tokens. This paper offers a first-of-its-kind exploration of LLMs' ability to solve novel tasks based on contextual signals from different task examples.
Authors: Ziyao Xu, Houfeng Wang
Abstract: Compositional generalization is an important ability of language models and has many different manifestations. For data-to-text generation, previous research on this ability is limited to a single manifestation called Systematicity and lacks consideration of large language models (LLMs), which cannot fully cover practical application scenarios. In this work, we propose SPOR, a comprehensive and practical evaluation method for compositional generalization in data-to-text generation. SPOR includes four aspects of manifestations (Systematicity, Productivity, Order invariance, and Rule learnability) and allows high-quality evaluation without additional manual annotations based on existing datasets. We demonstrate SPOR on two different datasets and evaluate some existing language models including LLMs. We find that the models are deficient in various aspects of the evaluation and need further improvement. Our work shows the necessity for comprehensive research on different manifestations of compositional generalization in data-to-text generation and provides a framework for evaluation.
Authors: Tinh Son Luong, Thanh-Thien Le, Linh Ngo Van, Thien Huu Nguyen
Abstract: Large language models (LLMs) have become integral to our professional workflows and daily lives. Nevertheless, these machine companions of ours have a critical flaw: the huge amount of data which endows them with vast and diverse knowledge, also exposes them to the inevitable toxicity and bias. While most LLMs incorporate defense mechanisms to prevent the generation of harmful content, these safeguards can be easily bypassed with minimal prompt engineering. In this paper, we introduce the new Thoroughly Engineered Toxicity (TET) dataset, comprising manually crafted prompts designed to nullify the protective layers of such models. Through extensive evaluations, we demonstrate the pivotal role of TET in providing a rigorous benchmark for evaluation of toxicity awareness in several popular LLMs: it highlights the toxicity in the LLMs that might remain hidden when using normal prompts, thus revealing subtler issues in their behavior.
Authors: Takuma Yoneda, Jiading Fang, Peng Li, Huanyu Zhang, Tianchong Jiang, Shengjie Lin, Ben Picker, David Yunis, Hongyuan Mei, Matthew R. Walter
Abstract: There has been a significant research interest in employing large language models to empower intelligent robots with complex reasoning. Existing work focuses on harnessing their abilities to reason about the histories of their actions and observations. In this paper, we explore a new dimension in which large language models may benefit robotics planning. In particular, we propose Statler, a framework in which large language models are prompted to maintain an estimate of the world state, which are often unobservable, and track its transition as new actions are taken. Our framework then conditions each action on the estimate of the current world state. Despite being conceptually simple, our Statler framework significantly outperforms strong competing methods (e.g., Code-as-Policies) on several robot planning tasks. Additionally, it has the potential advantage of scaling up to more challenging long-horizon planning tasks.
Authors: Sanad Aburass, Osama Dorgham, Jamil Al Shaqsi
Abstract: In our study, we introduce a novel hybrid ensemble model that synergistically combines LSTM, BiLSTM, CNN, GRU, and GloVe embeddings for the classification of gene mutations in cancer. This model was rigorously tested using Kaggle's Personalized Medicine: Redefining Cancer Treatment dataset, demonstrating exceptional performance across all evaluation metrics. Notably, our approach achieved a training accuracy of 80.6%, precision of 81.6%, recall of 80.6%, and an F1 score of 83.1%, alongside a significantly reduced Mean Squared Error (MSE) of 2.596. These results surpass those of advanced transformer models and their ensembles, showcasing our model's superior capability in handling the complexities of gene mutation classification. The accuracy and efficiency of gene mutation classification are paramount in the era of precision medicine, where tailored treatment plans based on individual genetic profiles can dramatically improve patient outcomes and save lives. Our model's remarkable performance highlights its potential in enhancing the precision of cancer diagnoses and treatments, thereby contributing significantly to the advancement of personalized healthcare.
Authors: Aiwei Liu, Leyi Pan, Xuming Hu, Shiao Meng, Lijie Wen
Abstract: Watermark algorithms for large language models (LLMs) have achieved extremely high accuracy in detecting text generated by LLMs. Such algorithms typically involve adding extra watermark logits to the LLM's logits at each generation step. However, prior algorithms face a trade-off between attack robustness and security robustness. This is because the watermark logits for a token are determined by a certain number of preceding tokens; a small number leads to low security robustness, while a large number results in insufficient attack robustness. In this work, we propose a semantic invariant watermarking method for LLMs that provides both attack robustness and security robustness. The watermark logits in our work are determined by the semantics of all preceding tokens. Specifically, we utilize another embedding LLM to generate semantic embeddings for all preceding tokens, and then these semantic embeddings are transformed into the watermark logits through our trained watermark model. Subsequent analyses and experiments demonstrated the attack robustness of our method in semantically invariant settings: synonym substitution and text paraphrasing settings. Finally, we also show that our watermark possesses adequate security robustness. Our code and data are available at \href{https://github.com/THU-BPM/Robust_Watermark}{https://github.com/THU-BPM/Robust\_Watermark}. Additionally, our algorithm could also be accessed through MarkLLM \citep{pan2024markllm} \footnote{https://github.com/THU-BPM/MarkLLM}.
URLs: https://github.com/THU-BPM/Robust_Watermark, https://github.com/THU-BPM/Robust\_Watermark, https://github.com/THU-BPM/MarkLLM
Authors: Xiangyu Zeng, Jie Lin, Piao Hu, Ruizheng Huang, Zhicheng Zhang
Abstract: How humans and machines make sense of current inputs for relation reasoning and question-answering while putting the perceived information into context of our past memories, has been a challenging conundrum in cognitive science and artificial intelligence. Inspired by human brain's memory system and cognitive architectures, we propose a PMI framework that consists of perception, memory and inference components. Notably, the memory module comprises working and long-term memory, with the latter endowed with a higher-order structure to retain extensive and complex relational knowledge and experience. Through a differentiable competitive write access, current perceptions update working memory, which is later merged with long-term memory via outer product associations, reducing information conflicts and averting memory overflow. In the inference module, relevant information is retrieved from two separate memory origins and associatively integrated to attain a more comprehensive and precise interpretation of current perceptions. We exploratively apply our PMI to improve prevailing Transformers and CNN models on question-answering tasks like bAbI-20k and Sort-of-CLEVR datasets, as well as detecting equilateral triangles, language modeling and image classification tasks, and in each case, our PMI enhancements consistently outshine their original counterparts significantly. Visualization analyses reveal that relational memory consolidation, along with the interaction and integration of information from diverse memory sources, substantially contributes to the model effectiveness on inference tasks.
Authors: Zongyu Wu, Hongcheng Gao, Yueze Wang, Xiang Zhang, Suhang Wang
Abstract: Text-to-Image (T2I) models have shown great performance in generating images based on textual prompts. However, these models are vulnerable to unsafe input to generate unsafe content like sexual, harassment and illegal-activity images. Existing studies based on image checker, model fine-tuning and embedding blocking are impractical in real-world applications. Hence, we propose the first universal prompt optimizer for safe T2I (POSI) generation in black-box scenario. We first construct a dataset consisting of toxic-clean prompt pairs by GPT-3.5 Turbo. To guide the optimizer to have the ability of converting toxic prompt to clean prompt while preserving semantic information, we design a novel reward function measuring toxicity and text alignment of generated images and train the optimizer through Proximal Policy Optimization. Experiments show that our approach can effectively reduce the likelihood of various T2I models in generating inappropriate images, with no significant impact on text alignment. It is also flexible to be combined with methods to achieve better performance. Our code is available at https://github.com/wzongyu/POSI.
Authors: Detai Xin, Xu Tan, Kai Shen, Zeqian Ju, Dongchao Yang, Yuancheng Wang, Shinnosuke Takamichi, Hiroshi Saruwatari, Shujie Liu, Jinyu Li, Sheng Zhao
Abstract: We present RALL-E, a robust language modeling method for text-to-speech (TTS) synthesis. While previous work based on large language models (LLMs) shows impressive performance on zero-shot TTS, such methods often suffer from poor robustness, such as unstable prosody (weird pitch and rhythm/duration) and a high word error rate (WER), due to the autoregressive prediction style of language models. The core idea behind RALL-E is chain-of-thought (CoT) prompting, which decomposes the task into simpler steps to enhance the robustness of LLM-based TTS. To accomplish this idea, RALL-E first predicts prosody features (pitch and duration) of the input text and uses them as intermediate conditions to predict speech tokens in a CoT style. Second, RALL-E utilizes the predicted duration prompt to guide the computing of self-attention weights in Transformer to enforce the model to focus on the corresponding phonemes and prosody features when predicting speech tokens. Results of comprehensive objective and subjective evaluations demonstrate that, compared to a powerful baseline method VALL-E, RALL-E significantly improves the WER of zero-shot TTS from $5.6\%$ (without reranking) and $1.7\%$ (with reranking) to $2.5\%$ and $1.0\%$, respectively. Furthermore, we demonstrate that RALL-E correctly synthesizes sentences that are hard for VALL-E and reduces the error rate from $68\%$ to $4\%$.
Authors: Haoting Zhang, Jinghai He, Rhonda Righter, Zeyu Zheng
Abstract: With the advancement in generative language models, the selection of prompts has gained significant attention in recent years. A prompt is an instruction or description provided by the user, serving as a guide for the generative language model in content generation. Despite existing methods for prompt selection that are based on human labor, we consider facilitating this selection through simulation optimization, aiming to maximize a pre-defined score for the selected prompt. Specifically, we propose a two-stage framework. In the first stage, we determine a feasible set of prompts in sufficient numbers, where each prompt is represented by a moderate-dimensional vector. In the subsequent stage for evaluation and selection, we construct a surrogate model of the score regarding the moderate-dimensional vectors that represent the prompts. We propose sequentially selecting the prompt for evaluation based on this constructed surrogate model. We prove the consistency of the sequential evaluation procedure in our framework. We also conduct numerical experiments to demonstrate the efficacy of our proposed framework, providing practical instructions for implementation.
Authors: Yang Bai, Ge Pei, Jindong Gu, Yong Yang, Xingjun Ma
Abstract: Large language models (LLMs) have achieved remarkable performance on a wide range of tasks. However, recent studies have shown that LLMs can memorize training data and simple repeated tokens can trick the model to leak the data. In this paper, we take a step further and show that certain special characters or their combinations with English letters are stronger memory triggers, leading to more severe data leakage. The intuition is that, since LLMs are trained with massive data that contains a substantial amount of special characters (e.g. structural symbols {, } of JSON files, and @, # in emails and online posts), the model may memorize the co-occurrence between these special characters and the raw texts. This motivates us to propose a simple but effective Special Characters Attack (SCA) to induce training data leakage. Our experiments verify the high effectiveness of SCA against state-of-the-art LLMs: they can leak diverse training data, such as code corpus, web pages, and personally identifiable information, and sometimes generate non-stop outputs as a byproduct. We further show that the composition of the training data corpus can be revealed by inspecting the leaked data -- one crucial piece of information for pre-training high-performance LLMs. Our work can help understand the sensitivity of LLMs to special characters and identify potential areas for improvement.
Authors: Suyuan Zhao, Jiahuan Zhang, Yizhen Luo, Yushuai Wu, Zaiqing Nie
Abstract: Cell identity encompasses various semantic aspects of a cell, including cell type, pathway information, disease information, and more, which are essential for biologists to gain insights into its biological characteristics. Understanding cell identity from the transcriptomic data, such as annotating cell types, have become an important task in bioinformatics. As these semantic aspects are determined by human experts, it is impossible for AI models to effectively carry out cell identity understanding tasks without the supervision signals provided by single-cell and label pairs. The single-cell pre-trained language models (PLMs) currently used for this task are trained only on a single modality, transcriptomics data, lack an understanding of cell identity knowledge. As a result, they have to be fine-tuned for downstream tasks and struggle when lacking labeled data with the desired semantic labels. To address this issue, we propose an innovative solution by constructing a unified representation of single-cell data and natural language during the pre-training phase, allowing the model to directly incorporate insights related to cell identity. More specifically, we introduce LangCell, the first Language-Cell pre-training framework. LangCell utilizes texts enriched with cell identity information to gain a profound comprehension of cross-modal knowledge. Results from experiments conducted on different benchmarks show that LangCell is the only single-cell PLM that can work effectively in zero-shot cell identity understanding scenarios, and also significantly outperforms existing models in few-shot and fine-tuning cell identity understanding scenarios.
Authors: Hector Zenil, Felipe S. Abrah\~ao
Abstract: Based on the principles of information theory, measure theory, and theoretical computer science, we introduce a univariate signal deconvolution method with a wide range of applications to coding theory, particularly in zero-knowledge one-way communication channels, such as in deciphering messages from unknown generating sources about which no prior knowledge is available and to which no return message can be sent. Our multidimensional space reconstruction method from an arbitrary received signal is proven to be agnostic vis-a-vis the encoding-decoding scheme, computation model, programming language, formal theory, the computable (or semi-computable) method of approximation to algorithmic complexity, and any arbitrarily chosen (computable) probability measure of the events. The method derives from the principles of an approach to Artificial General Intelligence capable of building a general-purpose model of models independent of any arbitrarily assumed prior probability distribution. We argue that this optimal and universal method of decoding non-random data has applications to signal processing, causal deconvolution, topological and geometric properties encoding, cryptography, and bio- and technosignature detection.
Authors: Pranab Sahoo, Prabhash Meharia, Akash Ghosh, Sriparna Saha, Vinija Jain, Aman Chadha
Abstract: The rapid advancement of foundation models (FMs) across language, image, audio, and video domains has shown remarkable capabilities in diverse tasks. However, the proliferation of FMs brings forth a critical challenge: the potential to generate hallucinated outputs, particularly in high-stakes applications. The tendency of foundation models to produce hallucinated content arguably represents the biggest hindrance to their widespread adoption in real-world scenarios, especially in domains where reliability and accuracy are paramount. This survey paper presents a comprehensive overview of recent developments that aim to identify and mitigate the problem of hallucination in FMs, spanning text, image, video, and audio modalities. By synthesizing recent advancements in detecting and mitigating hallucination across various modalities, the paper aims to provide valuable insights for researchers, developers, and practitioners. Essentially, it establishes a clear framework encompassing definition, taxonomy, and detection strategies for addressing hallucination in multimodal foundation models, laying the foundation for future research in this pivotal area.