Authors: Yizhan Li, Sifan Wu, Christopher Smith, Thomas Lo, Bang Liu
Abstract: Writing clinical notes and documenting medical exams is a critical task for healthcare professionals, serving as a vital component of patient care documentation. However, manually writing these notes is time-consuming and can impact the amount of time clinicians can spend on direct patient interaction and other tasks. Consequently, the development of automated clinical note generation systems has emerged as a clinically meaningful area of research within AI for health. In this paper, we present three key contributions to the field of clinical note generation using large language models (LLMs). First, we introduce CliniKnote, a comprehensive dataset consisting of 1,200 complex doctor-patient conversations paired with their full clinical notes. This dataset, created and curated by medical experts with the help of modern neural networks, provides a valuable resource for training and evaluating models in clinical note generation tasks. Second, we propose the K-SOAP (Keyword, Subjective, Objective, Assessment, and Plan) note format, which enhances traditional SOAP~\cite{podder2023soap} (Subjective, Objective, Assessment, and Plan) notes by adding a keyword section at the top, allowing for quick identification of essential information. Third, we develop an automatic pipeline to generate K-SOAP notes from doctor-patient conversations and benchmark various modern LLMs using various metrics. Our results demonstrate significant improvements in efficiency and performance compared to standard LLM finetuning methods.
Authors: Ian Stewart, Sameera Horawalavithana, Brendan Kennedy, Sai Munikoti, Karl Pazdernik
Abstract: Multimodal foundation models (MFMs) such as OFASys show the potential to unlock analysis of complex data such as images, videos, and audio data via text prompts alone. However, their performance may suffer in the face of text input that differs even slightly from their training distribution, which is surprising considering the use of modality-specific data to "ground" the text input. This study demonstrates that prompt instability is a major concern for MFMs, leading to a consistent drop in performance across all modalities, but that instability can be mitigated with additional training with augmented data. We evaluate several methods for grounded prompt perturbation, where we generate perturbations and filter based on similarity to text and/or modality data. After re-training the models on the augmented data, we find improved accuracy and more stable performance on the perturbed test data regardless of perturbation condition, suggesting that the data augmentation strategy helps the models handle domain shifts more effectively. In error analysis, we find consistent patterns of performance improvement across domains, suggesting that retraining on prompt perturbations tends to help general reasoning capabilities in MFMs.
Authors: Dingyi Yang, Qin Jin
Abstract: With the development of artificial intelligence, particularly the success of Large Language Models (LLMs), the quantity and quality of automatically generated stories have significantly increased. This has led to the need for automatic story evaluation to assess the generative capabilities of computing systems and analyze the quality of both automatic-generated and human-written stories. Evaluating a story can be more challenging than other generation evaluation tasks. While tasks like machine translation primarily focus on assessing the aspects of fluency and accuracy, story evaluation demands complex additional measures such as overall coherence, character development, interestingness, etc. This requires a thorough review of relevant research. In this survey, we first summarize existing storytelling tasks, including text-to-text, visual-to-text, and text-to-visual. We highlight their evaluation challenges, identify various human criteria to measure stories, and present existing benchmark datasets. Then, we propose a taxonomy to organize evaluation metrics that have been developed or can be adopted for story evaluation. We also provide descriptions of these metrics, along with the discussion of their merits and limitations. Later, we discuss the human-AI collaboration for story evaluation and generation. Finally, we suggest potential future research directions, extending from story evaluation to general evaluations.
Authors: James Liu, Pragaash Ponnusamy, Tianle Cai, Han Guo, Yoon Kim, Ben Athiwaratkun
Abstract: Activation sparsity can enable practical inference speedups in large language models (LLMs) by reducing the compute and memory-movement required for matrix multiplications during the forward pass. However, existing methods face limitations that inhibit widespread adoption. Some approaches are tailored towards older models with ReLU-based sparsity, while others require extensive continued pre-training on up to hundreds of billions of tokens. This paper describes TEAL, a simple training-free method that applies magnitude-based activation sparsity to hidden states throughout the entire model. TEAL achieves 40-50% model-wide sparsity with minimal performance degradation across Llama-2, Llama-3, and Mistral families, with sizes varying from 7B to 70B. We improve existing sparse kernels and demonstrate wall-clock decoding speed-ups of up to 1.53$\times$ and 1.8$\times$ at 40% and 50% model-wide sparsity. TEAL is compatible with weight quantization, enabling further efficiency gains.
Authors: Haven Kim, Kahyun Choi
Abstract: This paper addresses the unique challenge of conducting research in lyric studies, where direct use of lyrics is often restricted due to copyright concerns. Unlike typical data, internet-sourced lyrics are frequently protected under copyright law, necessitating alternative approaches. Our study introduces a novel method for generating copyright-free lyrics from publicly available Bag-of-Words (BoW) datasets, which contain the vocabulary of lyrics but not the lyrics themselves. Utilizing metadata associated with BoW datasets and large language models, we successfully reconstructed lyrics. We have compiled and made available a dataset of reconstructed lyrics, LyCon, aligned with metadata from renowned sources including the Million Song Dataset, Deezer Mood Detection Dataset, and AllMusic Genre Dataset, available for public access. We believe that the integration of metadata such as mood annotations or genres enables a variety of academic experiments on lyrics, such as conditional lyric generation.
Authors: Haley Lepp, Parth Sarin
Abstract: In this provocation, we discuss the English dominance of the AI research community, arguing that the requirement for English language publishing upholds and reinforces broader regimes of extraction in AI. While large language models and machine translation have been celebrated as a way to break down barriers, we regard their use as a symptom of linguistic exclusion of scientists and potential readers. We propose alternative futures for a healthier publishing culture, organized around three themes: administering conferences in the languages of the country in which they are held, instructing peer reviewers not to adjudicate the language appropriateness of papers, and offering opportunities to publish and present in multiple languages. We welcome new translations of this piece. Please contact the authors if you would like to contribute one.
Authors: Yijie Jin
Abstract: Multimodal Sentiment Analysis (MSA) leverages multiple modals to analyze sentiments. Typically, advanced fusion methods and representation learning-based methods are designed to tackle it. Our proposed GSIFN solves two key problems to be solved in MSA: (i) In multimodal fusion, the decoupling of modal combinations and tremendous parameter redundancy in existing fusion methods, which lead to poor fusion performance and efficiency. (ii) The trade-off between representation capability and computation overhead of the unimodal feature extractors and enhancers. GSIFN incorporates two main components to solve these problems: (i) Graph-Structured and Interlaced-Masked Multimodal Transformer. It adopts the Interlaced Mask mechanism to construct robust multimodal graph embedding, achieve all-modal-in-one Transformer-based fusion, and greatly reduce the computation overhead. (ii) A self-supervised learning framework with low computation overhead and high performance, which utilizes a parallelized LSTM with matrix memory to enhance non-verbal modal feature for unimodal label generation. Evaluated on the MSA datasets CMU-MOSI, CMU-MOSEI, and CH-SIMS, GSIFN demonstrates superior performance with significantly lower computation overhead compared with state-of-the-art methods.
Authors: Abhay Gupta, Philip Meng, Ece Yurtseven, Sean O'Brien, Kevin Zhu
Abstract: Detecting biases in natural language understanding (NLU) for African American Vernacular English (AAVE) is crucial to developing inclusive natural language processing (NLP) systems. To address dialect-induced performance discrepancies, we introduce AAVENUE ({AAVE} {N}atural Language {U}nderstanding {E}valuation), a benchmark for evaluating large language model (LLM) performance on NLU tasks in AAVE and Standard American English (SAE). AAVENUE builds upon and extends existing benchmarks like VALUE, replacing deterministic syntactic and morphological transformations with a more flexible methodology leveraging LLM-based translation with few-shot prompting, improving performance across our evaluation metrics when translating key tasks from the GLUE and SuperGLUE benchmarks. We compare AAVENUE and VALUE translations using five popular LLMs and a comprehensive set of metrics including fluency, BARTScore, quality, coherence, and understandability. Additionally, we recruit fluent AAVE speakers to validate our translations for authenticity. Our evaluations reveal that LLMs consistently perform better on SAE tasks than AAVE-translated versions, underscoring inherent biases and highlighting the need for more inclusive NLP models. We have open-sourced our source code on GitHub and created a website to showcase our work at https://aavenue.live.
URLs: https://aavenue.live.
Authors: Hanna Abi Akl
Abstract: We introduce SHADOW, a fine-tuned language model trained on an intermediate task using associative deductive reasoning, and measure its performance on a knowledge base construction task using Wikidata triple completion. We evaluate SHADOW on the LM-KBC 2024 challenge and show that it outperforms the baseline solution by 20% with a F1 score of 68.72%.
Authors: Yuhao Du, Zhuo Li, Pengyu Cheng, Xiang Wan, Anningzhe Gao
Abstract: Large Language Models (LLMs) have become a focal point in the rapidly evolving field of artificial intelligence. However, a critical concern is the presence of toxic content within the pre-training corpus of these models, which can lead to the generation of inappropriate outputs. Investigating methods for detecting internal faults in LLMs can help us understand their limitations and improve their security. Existing methods primarily focus on jailbreaking attacks, which involve manually or automatically constructing adversarial content to prompt the target LLM to generate unexpected responses. These methods rely heavily on prompt engineering, which is time-consuming and usually requires specially designed questions. To address these challenges, this paper proposes a target-driven attack paradigm that focuses on directly eliciting the target response instead of optimizing the prompts. We introduce the use of another LLM as the detector for toxic content, referred to as ToxDet. Given a target toxic response, ToxDet can generate a possible question and a preliminary answer to provoke the target model into producing desired toxic responses with meanings equivalent to the provided one. ToxDet is trained by interacting with the target LLM and receiving reward signals from it, utilizing reinforcement learning for the optimization process. While the primary focus of the target models is on open-source LLMs, the fine-tuned ToxDet can also be transferred to attack black-box models such as GPT-4o, achieving notable results. Experimental results on AdvBench and HH-Harmless datasets demonstrate the effectiveness of our methods in detecting the tendencies of target LLMs to generate harmful responses. This algorithm not only exposes vulnerabilities but also provides a valuable resource for researchers to strengthen their models against such attacks.
Authors: Hongfu Liu, Yuxi Xie, Ye Wang, Michael Shieh
Abstract: Language Language Models (LLMs) face safety concerns due to potential misuse by malicious users. Recent red-teaming efforts have identified adversarial suffixes capable of jailbreaking LLMs using the gradient-based search algorithm Greedy Coordinate Gradient (GCG). However, GCG struggles with computational inefficiency, limiting further investigations regarding suffix transferability and scalability across models and data. In this work, we bridge the connection between search efficiency and suffix transferability. We propose a two-stage transfer learning framework, DeGCG, which decouples the search process into behavior-agnostic pre-searching and behavior-relevant post-searching. Specifically, we employ direct first target token optimization in pre-searching to facilitate the search process. We apply our approach to cross-model, cross-data, and self-transfer scenarios. Furthermore, we introduce an interleaved variant of our approach, i-DeGCG, which iteratively leverages self-transferability to accelerate the search process. Experiments on HarmBench demonstrate the efficiency of our approach across various models and domains. Notably, our i-DeGCG outperforms the baseline on Llama2-chat-7b with ASRs of $43.9$ ($+22.2$) and $39.0$ ($+19.5$) on valid and test sets, respectively. Further analysis on cross-model transfer indicates the pivotal role of first target token optimization in leveraging suffix transferability for efficient searching.
Authors: Han Xia, Songyang Gao, Qiming Ge, Zhiheng Xi, Qi Zhang, Xuanjing Huang
Abstract: Reinforcement Learning from Human Feedback (RLHF) has proven effective in aligning large language models with human intentions, yet it often relies on complex methodologies like Proximal Policy Optimization (PPO) that require extensive hyper-parameter tuning and present challenges in sample efficiency and stability. In this paper, we introduce Inverse-Q*, an innovative framework that transcends traditional RL methods by optimizing token-level reinforcement learning without the need for additional reward or value models. Inverse-Q* leverages direct preference optimization techniques but extends them by estimating the conditionally optimal policy directly from the model's responses, facilitating more granular and flexible policy shaping. Our approach reduces reliance on human annotation and external supervision, making it especially suitable for low-resource settings. We present extensive experimental results demonstrating that Inverse-Q* not only matches but potentially exceeds the effectiveness of PPO in terms of convergence speed and the alignment of model responses with human preferences. Our findings suggest that Inverse-Q* offers a practical and robust alternative to conventional RLHF approaches, paving the way for more efficient and adaptable model training approaches.
Authors: Zhu Li, Xiyuan Gao, Yuqing Zhang, Shekhar Nayak, Matt Coler
Abstract: This study investigates the acoustic features of sarcasm and disentangles the interplay between the propensity of an utterance being used sarcastically and the presence of prosodic cues signaling sarcasm. Using a dataset of sarcastic utterances compiled from television shows, we analyze the prosodic features within utterances and key phrases belonging to three distinct sarcasm categories (embedded, propositional, and illocutionary), which vary in the degree of semantic cues present, and compare them to neutral expressions. Results show that in phrases where the sarcastic meaning is salient from the semantics, the prosodic cues are less relevant than when the sarcastic meaning is not evident from the semantics, suggesting a trade-off between prosodic and semantic cues of sarcasm at the phrase level. These findings highlight a lessened reliance on prosodic modulation in semantically dense sarcastic expressions and a nuanced interaction that shapes the communication of sarcastic intent.
Authors: Melisa Russak, Umar Jamil, Christopher Bryant, Kiran Kamble, Axel Magnuson, Mateusz Russak, Waseem AlShikh
Abstract: In this paper, we introduce Writing in the Margins (WiM), a new inference pattern for Large Language Models designed to optimize the handling of long input sequences in retrieval-oriented tasks. This approach leverages the chunked prefill of the key-value cache to perform segment-wise inference, which enables efficient processing of extensive contexts along with the generation and classification of intermediate information ("margins") that guide the model towards specific tasks. This method increases computational overhead marginally while significantly enhancing the performance of off-the-shelf models without the need for fine-tuning. Specifically, we observe that WiM provides an average enhancement of 7.5% in accuracy for reasoning skills (HotpotQA, MultiHop-RAG) and more than a 30.0% increase in the F1-score for aggregation tasks (CWE). Additionally, we show how the proposed pattern fits into an interactive retrieval design that provides end-users with ongoing updates about the progress of context processing, and pinpoints the integration of relevant information into the final response. We release our implementation of WiM using Hugging Face Transformers library at https://github.com/writer/writing-in-the-margins.
Authors: Shuaijie Shen, Chao Wang, Renzhuo Huang, Yan Zhong, Qinghai Guo, Zhichao Lu, Jianguo Zhang, Luziwei Leng
Abstract: Known as low energy consumption networks, spiking neural networks (SNNs) have gained a lot of attention within the past decades. While SNNs are increasing competitive with artificial neural networks (ANNs) for vision tasks, they are rarely used for long sequence tasks, despite their intrinsic temporal dynamics. In this work, we develop spiking state space models (SpikingSSMs) for long sequence learning by leveraging on the sequence learning abilities of state space models (SSMs). Inspired by dendritic neuron structure, we hierarchically integrate neuronal dynamics with the original SSM block, meanwhile realizing sparse synaptic computation. Furthermore, to solve the conflict of event-driven neuronal dynamics with parallel computing, we propose a light-weight surrogate dynamic network which accurately predicts the after-reset membrane potential and compatible to learnable thresholds, enabling orders of acceleration in training speed compared with conventional iterative methods. On the long range arena benchmark task, SpikingSSM achieves competitive performance to state-of-the-art SSMs meanwhile realizing on average 90\% of network sparsity. On language modeling, our network significantly surpasses existing spiking large language models (spikingLLMs) on the WikiText-103 dataset with only a third of the model size, demonstrating its potential as backbone architecture for low computation cost LLMs.
Authors: Ayomide Odumakinde, Daniel D'souza, Pat Verga, Beyza Ermis, Sara Hooker
Abstract: The use of synthetic data has played a critical role in recent state-of-art breakthroughs. However, overly relying on a single oracle teacher model to generate data has been shown to lead to model collapse and invite propagation of biases. These limitations are particularly evident in multilingual settings, where the absence of a universally effective teacher model that excels across all languages presents significant challenges. In this work, we address these extreme difference by introducing "multilingual arbitrage", which capitalizes on performance variations between multiple models for a given language. To do so, we strategically route samples through a diverse pool of models, each with unique strengths in different languages. Across exhaustive experiments on state-of-art models, our work suggests that arbitrage techniques allow for spectacular gains in performance that far outperform relying on a single teacher. In particular, compared to the best single teacher, we observe gains of up to 56.5% improvement in win rates averaged across all languages when switching to multilingual arbitrage. We observe the most significant gains for the least resourced languages in our pool.
Authors: Chi-Min Chan, Jianxuan Yu, Weize Chen, Chunyang Jiang, Xinyu Liu, Weijie Shi, Zhiyuan Liu, Wei Xue, Yike Guo
Abstract: The rapid advancement of large language models (LLMs) has led to the rise of LLM-based agents. Recent research shows that multi-agent systems (MAS), where each agent plays a specific role, can outperform individual LLMs. However, configuring an MAS for a task remains challenging, with performance only observable post-execution. Inspired by scaling laws in LLM development, we investigate whether MAS performance can be predicted beforehand. We introduce AgentMonitor, a framework that integrates at the agent level to capture inputs and outputs, transforming them into statistics for training a regression model to predict task performance. Additionally, it can further apply real-time corrections to address security risks posed by malicious agents, mitigating negative impacts and enhancing MAS security. Experiments demonstrate that an XGBoost model achieves a Spearman correlation of 0.89 in-domain and 0.58 in more challenging scenarios. Furthermore, using AgentMonitor reduces harmful content by 6.2% and increases helpful content by 1.8% on average, enhancing safety and reliability. Code is available at \url{https://github.com/chanchimin/AgentMonitor}.
Authors: Shruti Singh, Muskaan Singh, Virender Kadyan
Abstract: Transformers have evolved with great success in various artificial intelligence tasks. Thanks to our recent prevalence of self-attention mechanisms, which capture long-term dependency, phenomenal outcomes in speech processing and recognition tasks have been produced. The paper presents a comprehensive survey of transformer techniques oriented in speech modality. The main contents of this survey include (1) background of traditional ASR, end-to-end transformer ecosystem, and speech transformers (2) foundational models in a speech via lingualism paradigm, i.e., monolingual, bilingual, multilingual, and cross-lingual (3) dataset and languages, acoustic features, architecture, decoding, and evaluation metric from a specific topological lingualism perspective (4) popular speech transformer toolkit for building end-to-end ASR systems. Finally, highlight the discussion of open challenges and potential research directions for the community to conduct further research in this domain.
Authors: Haowei Du, Huishuai Zhang, Dongyan Zhao
Abstract: To address the hallucination in generative question answering (GQA) where the answer can not be derived from the document, we propose a novel evidence-enhanced triplet generation framework, EATQA, encouraging the model to predict all the combinations of (Question, Evidence, Answer) triplet by flipping the source pair and the target label to understand their logical relationships, i.e., predict Answer(A), Question(Q), and Evidence(E) given a QE, EA, and QA pairs, respectively. Furthermore, we bridge the distribution gap to distill the knowledge from evidence in inference stage. Our framework ensures the model to learn the logical relation between query, evidence and answer, which simultaneously improves the evidence generation and query answering. In this paper, we apply EATQA to LLama and it outperforms other LLMs-based methods and hallucination mitigation approaches on two challenging GQA benchmarks. Further analysis shows that our method not only keeps prior knowledge within LLM, but also mitigates hallucination and generates faithful answers.
Authors: Wazir Ali, Sampo Pyysalo
Abstract: Large Language Models (LLMs) have gained significant attention due to their high performance on a wide range of natural language tasks since the release of ChatGPT. The LLMs learn to understand and generate language by training billions of model parameters on vast volumes of text data. Despite being a relatively new field, LLM research is rapidly advancing in various directions. In this paper, we present an overview of LLM families, including LLaMA, PaLM, GPT, and MoE, and the methods developed to create and enhance LLMs for official European Union (EU) languages. We provide a comprehensive summary of common monolingual and multilingual datasets used for pretraining LLMs.
Authors: Yuyin Lu, Hegang Chen, Pengbo Mao, Yanghui Rao, Haoran Xie, Fu Lee Wang, Qing Li
Abstract: Topic taxonomy discovery aims at uncovering topics of different abstraction levels and constructing hierarchical relations between them. Unfortunately, most of prior work can hardly model semantic scopes of words and topics by holding the Euclidean embedding space assumption. What's worse, they infer asymmetric hierarchical relations by symmetric distances between topic embeddings. As a result, existing methods suffer from problems of low-quality topics at high abstraction levels and inaccurate hierarchical relations. To alleviate these problems, this paper develops a Box embedding-based Topic Model (BoxTM) that maps words and topics into the box embedding space, where the asymmetric metric is defined to properly infer hierarchical relations among topics. Additionally, our BoxTM explicitly infers upper-level topics based on correlation between specific topics through recursive clustering on topic boxes. Finally, extensive experiments validate high-quality of the topic taxonomy learned by BoxTM.
Authors: Guosheng Dong, Da Pan, Yiding Sun, Shusen Zhang, Zheng Liang, Xin Wu, Yanjun Shen, Fan Yang, Haoze Sun, Tianpeng Li, Mingan Lin, Jianhua Xu, Yufan Zhang, Xiaonan Nie, Lei Su, Bingning Wang, Wentao Zhang, Jiaxin Mao, Zenan Zhou, Weipeng Chen
Abstract: The general capabilities of Large Language Models (LLM) highly rely on the composition and selection on extensive pretraining datasets, treated as commercial secrets by several institutions. To mitigate this issue, we open-source the details of a universally applicable data processing pipeline and validate its effectiveness and potential by introducing a competitive LLM baseline. Specifically, the data processing pipeline consists of broad collection to scale up and reweighting to improve quality. We then pretrain a 7B model BaichuanSEED with 3T tokens processed by our pipeline without any deliberate downstream task-related optimization, followed by an easy but effective supervised fine-tuning stage. BaichuanSEED demonstrates consistency and predictability throughout training and achieves comparable performance on comprehensive benchmarks with several commercial advanced large language models, such as Qwen1.5 and Llama3. We also conduct several heuristic experiments to discuss the potential for further optimization of downstream tasks, such as mathematics and coding.
Authors: Xiyu Liu, Zhengxiao Liu, Naibin Gu, Zheng Lin, Wanli Ma, Ji Xiang, Weiping Wang
Abstract: The storage and recall of factual associations in auto-regressive transformer language models (LMs) have drawn a great deal of attention, inspiring knowledge editing by directly modifying the located model weights. Most editing works achieve knowledge editing under the guidance of existing interpretations of knowledge recall that mainly focus on subject knowledge. However, these interpretations are seriously flawed, neglecting relation information and leading to the over-generalizing problem for editing. In this work, we discover a novel relation-focused perspective to interpret the knowledge recall of transformer LMs during inference and apply it on knowledge editing to avoid over-generalizing. Experimental results on the dataset supplemented with a new R-Specificity criterion demonstrate that our editing approach significantly alleviates over-generalizing while remaining competitive on other criteria, breaking the domination of subject-focused editing for future research.
Authors: N. E. Kriman
Abstract: The use of large language models (LLMs) has significantly increased since the introduction of ChatGPT in 2022, demonstrating their value across various applications. However, a major challenge for enterprise and commercial adoption of LLMs is their tendency to generate inaccurate information, a phenomenon known as "hallucination." This project proposes a method for estimating the factuality of a summary generated by LLMs when compared to a source text. Our approach utilizes Naive Bayes classification to assess the accuracy of the content produced.
Authors: Kristina Gligori\'c, Tijana Zrnic, Cinoo Lee, Emmanuel J. Cand\`es, Dan Jurafsky
Abstract: Large language models (LLMs) have shown high agreement with human raters across a variety of tasks, demonstrating potential to ease the challenges of human data collection. In computational social science (CSS), researchers are increasingly leveraging LLM annotations to complement slow and expensive human annotations. Still, guidelines for collecting and using LLM annotations, without compromising the validity of downstream conclusions, remain limited. We introduce Confidence-Driven Inference: a method that combines LLM annotations and LLM confidence indicators to strategically select which human annotations should be collected, with the goal of producing accurate statistical estimates and provably valid confidence intervals while reducing the number of human annotations needed. Our approach comes with safeguards against LLM annotations of poor quality, guaranteeing that the conclusions will be both valid and no less accurate than if we only relied on human annotations. We demonstrate the effectiveness of Confidence-Driven Inference over baselines in statistical estimation tasks across three CSS settings--text politeness, stance, and bias--reducing the needed number of human annotations by over 25% in each. Although we use CSS settings for demonstration, Confidence-Driven Inference can be used to estimate most standard quantities across a broad range of NLP problems.
Authors: Olaf van der Veen, Semir Dzebo, Levi Littvay, Kirk Hawkins, Oren Dar
Abstract: Populism is a concept that is often used but notoriously difficult to measure. Common qualitative measurements like holistic grading or content analysis require great amounts of time and labour, making it difficult to quickly scope out which politicians should be classified as populist and which should not, while quantitative methods show mixed results when it comes to classifying populist rhetoric. In this paper, we develop a pipeline to train and validate an automated classification model to estimate the use of populist language. We train models based on sentences that were identified as populist and pluralist in 300 US governors' speeches from 2010 to 2018 and in 45 speeches of presidential candidates in 2016. We find that these models classify most speeches correctly, including 84% of governor speeches and 89% of presidential speeches. These results extend to different time periods (with 92% accuracy on more recent American governors), different amounts of data (with as few as 70 training sentences per category achieving similar results), and when classifying politicians instead of individual speeches. This pipeline is thus an effective tool that can optimise the systematic and swift classification of the use of populist language in politicians' speeches.
Authors: Yucheng Jiang, Yijia Shao, Dekun Ma, Sina J. Semnani, Monica S. Lam
Abstract: While language model (LM)-powered chatbots and generative search engines excel at answering concrete queries, discovering information in the terrain of unknown unknowns remains challenging for users. To emulate the common educational scenario where children/students learn by listening to and participating in conversations of their parents/teachers, we create Collaborative STORM (Co-STORM). Unlike QA systems that require users to ask all the questions, Co-STORM lets users observe and occasionally steer the discourse among several LM agents. The agents ask questions on the user's behalf, allowing the user to discover unknown unknowns serendipitously. To facilitate user interaction, Co-STORM assists users in tracking the discourse by organizing the uncovered information into a dynamic mind map, ultimately generating a comprehensive report as takeaways. For automatic evaluation, we construct the WildSeek dataset by collecting real information-seeking records with user goals. Co-STORM outperforms baseline methods on both discourse trace and report quality. In a further human evaluation, 70% of participants prefer Co-STORM over a search engine, and 78% favor it over a RAG chatbot.
Authors: Chidaksh Ravuru, Sagar Srinivas Sakhinana, Venkataramana Runkana
Abstract: Time series modeling is crucial for many applications, however, it faces challenges such as complex spatio-temporal dependencies and distribution shifts in learning from historical context to predict task-specific outcomes. To address these challenges, we propose a novel approach using an agentic Retrieval-Augmented Generation (RAG) framework for time series analysis. The framework leverages a hierarchical, multi-agent architecture where the master agent orchestrates specialized sub-agents and delegates the end-user request to the relevant sub-agent. The sub-agents utilize smaller, pre-trained language models (SLMs) customized for specific time series tasks through fine-tuning using instruction tuning and direct preference optimization, and retrieve relevant prompts from a shared repository of prompt pools containing distilled knowledge about historical patterns and trends to improve predictions on new data. Our proposed modular, multi-agent RAG approach offers flexibility and achieves state-of-the-art performance across major time series tasks by tackling complex challenges more effectively than task-specific customized methods across benchmark datasets.
Authors: Dibaloke Chanda, Milan Aryal, Nasim Yahya Soltani, Masoud Ganji
Abstract: Recent advances in deep learning have completely transformed the domain of computational pathology (CPath), which in turn altered the diagnostic workflow of pathologists by integrating foundation models (FMs) and vision-language models (VLMs) in their assessment and decision-making process. FMs overcome the limitations of existing deep learning approaches in CPath by learning a representation space that can be adapted to a wide variety of downstream tasks without explicit supervision. VLMs allow pathology reports written in natural language to be used as a rich semantic information source to improve existing models as well as generate predictions in natural language form. In this survey, a holistic and systematic overview of recent innovations in FMs and VLMs in CPath is presented. Furthermore, the tools, datasets and training schemes for these models are summarized in addition to categorizing them into distinct groups. This extensive survey highlights the current trends in CPath and the way it is going to be transformed through FMs and VLMs in the future.
Authors: Hao Wang, Jindong Han, Wei Fan, Hao Liu
Abstract: Spatio-temporal time series forecasting plays a critical role in various real-world applications, such as transportation optimization, energy management, and climate analysis. The recent advancements in Pre-trained Language Models (PLMs) have inspired efforts to reprogram these models for time series forecasting tasks, by leveraging their superior reasoning and generalization capabilities. However, existing approaches fall short in handling complex spatial inter-series dependencies and intrinsic intra-series frequency components, limiting their spatio-temporal forecasting performance. Moreover, the linear mapping of continuous time series to a compressed subset vocabulary in reprogramming constrains the spatio-temporal semantic expressivity of PLMs and may lead to potential information bottleneck. To overcome the above limitations, we propose \textsc{RePST}, a tailored PLM reprogramming framework for spatio-temporal forecasting. The key insight of \textsc{RePST} is to decouple the spatio-temporal dynamics in the frequency domain, allowing better alignment with the PLM text space. Specifically, we first decouple spatio-temporal data in Fourier space and devise a structural diffusion operator to obtain temporal intrinsic and spatial diffusion signals, making the dynamics more comprehensible and predictable for PLMs. To avoid information bottleneck from a limited vocabulary, we further propose a discrete reprogramming strategy that selects relevant discrete textual information from an expanded vocabulary space in a differentiable manner. Extensive experiments on four real-world datasets show that our proposed approach significantly outperforms state-of-the-art spatio-temporal forecasting models, particularly in data-scarce scenarios.
Authors: Xinyang Hu, Fengzhuo Zhang, Siyu Chen, Zhuoran Yang
Abstract: Chain-of-Thought (CoT) prompting and its variants have gained popularity as effective methods for solving multi-step reasoning problems using pretrained large language models (LLMs). In this work, we analyze CoT prompting from a statistical estimation perspective, providing a comprehensive characterization of its sample complexity. To this end, we introduce a multi-step latent variable model that encapsulates the reasoning process, where the latent variable encodes the task information. Under this framework, we demonstrate that when the pretraining dataset is sufficiently large, the estimator formed by CoT prompting is equivalent to a Bayesian estimator. This estimator effectively solves the multi-step reasoning problem by aggregating a posterior distribution inferred from the demonstration examples in the prompt. Moreover, we prove that the statistical error of the CoT estimator can be decomposed into two main components: (i) a prompting error, which arises from inferring the true task using CoT prompts, and (ii) the statistical error of the pretrained LLM. We establish that, under appropriate assumptions, the prompting error decays exponentially to zero as the number of demonstrations increases. Additionally, we explicitly characterize the approximation and generalization errors of the pretrained LLM. Notably, we construct a transformer model that approximates the target distribution of the multi-step reasoning problem with an error that decreases exponentially in the number of transformer blocks. Our analysis extends to other variants of CoT, including Self-Consistent CoT, Tree-of-Thought, and Selection-Inference, offering a broad perspective on the efficacy of these methods. We also provide numerical experiments to validate the theoretical findings.
Authors: Duo Wang, Yuan Zuo, Fengzhi Li, Junjie Wu
Abstract: Zero-shot graph machine learning, especially with graph neural networks (GNNs), has garnered significant interest due to the challenge of scarce labeled data. While methods like self-supervised learning and graph prompt learning have been extensively explored, they often rely on fine-tuning with task-specific labels, limiting their effectiveness in zero-shot scenarios. Inspired by the zero-shot capabilities of instruction-fine-tuned large language models (LLMs), we introduce a novel framework named Token Embedding-Aligned Graph Language Model (TEA-GLM) that leverages LLMs as cross-dataset and cross-task zero-shot learners for graph machine learning. Concretely, we pretrain a GNN, aligning its representations with token embeddings of an LLM. We then train a linear projector that transforms the GNN's representations into a fixed number of graph token embeddings without tuning the LLM. A unified instruction is designed for various graph tasks at different levels, such as node classification (node-level) and link prediction (edge-level). These design choices collectively enhance our method's effectiveness in zero-shot learning, setting it apart from existing methods. Experiments show that our graph token embeddings help the LLM predictor achieve state-of-the-art performance on unseen datasets and tasks compared to other methods using LLMs as predictors.
Authors: Nicholas Moratelli, Davide Caffagni, Marcella Cornia, Lorenzo Baraldi, Rita Cucchiara
Abstract: The conventional training approach for image captioning involves pre-training a network using teacher forcing and subsequent fine-tuning with Self-Critical Sequence Training to maximize hand-crafted captioning metrics. However, when attempting to optimize modern and higher-quality metrics like CLIP-Score and PAC-Score, this training method often encounters instability and fails to acquire the genuine descriptive capabilities needed to produce fluent and informative captions. In this paper, we propose a new training paradigm termed Direct CLIP-Based Optimization (DiCO). Our approach jointly learns and optimizes a reward model that is distilled from a learnable captioning evaluator with high human correlation. This is done by solving a weighted classification problem directly inside the captioner. At the same time, DiCO prevents divergence from the original model, ensuring that fluency is maintained. DiCO not only exhibits improved stability and enhanced quality in the generated captions but also aligns more closely with human preferences compared to existing methods, especially in modern metrics. Additionally, it maintains competitive performance in traditional metrics. Our source code and trained models are publicly available at https://github.com/aimagelab/DiCO.
Authors: Muhammad Fawi
Abstract: This paper introduces CURLoRA, a novel approach to fine-tuning large language models (LLMs) that leverages CUR matrix decomposition in the context of Low-Rank Adaptation (LoRA). Our method addresses two critical challenges in LLM fine-tuning: mitigating catastrophic forgetting during continual learning and reducing the number of trainable parameters. We propose a unique modification to the CUR decomposition process, utilizing inverted probabilities for column and row selection which acts as an implicit regularization, and initializing the $U$ matrix as a zero matrix, and only fine-tuning it. We demonstrate through experiments on multiple datasets that CURLoRA outperforms standard LoRA in mitigating catastrophic forgetting. It maintains model stability and performance across tasks while significantly reducing the number of trainable parameters. Our results show that CURLoRA achieves very good and stable task accuracy while maintaining base model's perplexity scores fixed compared to LoRA upon continual fine-tuning, particularly in scenarios with limited data.
Authors: Yingqiang Gao, Jhony Prada, Nianlong Gu, Jessica Lam, Richard H. R. Hahnloser
Abstract: Large Language Models (LLMs) produce eloquent texts but often the content they generate needs to be verified. Traditional information retrieval systems can assist with this task, but most systems have not been designed with LLM-generated queries in mind. As such, there is a compelling need for integrated systems that provide both retrieval and generation functionality within a single user interface. We present MODOC, a modular user interface that leverages the capabilities of LLMs and provides assistance with detecting their confabulations, promoting integrity in scientific writing. MODOC represents a significant step forward in scientific writing assistance. Its modular architecture supports flexible functions for retrieving information and for writing and generating text in a single, user-friendly interface.
Authors: Kate Lin, Tarfah Alrashed, Natasha Noy
Abstract: The Web today has millions of datasets, and the number of datasets continues to grow at a rapid pace. These datasets are not standalone entities; rather, they are intricately connected through complex relationships. Semantic relationships between datasets provide critical insights for research and decision-making processes. In this paper, we study dataset relationships from the perspective of users who discover, use, and share datasets on the Web: what relationships are important for different tasks? What contextual information might users want to know? We first present a comprehensive taxonomy of relationships between datasets on the Web and map these relationships to user tasks performed during dataset discovery. We develop a series of methods to identify these relationships and compare their performance on a large corpus of datasets generated from Web pages with schema.org markup. We demonstrate that machine-learning based methods that use dataset metadata achieve multi-class classification accuracy of 90%. Finally, we highlight gaps in available semantic markup for datasets and discuss how incorporating comprehensive semantics can facilitate the identification of dataset relationships. By providing a comprehensive overview of dataset relationships at scale, this paper sets a benchmark for future research.
Authors: Cherag Aroraa, Tracy Holloway King, Jayant Kumar, Yi Lu, Sanat Sharma, Arvind Srikantan, David Uvalle, Josep Valls-Vargas, Harsha Vardhan
Abstract: As user content and queries become increasingly multi-modal, the need for effective multi-modal search systems has grown. Traditional search systems often rely on textual and metadata annotations for indexed images, while multi-modal embeddings like CLIP enable direct search using text and image embeddings. However, embedding-based approaches face challenges in integrating contextual features such as user locale and recency. Building a scalable multi-modal search system requires fine-tuning several components. This paper presents a multi-modal search architecture and a series of AB tests that optimize embeddings and multi-modal technologies in Adobe Express template search. We address considerations such as embedding model selection, the roles of embeddings in matching and ranking, and the balance between dense and sparse embeddings. Our iterative approach demonstrates how utilizing sparse, dense, and contextual features enhances short and long query search, significantly reduces null rates (over 70\%), and increases click-through rates (CTR). Our findings provide insights into developing robust multi-modal search systems, thereby enhancing relevance for complex queries.
Authors: Yijiang Liu, Huanrui Yang, Youxin Chen, Rongyu Zhang, Miao Wang, Yuan Du, Li Du
Abstract: Large language models (LLMs) excel in language tasks, especially with supervised fine-tuning after pre-training. However, their substantial memory and computational requirements hinder practical applications. Structural pruning, which reduces less significant weight dimensions, is one solution. Yet, traditional post-hoc pruning often leads to significant performance loss, with limited recovery from further fine-tuning due to reduced capacity. Since the model fine-tuning refines the general and chaotic knowledge in pre-trained models, we aim to incorporate structural pruning with the fine-tuning, and propose the Pruning-Aware Tuning (PAT) paradigm to eliminate model redundancy while preserving the model performance to the maximum extend. Specifically, we insert the innovative Hybrid Sparsification Modules (HSMs) between the Attention and FFN components to accordingly sparsify the upstream and downstream linear modules. The HSM comprises a lightweight operator and a globally shared trainable mask. The lightweight operator maintains a training overhead comparable to that of LoRA, while the trainable mask unifies the channels to be sparsified, ensuring structural pruning. Additionally, we propose the Identity Loss which decouples the transformation and scaling properties of the HSMs to enhance training robustness. Extensive experiments demonstrate that PAT excels in both performance and efficiency. For example, our Llama2-7b model with a 25\% pruning ratio achieves 1.33$\times$ speedup while outperforming the LoRA-finetuned model by up to 1.26\% in accuracy with a similar training cost. Code: https://github.com/kriskrisliu/PAT_Pruning-Aware-Tuning
URLs: https://github.com/kriskrisliu/PAT_Pruning-Aware-Tuning
Authors: Simran Kaur, Simon Park, Anirudh Goyal, Sanjeev Arora
Abstract: We introduce Instruct-SkillMix, an automated approach for creating diverse, high quality SFT data. The Instruct-SkillMix pipeline involves two stages, each leveraging an existing powerful LLM: (1) Skill extraction: uses the LLM to extract core "skills" for instruction-following, either from existing datasets, or by directly prompting the model; (2) Data generation: uses the powerful LLM to generate (instruction, response) data that exhibit a randomly chosen pair of these skills. Here, the use of random skill combinations promotes diversity and difficulty. Vanilla SFT (i.e., no PPO, DPO, or RL methods) on data generated from Instruct-SkillMix leads to strong gains on instruction following benchmarks such as AlpacaEval 2.0, MT-Bench, and WildBench. With just $4$K examples, LLaMA-3-8B-Base achieves 42.76% length-controlled win rate on AlpacaEval 2.0. To our knowledge, this achieves state-of-the-art performance among all models that have only undergone SFT (no RL methods) and competes with proprietary models such as Claude 3 Opus and LLaMA-3.1-405B-Instruct. Ablation studies also suggest plausible reasons for why creating open instruction-tuning datasets via naive crowd-sourcing has proved difficult. Introducing low quality answers ("shirkers") in $20\%$ of Instruct-SkillMix examples causes performance to plummet, sometimes catastrophically. The Instruct-SkillMix pipeline is flexible and is adaptable to other settings.
Authors: Nada Shahin, Leila Ismail
Abstract: With the growing Deaf and Hard of Hearing population worldwide and the persistent shortage of certified sign language interpreters, there is a pressing need for an efficient, signs-driven, integrated end-to-end translation system, from sign to gloss to text and vice-versa. There has been a wealth of research on machine translations and related reviews. However, there are few works on sign language machine translation considering the particularity of the language being continuous and dynamic. This paper aims to address this void, providing a retrospective analysis of the temporal evolution of sign language machine translation algorithms and a taxonomy of the Transformers architectures, the most used approach in language translation. We also present the requirements of a real-time Quality-of-Service sign language ma-chine translation system underpinned by accurate deep learning algorithms. We propose future research directions for sign language translation systems.
Authors: Ashish Hooda, Rishabh Khandelwal, Prasad Chalasani, Kassem Fawaz, Somesh Jha
Abstract: Privacy policies are crucial in the online ecosystem, defining how services handle user data and adhere to regulations such as GDPR and CCPA. However, their complexity and frequent updates often make them difficult for stakeholders to understand and analyze. Current automated analysis methods, which utilize natural language processing, have limitations. They typically focus on individual tasks and fail to capture the full context of the policies. We propose PolicyLR, a new paradigm that offers a comprehensive machine-readable representation of privacy policies, serving as an all-in-one solution for multiple downstream tasks. PolicyLR converts privacy policies into a machine-readable format using valuations of atomic formulae, allowing for formal definitions of tasks like compliance and consistency. We have developed a compiler that transforms unstructured policy text into this format using off-the-shelf Large Language Models (LLMs). This compiler breaks down the transformation task into a two-stage translation and entailment procedure. This procedure considers the full context of the privacy policy to infer a complex formula, where each formula consists of simpler atomic formulae. The advantage of this model is that PolicyLR is interpretable by design and grounded in segments of the privacy policy. We evaluated the compiler using ToS;DR, a community-annotated privacy policy entailment dataset. Utilizing open-source LLMs, our compiler achieves precision and recall values of 0.91 and 0.88, respectively. Finally, we demonstrate the utility of PolicyLR in three privacy tasks: Policy Compliance, Inconsistency Detection, and Privacy Comparison Shopping.
Authors: Yang Liu, Chuan Zhou, Peng Zhang, Yanan Cao, Yongchao Liu, Zhao Li, Hongyang Chen
Abstract: Knowledge graph embedding (KGE) constitutes a foundational task, directed towards learning representations for entities and relations within knowledge graphs (KGs), with the objective of crafting representations comprehensive enough to approximate the logical and symbolic interconnections among entities. In this paper, we define a metric Z-counts to measure the difficulty of training each triple ($<$head entity, relation, tail entity$>$) in KGs with theoretical analysis. Based on this metric, we propose \textbf{CL4KGE}, an efficient \textbf{C}urriculum \textbf{L}earning based training strategy for \textbf{KGE}. This method includes a difficulty measurer and a training scheduler that aids in the training of KGE models. Our approach possesses the flexibility to act as a plugin within a wide range of KGE models, with the added advantage of adaptability to the majority of KGs in existence. The proposed method has been evaluated on popular KGE models, and the results demonstrate that it enhances the state-of-the-art methods. The use of Z-counts as a metric has enabled the identification of challenging triples in KGs, which helps in devising effective training strategies.
Authors: Shusaku Egami, Takahiro Ugai, Ken Fukuda
Abstract: Multi-modal knowledge graphs (MMKGs), which ground various non-symbolic data (e.g., images and videos) into symbols, have attracted attention as resources enabling knowledge processing and machine learning across modalities. However, the construction of MMKGs for videos consisting of multiple events, such as daily activities, is still in the early stages. In this paper, we construct an MMKG based on synchronized multi-view simulated videos of daily activities. Besides representing the content of daily life videos as event-centric knowledge, our MMKG also includes frame-by-frame fine-grained changes, such as bounding boxes within video frames. In addition, we provide support tools for querying our MMKG. As an application example, we demonstrate that our MMKG facilitates benchmarking vision-language models by providing the necessary vision-language datasets for a tailored task.
Authors: Vanni Zavarella, Sergio Consoli, Diego Reforgiato Recupero, Gianni Fenu, Simone Angioni, Davide Buscaldi, Danilo Dess\`i, Francesco Osborne
Abstract: Numerous methods and pipelines have recently emerged for the automatic extraction of knowledge graphs from documents such as scientific publications and patents. However, adapting these methods to incorporate alternative text sources like micro-blogging posts and news has proven challenging as they struggle to model open-domain entities and relations, typically found in these sources. In this paper, we propose an enhanced information extraction pipeline tailored to the extraction of a knowledge graph comprising open-domain entities from micro-blogging posts on social media platforms. Our pipeline leverages dependency parsing and classifies entity relations in an unsupervised manner through hierarchical clustering over word embeddings. We provide a use case on extracting semantic triples from a corpus of 100 thousand tweets about digital transformation and publicly release the generated knowledge graph. On the same dataset, we conduct two experimental evaluations, showing that the system produces triples with precision over 95% and outperforms similar pipelines of around 5% in terms of precision, while generating a comparatively higher number of triples.
Authors: Hao Jiang, Haoxiang Zhang, Qingshan Hou, Chaofeng Chen, Weisi Lin, Jingchang Zhang, Annan Wang
Abstract: Providing high-quality item recall for text queries is crucial in large-scale e-commerce search systems. Current Embedding-based Retrieval Systems (ERS) embed queries and items into a shared low-dimensional space, but uni-modality ERS rely too heavily on textual features, making them unreliable in complex contexts. While multi-modality ERS incorporate various data sources, they often overlook individual preferences for different modalities, leading to suboptimal results. To address these issues, we propose MRSE, a Multi-modality Retrieval System that integrates text, item images, and user preferences through lightweight mixture-of-expert (LMoE) modules to better align features across and within modalities. MRSE also builds user profiles at a multi-modality level and introduces a novel hybrid loss function that enhances consistency and robustness using hard negative sampling. Experiments on a large-scale dataset from Shopee and online A/B testing show that MRSE achieves an 18.9% improvement in offline relevance and a 3.7% gain in online core metrics compared to Shopee's state-of-the-art uni-modality system.
Authors: Jerome Garnier-Brun, Marc M\'ezard, Emanuele Moscato, Luca Saglietti
Abstract: We introduce a hierarchical filtering procedure for generative models of sequences on trees, enabling control over the range of positional correlations in the data. Leveraging this controlled setting, we provide evidence that vanilla encoder-only transformer architectures can implement the optimal Belief Propagation algorithm on both root classification and masked language modeling tasks. Correlations at larger distances corresponding to increasing layers of the hierarchy are sequentially included as the network is trained. We analyze how the transformer layers succeed by focusing on attention maps from models trained with varying degrees of filtering. These attention maps show clear evidence for iterative hierarchical reconstruction of correlations, and we can relate these observations to a plausible implementation of the exact inference algorithm for the network sizes considered.
Authors: Hanjia Lyu, Ryan Rossi, Xiang Chen, Md Mehrab Tanjim, Stefano Petrangeli, Somdeb Sarkhel, Jiebo Luo
Abstract: Large Language Models (LLMs) and Large Multimodal Models (LMMs) have been shown to enhance the effectiveness of enriching item descriptions, thereby improving the accuracy of recommendation systems. However, most existing approaches either rely on text-only prompting or employ basic multimodal strategies that do not fully exploit the complementary information available from both textual and visual modalities. This paper introduces a novel framework, Cross-Reflection Prompting, termed X-Reflect, designed to address these limitations by prompting LMMs to explicitly identify and reconcile supportive and conflicting information between text and images. By capturing nuanced insights from both modalities, this approach generates more comprehensive and contextually richer item representations. Extensive experiments conducted on two widely used benchmarks demonstrate that our method outperforms existing prompting baselines in downstream recommendation accuracy. Additionally, we evaluate the generalizability of our framework across different LMM backbones and the robustness of the prompting strategies, offering insights for optimization. This work underscores the importance of integrating multimodal information and presents a novel solution for improving item understanding in multimodal recommendation systems.
Authors: Longshen Ou, Jingwei Zhao, Ziyu Wang, Gus Xia, Ye Wang
Abstract: Large language models have shown significant capabilities across various domains, including symbolic music generation. However, leveraging these pre-trained models for controllable music arrangement tasks, each requiring different forms of musical information as control, remains a novel challenge. In this paper, we propose a unified sequence-to-sequence framework that enables the fine-tuning of a symbolic music language model for multiple multi-track arrangement tasks, including band arrangement, piano reduction, drum arrangement, and voice separation. Our experiments demonstrate that the proposed approach consistently achieves higher musical quality compared to task-specific baselines across all four tasks. Furthermore, through additional experiments on probing analysis, we show the pre-training phase equips the model with essential knowledge to understand musical conditions, which is hard to acquired solely through task-specific fine-tuning.
Authors: Franziska Braun, Sebastian P. Bayerl, Florian H\"onig, Hartmut Lehfeld, Thomas Hillemacher, Tobias Bocklet, Korbinian Riedhammer
Abstract: Speech pauses, alongside content and structure, offer a valuable and non-invasive biomarker for detecting dementia. This work investigates the use of pause-enriched transcripts in transformer-based language models to differentiate the cognitive states of subjects with no cognitive impairment, mild cognitive impairment, and Alzheimer's dementia based on their speech from a clinical assessment. We address three binary classification tasks: Onset, monitoring, and dementia exclusion. The performance is evaluated through experiments on a German Verbal Fluency Test and a Picture Description Test, comparing the model's effectiveness across different speech production contexts. Starting from a textual baseline, we investigate the effect of incorporation of pause information and acoustic context. We show the test should be chosen depending on the task, and similarly, lexical pause information and acoustic cross-attention contribute differently.
Authors: Nathaniel Li, Ziwen Han, Ian Steneker, Willow Primack, Riley Goodside, Hugh Zhang, Zifan Wang, Cristina Menghini, Summer Yue
Abstract: Recent large language model (LLM) defenses have greatly improved models' ability to refuse harmful queries, even when adversarially attacked. However, LLM defenses are primarily evaluated against automated adversarial attacks in a single turn of conversation, an insufficient threat model for real-world malicious use. We demonstrate that multi-turn human jailbreaks uncover significant vulnerabilities, exceeding 70% attack success rate (ASR) on HarmBench against defenses that report single-digit ASRs with automated single-turn attacks. Human jailbreaks also reveal vulnerabilities in machine unlearning defenses, successfully recovering dual-use biosecurity knowledge from unlearned models. We compile these results into Multi-Turn Human Jailbreaks (MHJ), a dataset of 2,912 prompts across 537 multi-turn jailbreaks. We publicly release MHJ alongside a compendium of jailbreak tactics developed across dozens of commercial red teaming engagements, supporting research towards stronger LLM defenses.
Authors: Takanori Ashihara, Takafumi Moriya, Kohei Matsuura, Tomohiro Tanaka, Yusuke Ijima, Taichi Asami, Marc Delcroix, Yukinori Honma
Abstract: Self-supervised learning (SSL) for speech representation has been successfully applied in various downstream tasks, such as speech and speaker recognition. More recently, speech SSL models have also been shown to be beneficial in advancing spoken language understanding tasks, implying that the SSL models have the potential to learn not only acoustic but also linguistic information. In this paper, we aim to clarify if speech SSL techniques can well capture linguistic knowledge. For this purpose, we introduce SpeechGLUE, a speech version of the General Language Understanding Evaluation (GLUE) benchmark. Since GLUE comprises a variety of natural language understanding tasks, SpeechGLUE can elucidate the degree of linguistic ability of speech SSL models. Experiments demonstrate that speech SSL models, although inferior to text-based SSL models, perform better than baselines, suggesting that they can acquire a certain amount of general linguistic knowledge from just unlabeled speech data.
Authors: Jihyeon Lee, Dain Kim, Doohae Jung, Boseop Kim, Kyoung-Woon On
Abstract: In-context learning, which offers substantial advantages over fine-tuning, is predominantly observed in decoder-only models, while encoder-decoder (i.e., seq2seq) models excel in methods that rely on weight updates. Recently, a few studies have demonstrated the feasibility of few-shot learning with seq2seq models; however, this has been limited to tasks that align well with the seq2seq architecture, such as summarization and translation. Inspired by these initial studies, we provide a first-ever extensive experiment comparing the in-context few-shot learning capabilities of decoder-only and encoder-decoder models on a broad range of tasks. Furthermore, we propose two methods to more effectively elicit in-context learning ability in seq2seq models: objective-aligned prompting and a fusion-based approach. Remarkably, our approach outperforms a decoder-only model that is six times larger and exhibits significant performance improvements compared to conventional seq2seq models across a variety of settings. We posit that, with the right configuration and prompt design, seq2seq models can be highly effective few-shot learners for a wide spectrum of applications.
Authors: Kilichbek Haydarov, Xiaoqian Shen, Avinash Madasu, Mahmoud Salem, Li-Jia Li, Gamaleldin Elsayed, Mohamed Elhoseiny
Abstract: We introduce Affective Visual Dialog, an emotion explanation and reasoning task as a testbed for research on understanding the formation of emotions in visually grounded conversations. The task involves three skills: (1) Dialog-based Question Answering (2) Dialog-based Emotion Prediction and (3) Affective emotion explanation generation based on the dialog. Our key contribution is the collection of a large-scale dataset, dubbed AffectVisDial, consisting of 50K 10-turn visually grounded dialogs as well as concluding emotion attributions and dialog-informed textual emotion explanations, resulting in a total of 27,180 working hours. We explain our design decisions in collecting the dataset and introduce the questioner and answerer tasks that are associated with the participants in the conversation. We train and demonstrate solid Affective Visual Dialog baselines adapted from state-of-the-art models. Remarkably, the responses generated by our models show promising emotional reasoning abilities in response to visually grounded conversations. Our project page is available at https://affective-visual-dialog.github.io.
Authors: Daniela Occhipinti, Serra Sinem Tekiroglu, Marco Guerini
Abstract: Providing dialogue agents with a profile representation can improve their consistency and coherence, leading to better conversations. However, current profile-based dialogue datasets for training such agents contain either explicit profile representations that are simple and dialogue-specific, or implicit representations that are difficult to collect. In this work, we propose a unified framework in which we bring together both standard and more sophisticated profile representations by creating a new resource where each dialogue is aligned with all possible speaker representations such as communication style, biographies, and personality. This framework allows to test several baselines built using generative language models with several profile configurations. The automatic evaluation shows that profile-based models have better generalisation capabilities than models trained on dialogues only, both in-domain and cross-domain settings. These results are consistent for fine-tuned models and instruction-based LLMs. Additionally, human evaluation demonstrates a clear preference for generations consistent with both profile and context. Finally, to account for possible privacy concerns, all experiments are done under two configurations: inter-character and intra-character. In the former, the LM stores the information about the character in its internal representation, while in the latter, the LM does not retain any personal information but uses it only at inference time.
Authors: Han Guo, Philip Greengard, Eric P. Xing, Yoon Kim
Abstract: We propose a simple approach for memory-efficient adaptation of pretrained language models. Our approach uses an iterative algorithm to decompose each pretrained matrix into a high-precision low-rank component and a memory-efficient quantized component. During finetuning, the quantized component remains fixed and only the low-rank component is updated. We present an integer linear programming formulation of the quantization component which enables dynamic configuration of quantization parameters (e.g., bit-width, block size) for each matrix given an overall target memory budget. We further explore a data-aware version of the algorithm which uses an approximation of the Fisher information matrix to weight the reconstruction objective during matrix decomposition. Experiments on finetuning RoBERTa and LLaMA-2 (7B and 70B) demonstrate that our low-rank plus quantized matrix decomposition approach (LQ-LoRA) outperforms strong QLoRA and GPTQ-LoRA baselines and enables aggressive quantization to sub-3 bits with only minor performance degradations. When finetuned on a language modeling calibration dataset, LQ-LoRA can also be used for model compression; in this setting our 2.75-bit LLaMA-2-70B model (which has 2.85 bits on average when including the low-rank components and requires 27GB of GPU memory) performs respectably compared to the 16-bit baseline.
Authors: Yifan Peng, Jinchuan Tian, William Chen, Siddhant Arora, Brian Yan, Yui Sudo, Muhammad Shakeel, Kwanghee Choi, Jiatong Shi, Xuankai Chang, Jee-weon Jung, Shinji Watanabe
Abstract: Recent studies have highlighted the importance of fully open foundation models. The Open Whisper-style Speech Model (OWSM) is an initial step towards reproducing OpenAI Whisper using public data and open-source toolkits. However, previous versions of OWSM (v1 to v3) are still based on standard Transformer, which might lead to inferior performance compared to state-of-the-art speech encoder architectures. This work aims to improve the performance and efficiency of OWSM without additional data. We present a series of E-Branchformer-based models named OWSM v3.1, ranging from 100M to 1B parameters. OWSM v3.1 outperforms its predecessor, OWSM v3, in most evaluation benchmarks, while showing an improved inference speed of up to 25%. We further reveal the emergent ability of OWSM v3.1 in zero-shot contextual biasing speech recognition. We also provide a model trained on a subset of data with low license restrictions. We will publicly release the code, pre-trained models, and training logs.
Authors: Heeseung Kim, Soonshin Seo, Kyeongseok Jeong, Ohsung Kwon, Soyoon Kim, Jungwhan Kim, Jaehong Lee, Eunwoo Song, Myungwoo Oh, Jung-Woo Ha, Sungroh Yoon, Kang Min Yoo
Abstract: Recent work shows promising results in expanding the capabilities of large language models (LLM) to directly understand and synthesize speech. However, an LLM-based strategy for modeling spoken dialogs remains elusive, calling for further investigation. This paper introduces an extensive speech-text LLM framework, the Unified Spoken Dialog Model (USDM), designed to generate coherent spoken responses with naturally occurring prosodic features relevant to the given input speech without relying on explicit automatic speech recognition (ASR) or text-to-speech (TTS) systems. We have verified the inclusion of prosody in speech tokens that predominantly contain semantic information and have used this foundation to construct a prosody-infused speech-text model. Additionally, we propose a generalized speech-text pretraining scheme that enhances the capture of cross-modal semantics. To construct USDM, we fine-tune our speech-text model on spoken dialog data using a multi-step spoken dialog template that stimulates the chain-of-reasoning capabilities exhibited by the underlying LLM. Automatic and human evaluations on the DailyTalk dataset demonstrate that our approach effectively generates natural-sounding spoken responses, surpassing previous and cascaded baselines. We will make our code and checkpoints publicly available.
Authors: Yifan Peng, Yui Sudo, Muhammad Shakeel, Shinji Watanabe
Abstract: There has been an increasing interest in large speech models that can perform multiple tasks in a single model. Such models usually adopt an encoder-decoder or decoder-only architecture due to their popularity and good performance in many domains. However, autoregressive models can be slower during inference compared to non-autoregressive models and also have potential risks of hallucination. Though prior studies observed promising results of non-autoregressive models for certain tasks at small scales, it remains unclear if they can be scaled to speech-to-text generation in diverse languages and tasks. Inspired by the Open Whisper-style Speech Model (OWSM) project, we propose OWSM-CTC, a novel encoder-only speech foundation model based on Connectionist Temporal Classification (CTC). It is trained on 180k hours of public audio data for multilingual automatic speech recognition (ASR), speech translation (ST), and language identification (LID). Compared to encoder-decoder OWSM, our OWSM-CTC achieves competitive results on ASR and up to 24% relative improvement on ST, while it is more robust and 3 to 4 times faster for inference. OWSM-CTC also improves the long-form ASR result with 20x speed-up. We will publicly release our code, pre-trained model, and training logs to promote open science in speech foundation models.
Authors: Jing Huang, Zhengxuan Wu, Christopher Potts, Mor Geva, Atticus Geiger
Abstract: Individual neurons participate in the representation of multiple high-level concepts. To what extent can different interpretability methods successfully disentangle these roles? To help address this question, we introduce RAVEL (Resolving Attribute-Value Entanglements in Language Models), a dataset that enables tightly controlled, quantitative comparisons between a variety of existing interpretability methods. We use the resulting conceptual framework to define the new method of Multi-task Distributed Alignment Search (MDAS), which allows us to find distributed representations satisfying multiple causal criteria. With Llama2-7B as the target language model, MDAS achieves state-of-the-art results on RAVEL, demonstrating the importance of going beyond neuron-level analyses to identify features distributed across activations. We release our benchmark at https://github.com/explanare/ravel.
Authors: Iain J. Cruickshank, Amir Soofi, Lynnette Hui Xian Ng
Abstract: Stance detection of social media text is a key component of many real-world applications like evaluating marketing campaigns, evaluating political policies or candidates, or evaluating information environments. However, creating automatic stance labeling systems requires the manual annotation of stances, which is both tedious and resource-intensive. This paper introduces a stance labeling method that makes use of weak signals of sentence tone, then consolidating these signals with a Data Programmingmodel for the final stance label. In a time of international conflict, understanding the public opinion towards the country's military is crucial for recruitment. We present DIVERSE, a dataset involve stances towards YouTube videos of the US military (Dataset available at https://doi.org/10.5281/zenodo.10493803). On average, the videos have 200 comments each, and the stances skew slightly towards the "against" characterization for both the US army and the video.
Authors: Shannon Zejiang Shen, Hunter Lang, Bailin Wang, Yoon Kim, David Sontag
Abstract: We propose a method to teach multiple large language models (LLM) to collaborate by interleaving their generations at the token level. We model the decision of which LLM generates the next token as a latent variable. By optimizing the marginal likelihood of a training set under our latent variable model, the base LLM automatically learns when to generate itself and when to call on one of the ``assistant'' language models to generate, all without direct supervision. Token-level collaboration during decoding allows for a fusion of each model's expertise in a manner tailored to the specific task at hand. Our collaborative decoding is especially useful in cross-domain settings where a generalist base LLM learns to invoke domain expert models. On instruction-following, domain-specific QA, and reasoning tasks, we show that the performance of the joint system exceeds that of the individual models. Through qualitative analysis of the learned latent decisions, we show models trained with our method exhibit several interesting collaboration patterns, e.g., template-filling. Our code is available at https://github.com/clinicalml/co-llm.
Authors: Xuechen Liang, Meiling Tao, Yinghui Xia, Tianyu Shi, Jun Wang, JingSong Yang
Abstract: Open large language models (LLMs) have significantly advanced the field of natural language processing, showcasing impressive performance across various tasks.Despite the significant advancements in LLMs, their effective operation still relies heavily on human input to accurately guide the dialogue flow, with agent tuning being a crucial optimization technique that involves human adjustments to the model for better response to such guidance.Addressing this dependency, our work introduces the TinyAgent model, trained on a meticulously curated high-quality dataset. We also present the Collaborative Multi-Agent Tuning (CMAT) framework, an innovative system designed to augment language agent capabilities through adaptive weight updates based on environmental feedback. This framework fosters collaborative learning and real-time adaptation among multiple intelligent agents, enhancing their context-awareness and long-term memory. In this research, we propose a new communication agent framework that integrates multi-agent systems with environmental feedback mechanisms, offering a scalable method to explore cooperative behaviors. Notably, our TinyAgent-7B model exhibits performance on par with GPT-3.5, despite having fewer parameters, signifying a substantial improvement in the efficiency and effectiveness of LLMs.
Authors: Haven Kim, Taketo Akama
Abstract: In musical compositions that include vocals, lyrics significantly contribute to artistic expression. Consequently, previous studies have introduced the concept of a recommendation system that suggests lyrics similar to a user's favorites or personalized preferences, aiding in the discovery of lyrics among millions of tracks. However, many of these systems do not fully consider human perceptions of lyric similarity, primarily due to limited research in this area. To bridge this gap, we conducted a comparative analysis of computational methods for modeling lyric similarity with human perception. Results indicated that computational models based on similarities between embeddings from pre-trained BERT-based models, the audio from which the lyrics are derived, and phonetic components are indicative of perceptual lyric similarity. This finding underscores the importance of semantic, stylistic, and phonetic similarities in human perception about lyric similarity. We anticipate that our findings will enhance the development of similarity-based lyric recommendation systems by offering pseudo-labels for neural network development and introducing objective evaluation metrics.
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: Sourabrata Mukherjee, Atul Kr. Ojha, Akanksha Bansal, Deepak Alok, John P. McCrae, Ond\v{r}ej Du\v{s}ek
Abstract: Text style transfer (TST) involves altering the linguistic style of a text while preserving its core content. This paper focuses on sentiment transfer, a popular TST subtask, across a spectrum of Indian languages: Hindi, Magahi, Malayalam, Marathi, Punjabi, Odia, Telugu, and Urdu, expanding upon previous work on English-Bangla sentiment transfer (Mukherjee et al., 2023). We introduce dedicated datasets of 1,000 positive and 1,000 negative style-parallel sentences for each of these eight languages. We then evaluate the performance of various benchmark models categorized into parallel, non-parallel, cross-lingual, and shared learning approaches, including the Llama2 and GPT-3.5 large language models (LLMs). Our experiments highlight the significance of parallel data in TST and demonstrate the effectiveness of the Masked Style Filling (MSF) approach (Mukherjee et al., 2023) in non-parallel techniques. Moreover, cross-lingual and joint multilingual learning methods show promise, offering insights into selecting optimal models tailored to the specific language and task requirements. To the best of our knowledge, this work represents the first comprehensive exploration of the TST task as sentiment transfer across a diverse set of languages.
Authors: Sourabrata Mukherjee, Atul Kr. Ojha, Ond\v{r}ej Du\v{s}ek
Abstract: We analyze the performance of large language models (LLMs) on Text Style Transfer (TST), specifically focusing on sentiment transfer and text detoxification across three languages: English, Hindi, and Bengali. Text Style Transfer involves modifying the linguistic style of a text while preserving its core content. We evaluate the capabilities of pre-trained LLMs using zero-shot and few-shot prompting as well as parameter-efficient finetuning on publicly available datasets. Our evaluation using automatic metrics, GPT-4 and human evaluations reveals that while some prompted LLMs perform well in English, their performance in on other languages (Hindi, Bengali) remains average. However, finetuning significantly improves results compared to zero-shot and few-shot prompting, making them comparable to previous state-of-the-art. This underscores the necessity of dedicated datasets and specialized models for effective TST.
Authors: Jiaming Tang, Yilong Zhao, Kan Zhu, Guangxuan Xiao, Baris Kasikci, Song Han
Abstract: As the demand for long-context large language models (LLMs) increases, models with context windows of up to 128K or 1M tokens are becoming increasingly prevalent. However, long-context LLM inference is challenging since the inference speed decreases significantly as the sequence length grows. This slowdown is primarily caused by loading a large KV cache during self-attention. Previous works have shown that a small portion of critical tokens will dominate the attention outcomes. However, we observe the criticality of a token highly depends on the query. To this end, we propose Quest, a query-aware KV cache selection algorithm. Quest keeps track of the minimal and maximal Key values in KV cache pages and estimates the criticality of a given page using Query vectors. By only loading the Top-K critical KV cache pages for attention, Quest significantly speeds up self-attention without sacrificing accuracy. We show that Quest can achieve up to 2.23x self-attention speedup, which reduces inference latency by 7.03x while performing well on tasks with long dependencies with negligible accuracy loss. Code is available at http://github.com/mit-han-lab/Quest .
Authors: Zipeng Liu, Likang Wu, Ming He, Zhong Guan, Hongke Zhao, Nan Feng
Abstract: Significant efforts have been dedicated to integrating the powerful Large Language Models (LLMs) with diverse modalities, particularly focusing on the fusion of language, vision and audio data. However, the graph-structured data, which is inherently rich in structural and domain-specific knowledge, has not yet been gracefully adapted to LLMs. Existing methods either describe the graph with raw text, suffering the loss of graph structural information, or feed Graph Neural Network (GNN) embeddings into LLMs at the cost of losing explainable prompt semantics. To bridge this gap, we introduce an end-to-end modality-aligning framework for LLM-graph alignment: Dual-Residual Vector Quantized-Variational AutoEncoder, namely Dr.E. Our approach is purposefully designed to facilitate token-level alignment with LLMs, enabling an effective translation of the intrinsic `language' of graphs into comprehensible natural language. We also manage to enhance LLMs' more robust structural understanding of graphs by incorporating multiple views of the central nodes based on their surrounding nodes at various distances. Our experimental evaluations on standard graph tasks demonstrate competitive performance against other state-of-the-art (SOTA) approaches. Additionally, our framework ensures certain visual interpretability, efficiency, and robustness, marking the promising successful endeavor to achieve token-level alignment between LLMs and GNNs. Our code is available at: https://anonymous.4open.science/r/dre-817.
Authors: Jay Shim, Grant Kruttschnitt, Alyssa Ma, Daniel Kim, Benjamin Chek, Athul Anand, Kevin Zhu, Sean O'Brien
Abstract: Rapidly increasing model scales coupled with steering methods such as chain-of-thought prompting have led to drastic improvements in language model reasoning. At the same time, models struggle with compositional generalization and are far from human performance on many reasoning-based benchmarks. Leveraging the success of chain-of-thought prompting, and also taking inspiration from context-aware decoding (CAD), we explore input-based contrasting methods to further encourage the type of reasoning induced by chain-of-thought prompting. While work remains to stabilize these results across datasets and models, the improvements we find warrant further investigation into input-based steering methods for context-aware reasoning.
Authors: Youngkyung Seo, Yoonseok Heo, Jun-Seok Koh, Du-Seong Chang
Abstract: In multi-session dialog system, it is essential to continuously update the memory as the session progresses. Simply accumulating memory can make it difficult to focus on the content of the conversation for inference due to the limited input sentence size. Therefore, efficient and accurate conversation model that is capable of managing memory to reflect the conversation history continuously is necessary. This paper presents a conversation model that efficiently manages memory as sessions progress and incorporates this into the model to reflect the conversation history accurately with 3 methodologies: SFT, DPO and DPO with SFT model. Our model using DPO algorithm shows an improvement about 0.0591 of BERTScore in memory accuracy, and the rate of responses reflecting the memory increased as well. Also, response generation performance enhanced about 4.292 in fluency, 3.935 in coherence, and 2.896 in consistency. This paper describes a training method that yields better performance than models with more than twice the parameter size, even when the model size is smaller. Thus, our model demonstrates efficiency not only in terms of accuracy but also in resource utilization.
Authors: Kunlun Zhu, Yifan Luo, Dingling Xu, Ruobing Wang, Shi Yu, Shuo Wang, Yukun Yan, Zhenghao Liu, Xu Han, Zhiyuan Liu, Maosong Sun
Abstract: Retrieval-Augmented Generation (RAG) systems have demonstrated their advantages in alleviating the hallucination of Large Language Models (LLMs). Existing RAG benchmarks mainly focus on evaluating whether LLMs can correctly answer the general knowledge. However, they are unable to evaluate the effectiveness of the RAG system in dealing with the data from different vertical domains. This paper introduces RAGEval, a framework for automatically generating evaluation datasets to evaluate the knowledge usage ability of different LLMs in different scenarios. Specifically, RAGEval summarizes a schema from seed documents, applies the configurations to generate diverse documents, and constructs question-answering pairs according to both articles and configurations. We propose three novel metrics, Completeness, Hallucination, and Irrelevance, to carefully evaluate the responses generated by LLMs. By benchmarking RAG models in vertical domains, RAGEval has the ability to better evaluate the knowledge usage ability of LLMs, which avoids the confusion regarding the source of knowledge in answering question in existing QA datasets--whether it comes from parameterized memory or retrieval. The code and dataset will be released.
Authors: Yiming Liang, Ge Zhang, Xingwei Qu, Tianyu Zheng, Jiawei Guo, Xinrun Du, Zhenzhu Yang, Jiaheng Liu, Chenghua Lin, Lei Ma, Wenhao Huang, Jiajun Zhang
Abstract: Large Language Models (LLMs) have achieved significant advancements, however, the common learning paradigm treats LLMs as passive information repositories, neglecting their potential for active learning and alignment. Some approaches train LLMs using their own generated synthetic data, exploring the possibility of active alignment. However, there is still a huge gap between these one-time alignment methods and the continuous automatic alignment of humans. In this paper, we introduce \textbf{I-SHEEP}, an \textbf{I}terative \textbf{S}elf-En\textbf{H}anc\textbf{E}m\textbf{E}nt \textbf{P}aradigm.This human-like paradigm enables LLMs to \textbf{continuously self-align from scratch with nothing}. Compared to the one-time alignment method Dromedary \cite{sun2023principledriven}, which refers to the first iteration in this paper, I-SHEEP can significantly enhance capacities on both Qwen and Llama models. I-SHEEP achieves a maximum relative improvement of 78.2\% in the Alpaca Eval, 24.0\% in the MT Bench, and an absolute increase of 8.88\% in the IFEval accuracy over subsequent iterations in Qwen-1.5 72B model. Additionally, I-SHEEP surpasses the base model in various standard benchmark generation tasks, achieving an average improvement of 24.77\% in code generation tasks, 12.04\% in TrivialQA, and 20.29\% in SQuAD. We also provide new insights based on the experiment results. Our codes, datasets, and models are available at \textbf{https://anonymous.4open.science/r/I-SHEEP}.
Authors: Dingzirui Wang, Longxu Dou, Xuanliang Zhang, Qingfu Zhu, Wanxiang Che
Abstract: Text-to-SQL is an important task that helps people obtain information from databases by automatically generating SQL queries. Considering the brilliant performance, approaches based on Large Language Models (LLMs) become the mainstream for text-to-SQL. Among these approaches, automated correction is an effective approach that further enhances performance by correcting the mistakes in the generated results. The existing correction methods require LLMs to directly correct with generated SQL, while previous research shows that LLMs do not know how to detect mistakes, leading to poor performance. Therefore, in this paper, we propose to employ the decomposed correction to enhance text-to-SQL performance. We first demonstrate that decomposed correction outperforms direct correction since detecting and fixing mistakes with the results of the decomposed sub-tasks is easier than with SQL. Based on this analysis, we introduce Decomposed Automation Correction (DAC), which corrects SQL by decomposing text-to-SQL into entity linking and skeleton parsing. DAC first generates the entity and skeleton corresponding to the question and then compares the differences between the initial SQL and the generated entities and skeleton as feedback for correction. Experimental results show that our method improves performance by $3.7\%$ on average of Spider, Bird, and KaggleDBQA compared with the baseline method, demonstrating the effectiveness of DAC.
Authors: Xuanliang Zhang, Dingzirui Wang, Longxu Dou, Baoxin Wang, Dayong Wu, Qingfu Zhu, Wanxiang Che
Abstract: The table reasoning task aims to answer the question according to the given table. Currently, using Large Language Models (LLMs) is the predominant method for table reasoning. Most existing methods employ a fixed tabular format to represent the table, which could limit the performance. Given that each instance requires different capabilities and models possess varying abilities, we assert that different instances and models suit different tabular formats. We prove the aforementioned claim through quantitative analysis of experimental results, where different instances and models achieve different performances using various tabular formats. Building on this discussion, we propose FLEXTAF-Single and FLEXTAF-Vote to enhance table reasoning performance by employing flexible tabular formats. Specifically, (i) FLEXTAF-Single trains a classifier to predict the most suitable tabular format based on the instance and the LLM. (ii) FLEXTAF-Vote integrates the results across different formats. Our experiments on WikiTableQuestions and TabFact reveal significant improvements, with average gains of 2.3% and 4.8% compared to the best performance achieved using a fixed tabular format with greedy decoding and self-consistency decoding, thereby validating the effectiveness of our methods.
Authors: Yeyong Yu, Runsheng Yu, Haojie Wei, Zhanqiu Zhang, Quan Qian
Abstract: The rapid advancement of large language models (LLMs) has revolutionized role-playing, enabling the development of general role-playing models. However, current role-playing training has two significant issues: (I) Using a predefined role profile to prompt dialogue training for specific scenarios usually leads to inconsistencies and even conflicts between the dialogue and the profile, resulting in training biases. (II) The model learns to imitate the role based solely on the profile, neglecting profile-dialogue alignment at the sentence level. In this work, we propose a simple yet effective framework called BEYOND DIALOGUE, designed to overcome these hurdles. This framework innovatively introduces "beyond dialogue" tasks to align dialogue with profile traits based on each specific scenario, thereby eliminating biases during training. Furthermore, by adopting an innovative prompting mechanism that generates reasoning outcomes for training, the framework allows the model to achieve fine-grained alignment between profile and dialogue at the sentence level. The aforementioned methods are fully automated and low-cost. Additionally, the integration of automated dialogue and objective evaluation methods forms a comprehensive framework, paving the way for general role-playing. Experimental results demonstrate that our model excels in adhering to and reflecting various dimensions of role profiles, outperforming most proprietary general and specialized role-playing baselines. All code and datasets are available at https://github.com/yuyouyu32/BeyondDialogue.
Authors: Atmika Gorti, Manas Gaur, Aman Chadha
Abstract: Large Language Models (LLMs) are prone to inheriting and amplifying societal biases embedded within their training data, potentially reinforcing harmful stereotypes related to gender, occupation, and other sensitive categories. This issue becomes particularly problematic as biased LLMs can have far-reaching consequences, leading to unfair practices and exacerbating social inequalities across various domains, such as recruitment, online content moderation, or even the criminal justice system. Although prior research has focused on detecting bias in LLMs using specialized datasets designed to highlight intrinsic biases, there has been a notable lack of investigation into how these findings correlate with authoritative datasets, such as those from the U.S. National Bureau of Labor Statistics (NBLS). To address this gap, we conduct empirical research that evaluates LLMs in a ``bias-out-of-the-box" setting, analyzing how the generated outputs compare with the distributions found in NBLS data. Furthermore, we propose a straightforward yet effective debiasing mechanism that directly incorporates NBLS instances to mitigate bias within LLMs. Our study spans seven different LLMs, including instructable, base, and mixture-of-expert models, and reveals significant levels of bias that are often overlooked by existing bias detection techniques. Importantly, our debiasing method, which does not rely on external datasets, demonstrates a substantial reduction in bias scores, highlighting the efficacy of our approach in creating fairer and more reliable LLMs.
Authors: Hourui Deng, Hongjie Zhang, Jie Ou, Chaosheng Feng
Abstract: Spatial reasoning in Large Language Models (LLMs) is the foundation for embodied intelligence. However, even in simple maze environments, LLMs still encounter challenges in long-term path-planning, primarily influenced by their spatial hallucination and context inconsistency hallucination by long-term reasoning. To address this challenge, this study proposes an innovative model, Spatial-to-Relational Transformation and Curriculum Q-Learning (S2RCQL). To address the spatial hallucination of LLMs, we propose the Spatial-to-Relational approach, which transforms spatial prompts into entity relations and paths representing entity relation chains. This approach fully taps the potential of LLMs in terms of sequential thinking. As a result, we design a path-planning algorithm based on Q-learning to mitigate the context inconsistency hallucination, which enhances the reasoning ability of LLMs. Using the Q-value of state-action as auxiliary information for prompts, we correct the hallucinations of LLMs, thereby guiding LLMs to learn the optimal path. Finally, we propose a reverse curriculum learning technique based on LLMs to further mitigate the context inconsistency hallucination. LLMs can rapidly accumulate successful experiences by reducing task difficulty and leveraging them to tackle more complex tasks. We performed comprehensive experiments based on Baidu's self-developed LLM: ERNIE-Bot 4.0. The results showed that our S2RCQL achieved a 23%--40% improvement in both success and optimality rates compared with advanced prompt engineering.
Authors: Elysia Shi, Adithri Manda, London Chowdhury, Runeema Arun, Kevin Zhu, Michael Lam
Abstract: When using AI to detect signs of depressive disorder, AI models habitually draw preemptive conclusions. We theorize that using chain-of-thought (CoT) prompting to evaluate Patient Health Questionnaire-8 (PHQ-8) scores will improve the accuracy of the scores determined by AI models. In our findings, when the models reasoned with CoT, the estimated PHQ-8 scores were consistently closer on average to the accepted true scores reported by each participant compared to when not using CoT. Our goal is to expand upon AI models' understanding of the intricacies of human conversation, allowing them to more effectively assess a patient's feelings and tone, therefore being able to more accurately discern mental disorder symptoms; ultimately, we hope to augment AI models' abilities, so that they can be widely accessible and used in the medical field.
Authors: Aradhye Agarwal, Suhas K Ramesh, Ayan Sengupta, Tanmoy Chakraborty
Abstract: Fine-tuning large language models (LLMs) on downstream tasks requires substantial computational resources. A class of parameter-efficient fine-tuning (PEFT) aims to mitigate these computational challenges by selectively fine-tuning only a small fraction of the model parameters. Although computationally efficient, these techniques often fail to match the performance of fully fine-tuned models, primarily due to inherent biases introduced during parameter selection. Traditional selective PEFT techniques use a fixed set of parameters based on a predefined budget (a process also known as unmasking), failing to capture parameter importance dynamically and often ending up exceeding the budget. We introduce $\text{ID}^3$, a novel selective PEFT method that calculates parameter importance continually and dynamically unmasks parameters by balancing exploration and exploitation in parameter selection. Our empirical study on 15 tasks spanning natural language understanding and generative tasks demonstrates the effectiveness of our method compared to fixed-masking-based PEFT techniques. We analytically show that $\text{ID}^3$ reduces the number of gradient updates by a factor of two, enhancing computational efficiency. $\text{ID}^3$ is robust to random initialization of neurons and, therefore, can be seamlessly integrated into existing additive and reparametrization-based PEFT modules such as adapters and LoRA for dynamic sparsification.
Authors: Unggi Lee, Minji Jeon, Yunseo Lee, Gyuri Byun, Yoorim Son, Jaeyoon Shin, Hongkyu Ko, Hyeoncheol Kim
Abstract: Art appreciation is vital in nurturing critical thinking and emotional intelligence among learners. However, traditional art appreciation education has often been hindered by limited access to art resources, especially for disadvantaged students, and an imbalanced emphasis on STEM subjects in mainstream education. In response to these challenges, recent technological advancements have paved the way for innovative solutions. This study explores the application of multi-modal large language models (MLLMs) in art appreciation education, focusing on developing LLaVA-Docent, a model that leverages these advancements. Our approach involved a comprehensive literature review and consultations with experts in the field, leading to developing a robust data framework. Utilizing this framework, we generated a virtual dialogue dataset that was leveraged by GPT-4. This dataset was instrumental in training the MLLM, named LLaVA-Docent. Six researchers conducted quantitative and qualitative evaluations of LLaVA-Docent to assess its effectiveness, benchmarking it against the GPT-4 model in a few-shot setting. The evaluation process revealed distinct strengths and weaknesses of the LLaVA-Docent model. Our findings highlight the efficacy of LLaVA-Docent in enhancing the accessibility and engagement of art appreciation education. By harnessing the potential of MLLMs, this study makes a significant contribution to the field of art education, proposing a novel methodology that reimagines the way art appreciation is taught and experienced.
Authors: Alexandra Souly, Qingyuan Lu, Dillon Bowen, Tu Trinh, Elvis Hsieh, Sana Pandey, Pieter Abbeel, Justin Svegliato, Scott Emmons, Olivia Watkins, Sam Toyer
Abstract: Most jailbreak papers claim the jailbreaks they propose are highly effective, often boasting near-100% attack success rates. However, it is perhaps more common than not for jailbreak developers to substantially exaggerate the effectiveness of their jailbreaks. We suggest this problem arises because jailbreak researchers lack a standard, high-quality benchmark for evaluating jailbreak performance, leaving researchers to create their own. To create a benchmark, researchers must choose a dataset of forbidden prompts to which a victim model will respond, along with an evaluation method that scores the harmfulness of the victim model's responses. We show that existing benchmarks suffer from significant shortcomings and introduce the StrongREJECT benchmark to address these issues. StrongREJECT's dataset contains prompts that victim models must answer with specific, harmful information, while its automated evaluator measures the extent to which a response gives useful information to forbidden prompts. In doing so, the StrongREJECT evaluator achieves state-of-the-art agreement with human judgments of jailbreak effectiveness. Notably, we find that existing evaluation methods significantly overstate jailbreak effectiveness compared to human judgments and the StrongREJECT evaluator. We describe a surprising and novel phenomenon that explains this discrepancy: jailbreaks bypassing a victim model's safety fine-tuning tend to reduce its capabilities. Together, our findings underscore the need for researchers to use a high-quality benchmark, such as StrongREJECT, when developing new jailbreak attacks. We release the StrongREJECT code and data at https://strong-reject.readthedocs.io/en/latest/.
Authors: Aniruddha Nrusimha, Mayank Mishra, Naigang Wang, Dan Alistarh, Rameswar Panda, Yoon Kim
Abstract: We consider the problem of accurate quantization for language models, where both the weights and activations are uniformly quantized to 4 bits per parameter, the lowest bitwidth format natively supported by GPU hardware. In this context, the key challenge is activation quantization: it is known that language models contain outlier channels whose values on average are orders of magnitude higher than than other channels, which prevents accurate low-bitwidth quantization with known techniques. We systematically study this phenomena and find that these outlier channels emerge early in training, and that they occur more frequently in layers with residual streams. We then propose a simple strategy which regularizes a layer's inputs via quantization-aware training (QAT) and its outputs via activation kurtosis regularization. We show that regularizing both the inputs and outputs is crucial for preventing a model's "migrating" the difficulty in input quantization to the weights, which makes post-training quantization (PTQ) of weights more difficult. When combined with weight PTQ, we show that our approach can obtain a W4A4 model that performs competitively to the standard-precision W16A16 baseline.
Authors: Songlin Yang, Bailin Wang, Yu Zhang, Yikang Shen, Yoon Kim
Abstract: Transformers with linear attention (i.e., linear transformers) and state-space models have recently been suggested as a viable linear-time alternative to transformers with softmax attention. However, these models still underperform transformers especially on tasks that require in-context retrieval. While more expressive variants of linear transformers which replace the additive outer-product update in linear transformers with the delta rule have been found to be more effective at associative recall, existing algorithms for training such models do not parallelize over sequence length and are thus inefficient to train on modern hardware. This work describes a hardware-efficient algorithm for training linear transformers with the delta rule, which exploits a memory-efficient representation for computing products of Householder matrices. This algorithm allows us to scale up DeltaNet to standard language modeling settings. We train a 1.3B model for 100B tokens and find that it outperforms recent linear-time baselines such as Mamba and GLA in terms of perplexity and zero-shot performance on downstream tasks (including on tasks that focus on recall). We also experiment with two hybrid models which combine DeltaNet layers with (1) sliding-window attention layers every other layer or (2) two global attention layers, and find that these hybrid models outperform strong transformer baselines.
Authors: Yueqing Liang, Liangwei Yang, Chen Wang, Xiongxiao Xu, Philip S. Yu, Kai Shu
Abstract: With the emergence of large language models (LLMs) and their ability to perform a variety of tasks, their application in recommender systems (RecSys) has shown promise. However, we are facing significant challenges when deploying LLMs into RecSys, such as limited prompt length, unstructured item information, and un-constrained generation of recommendations, leading to sub-optimal performance. To address these issues, we propose a novel method using a taxonomy dictionary. This method provides a systematic framework for categorizing and organizing items, improving the clarity and structure of item information. By incorporating the taxonomy dictionary into LLM prompts, we achieve efficient token utilization and controlled feature generation, leading to more accurate and contextually relevant recommendations. Our Taxonomy-guided Recommendation (TaxRec) approach features a two-step process: one-time taxonomy categorization and LLM-based recommendation, enabling zero-shot recommendations without the need for domain-specific fine-tuning. Experimental results demonstrate TaxRec significantly enhances recommendation quality compared to traditional zero-shot approaches, showcasing its efficacy as personal recommender with LLMs. Code is available at https://github.com/yueqingliang1/TaxRec.
Authors: Han Guo, William Brandon, Radostin Cholakov, Jonathan Ragan-Kelley, Eric P. Xing, Yoon Kim
Abstract: The deployment of large language models (LLMs) is often constrained by memory bandwidth, where the primary bottleneck is the cost of transferring model parameters from the GPU's global memory to its registers. When coupled with custom kernels that fuse the dequantization and matmul operations, weight-only quantization can thus enable faster inference by reducing the amount of memory movement. However, developing high-performance kernels for weight-quantized LLMs presents substantial challenges, especially when the weights are compressed to non-evenly-divisible bit widths (e.g., 3 bits) with non-uniform, lookup table (LUT) quantization. This paper describes FLUTE, a flexible lookup table engine for LUT-quantized LLMs, which uses offline restructuring of the quantized weight matrix to minimize bit manipulations associated with unpacking, and vectorization and duplication of the lookup table to mitigate shared memory bandwidth constraints. At batch sizes < 32 and quantization group size of 128 (typical in LLM inference), the FLUTE kernel can be 2-4x faster than existing GEMM kernels. As an application of FLUTE, we explore a simple extension to lookup table-based NormalFloat quantization and apply it to quantize LLaMA3 to various configurations, obtaining competitive quantization performance against strong baselines while obtaining an end-to-end throughput increase of 1.5 to 2 times.
Authors: Ho Yin Ng, Zeyu He, Ting-Hao 'Kenneth' Huang
Abstract: Color coding, a technique assigning specific colors to cluster information types, has proven advantages in aiding human cognitive activities, especially reading and comprehension. The rise of Large Language Models (LLMs) has streamlined document coding, enabling simple automatic text labeling with various schemes. This has the potential to make color-coding more accessible and benefit more users. However, the impact of color choice on information seeking is understudied. We conducted a user study assessing various color schemes' effectiveness in LLM-coded text documents, standardizing contrast ratios to approximately 5.55:1 across schemes. Participants performed timed information-seeking tasks in color-coded scholarly abstracts. Results showed non-analogous and yellow-inclusive color schemes improved performance, with the latter also being more preferred by participants. These findings can inform better color scheme choices for text annotation. As LLMs advance document coding, we advocate for more research focusing on the "color" aspect of color-coding techniques.
Authors: Seungjun Han, Wongyung Choi
Abstract: Emergency department (ED) overcrowding and the complexity of rapid decision-making in critical care settings pose significant challenges to healthcare systems worldwide. While clinical decision support systems (CDSS) have shown promise, the integration of large language models (LLMs) offers new possibilities for enhancing triage accuracy and clinical decision-making. This study presents an LLM-driven CDSS designed to assist ED physicians and nurses in patient triage, treatment planning, and overall emergency care management. We developed a multi-agent CDSS utilizing Llama-3-70b as the base LLM, orchestrated by CrewAI and Langchain. The system comprises four AI agents emulating key ED roles: Triage Nurse, Emergency Physician, Pharmacist, and ED Coordinator. It incorporates the Korean Triage and Acuity Scale (KTAS) for triage assessment and integrates with the RxNorm API for medication management. The model was evaluated using the Asclepius dataset, with performance assessed by a clinical emergency medicine specialist. The CDSS demonstrated high accuracy in triage decision-making compared to the baseline of a single-agent system. Furthermore, the system exhibited strong performance in critical areas, including primary diagnosis, critical findings identification, disposition decision-making, treatment planning, and resource allocation. Our multi-agent CDSS demonstrates significant potential for supporting comprehensive emergency care management. By leveraging state-of-the-art AI technologies, this system offers a scalable and adaptable tool that could enhance emergency medical care delivery, potentially alleviating ED overcrowding and improving patient outcomes. This work contributes to the growing field of AI applications in emergency medicine and offers a promising direction for future research and clinical implementation.
Authors: Yinghao Ma, Anders {\O}land, Anton Ragni, Bleiz MacSen Del Sette, Charalampos Saitis, Chris Donahue, Chenghua Lin, Christos Plachouras, Emmanouil Benetos, Elio Quinton, Elona Shatri, Fabio Morreale, Ge Zhang, Gy\"orgy Fazekas, Gus Xia, Huan Zhang, Ilaria Manco, Jiawen Huang, Julien Guinot, Liwei Lin, Luca Marinelli, Max W. Y. Lam, Megha Sharma, Qiuqiang Kong, Roger B. Dannenberg, Ruibin Yuan, Shangda Wu, Shih-Lun Wu, Shuqi Dai, Shun Lei, Shiyin Kang, Simon Dixon, Wenhu Chen, Wenhao Huang, Xingjian Du, Xingwei Qu, Xu Tan, Yizhi Li, Zeyue Tian, Zhiyong Wu, Zhizheng Wu, Ziyang Ma, Ziyu Wang
Abstract: In recent years, foundation models (FMs) such as large language models (LLMs) and latent diffusion models (LDMs) have profoundly impacted diverse sectors, including music. This comprehensive review examines state-of-the-art (SOTA) pre-trained models and foundation models in music, spanning from representation learning, generative learning and multimodal learning. We first contextualise the significance of music in various industries and trace the evolution of AI in music. By delineating the modalities targeted by foundation models, we discover many of the music representations are underexplored in FM development. Then, emphasis is placed on the lack of versatility of previous methods on diverse music applications, along with the potential of FMs in music understanding, generation and medical application. By comprehensively exploring the details of the model pre-training paradigm, architectural choices, tokenisation, finetuning methodologies and controllability, we emphasise the important topics that should have been well explored, like instruction tuning and in-context learning, scaling law and emergent ability, as well as long-sequence modelling etc. A dedicated section presents insights into music agents, accompanied by a thorough analysis of datasets and evaluations essential for pre-training and downstream tasks. Finally, by underscoring the vital importance of ethical considerations, we advocate that following research on FM for music should focus more on such issues as interpretability, transparency, human responsibility, and copyright issues. The paper offers insights into future challenges and trends on FMs for music, aiming to shape the trajectory of human-AI collaboration in the music realm.