llm - 2025_03
Navigation
Papers
Recently, Large Language Models (LLMs) have witnessed remarkable performance as zero-shot task planners for robotic manipulation tasks. However, the open-loop nature of previous works makes LLM-based planning error-prone and fragile. On the other hand, failure detection approaches for closed-loop planning are often limited by task-specific heuristics or following an unrealistic assumption that the prediction is trustworthy all the time. As a general-purpose reasoning machine, LLMs or Multimodal Large Language Models (MLLMs) are promising for detecting failures. However, However, the appropriateness of the aforementioned assumption diminishes due to the notorious hullucination problem. In this work, we attempt to mitigate these issues by introducing a framework for closed-loop LLM-based planning called KnowLoop, backed by an uncertainty-based MLLMs failure detector, which is agnostic to any used MLLMs or LLMs. Specifically, we evaluate three different ways for quantifying the uncertainty of MLLMs, namely token probability, entropy, and self-explained confidence as primary metrics based on three carefully designed representative prompting strategies. With a self-collected dataset including various manipulation tasks and an LLM-based robot system, our experiments demonstrate that token probability and entropy are more reflective compared to self-explained confidence. By setting an appropriate threshold to filter out uncertain predictions and seek human help actively, the accuracy of failure detection can be significantly enhanced. This improvement boosts the effectiveness of closed-loop planning and the overall success rate of tasks.
Large Language Models (LLMs) generate text by sampling the next token from a probability distribution over the vocabulary at each decoding step. However, popular sampling methods like top-p (nucleus sampling) often struggle to balance quality and diversity, especially at higher temperatures, leading to incoherent or repetitive outputs. To address this challenge, we propose min-p sampling, a dynamic truncation method that adjusts the sampling threshold based on the model's confidence by scaling according to the top token's probability. We conduct extensive experiments on benchmarks including GPQA, GSM8K, and AlpacaEval Creative Writing, demonstrating that min-p sampling improves both the quality and diversity of generated text, particularly at high temperatures. Moreover, human evaluations reveal a clear preference for min-p sampling in terms of both text quality and diversity. Min-p sampling has been adopted by multiple open-source LLM implementations, highlighting its practical utility and potential impact.
In multi-turn dialogues, large language models (LLM) face a critical challenge of ensuring coherence while adapting to user-specific information. This study introduces the persona knowledge gap, the discrepancy between a model's internal understanding and the knowledge required for coherent, personalized conversations. While prior research has recognized these gaps, computational methods for their identification and resolution remain underexplored. We propose Conversation Preference Elicitation and Recommendation (CPER), a novel framework that dynamically detects and resolves persona knowledge gaps using intrinsic uncertainty quantification and feedback-driven refinement. CPER consists of three key modules: a Contextual Understanding Module for preference extraction, a Dynamic Feedback Module for measuring uncertainty and refining persona alignment, and a Persona-Driven Response Generation module for adapting responses based on accumulated user context. We evaluate CPER on two real-world datasets: CCPE-M for preferential movie recommendations and ESConv for mental health support. Using A/B testing, human evaluators preferred CPER's responses 42% more often than baseline models in CCPE-M and 27% more often in ESConv. A qualitative human evaluation confirms that CPER's responses are preferred for maintaining contextual relevance and coherence, particularly in longer (12+ turn) conversations.
In many safety-critical engineering domains, hazard analysis techniques are an essential part of requirement elicitation. Of the methods proposed for this task, STPA (System-Theoretic Process Analysis) represents a relatively recent development in the field. The completion, management, and traceability of this hazard analysis technique present a time-consuming challenge to the requirements and safety engineers involved. In this paper, we introduce a free, open-source software framework to build STPA models with several automated workflows powered by large language models (LLMs). In past works, LLMs have been successfully integrated into a myriad of workflows across various fields. Here, we demonstrate that LLMs can be used to complete tasks associated with STPA with a high degree of accuracy, saving the time and effort of the human engineers involved. We experimentally validate our method on real-world STPA models built by requirement engineers and researchers. The source code of our software framework is available at the following link: https://github.com/blueskysolarracing/stpa.
X-ray images are vital in medical diagnostics, but their effectiveness is limited without clinical context. Radiologists often find chest X-rays insufficient for diagnosing underlying diseases, necessitating the integration of structured clinical features with radiology reports. To address this, we introduce DALL-M, a novel framework that enhances clinical datasets by generating contextual synthetic data. DALL-M augments structured patient data, including vital signs (e.g., heart rate, oxygen saturation), radiology findings (e.g., lesion presence), and demographic factors. It integrates this tabular data with contextual knowledge extracted from radiology reports and domain-specific resources (e.g., Radiopaedia, Wikipedia), ensuring clinical consistency and reliability. DALL-M follows a three-phase process: (i) clinical context storage, (ii) expert query generation, and (iii) context-aware feature augmentation. Using large language models (LLMs), it generates both contextual synthetic values for existing clinical features and entirely new, clinically relevant features. Applied to 799 cases from the MIMIC-IV dataset, DALL-M expanded the original 9 clinical features to 91. Empirical validation with machine learning models (including Decision Trees, Random Forests, XGBoost, and TabNET) demonstrated a 16.5% improvement in F1 score and a 25% increase in Precision and Recall. DALL-M bridges an important gap in clinical data augmentation by preserving data integrity while enhancing predictive modeling in healthcare. Our results show that integrating LLM-generated synthetic features significantly improves model performance, making DALL-M a scalable and practical approach for AI-driven medical diagnostics.
In this paper, we propose a new decoding method called Permute-and-Flip (PF) decoder. It enjoys stability properties similar to the standard sampling decoder, but is provably up to 2x better in its quality-stability tradeoff than sampling and never worse than any other decoder. We also design a cryptographic watermarking scheme analogous to Aaronson (2023)'s Gumbel watermark, but naturally tailored for PF decoder. The watermarking scheme does not change the distribution to sample, while allowing arbitrarily low false positive rate and high recall whenever the generated text has high entropy. Our experiments show that the PF decoder (and its watermarked counterpart) significantly outperform(s) naive sampling (and its Gumbel watermarked counterpart) in terms of perplexity, while retaining the same stability (and detectability), hence making it a promising new approach for LLM decoding. The code is available at https://github.com/XuandongZhao/pf-decoding
Large Language Models (LLMs) are widely adopted for automated code generation with promising results. Although prior research has assessed LLM-generated code and identified various quality issues -- such as redundancy, poor maintainability, and sub-optimal performance a systematic understanding and categorization of these inefficiencies remain unexplored. Without such knowledge, practitioners struggle to optimize LLM-generated code for real-world applications, limiting its adoption. This study can also guide improving code LLMs, enhancing the quality and efficiency of code generation. Therefore, in this study, we empirically investigate inefficiencies in LLM-generated code by state-of-the-art models, i.e., CodeLlama, DeepSeek-Coder, and CodeGemma. To do so, we analyze 492 generated code snippets in the HumanEval++ dataset. We then construct a taxonomy of inefficiencies in LLM-generated code that includes 5 categories General Logic, Performance, Readability, Maintainability, and Errors) and 19 subcategories of inefficiencies. We then validate the proposed taxonomy through an online survey with 58 LLM practitioners and researchers. Our study indicates that logic and performance-related inefficiencies are the most popular, relevant, and frequently co-occur and impact overall code quality inefficiency. Our taxonomy provides a structured basis for evaluating the quality LLM-generated code and guiding future research to improve code generation efficiency.
Advancements in Large Language Models (LLMs) have increased the performance of different natural language understanding as well as generation tasks. Although LLMs have breached the state-of-the-art performance in various tasks, they often reflect different forms of bias present in the training data. In the light of this perceived limitation, we provide a unified evaluation of benchmarks using a set of representative LLMs that cover different forms of biases starting from physical characteristics to socio-economic categories. Moreover, we propose five prompting approaches to carry out the bias detection task across different aspects of bias. Further, we formulate three research questions to gain valuable insight in detecting biases in LLMs using different approaches and evaluation metrics across benchmarks. The results indicate that each of the selected LLMs suffer from one or the other form of bias with the LLaMA3.1-8B model being the least biased. Finally, we conclude the paper with the identification of key challenges and possible future directions.
As users increasingly seek guidance from LLMs for decision-making in daily life, many of these decisions are not clear-cut and depend significantly on the personal values and ethical standards of people. We present DailyDilemmas, a dataset of 1,360 moral dilemmas encountered in everyday life. Each dilemma presents two possible actions, along with affected parties and relevant human values for each action. Based on these dilemmas, we gather a repository of human values covering diverse everyday topics, such as interpersonal relationships, workplace, and environmental issues. With DailyDilemmas, we evaluate LLMs on these dilemmas to determine what action they will choose and the values represented by these action choices. Then, we analyze values through the lens of five theoretical frameworks inspired by sociology, psychology, and philosophy, including the World Values Survey, Moral Foundations Theory, Maslow's Hierarchy of Needs, Aristotle's Virtues, and Plutchik's Wheel of Emotions. For instance, we find LLMs are most aligned with self-expression over survival in World Values Survey and care over loyalty in Moral Foundations Theory. Interestingly, we find substantial preference differences in models for some core values. For example, for truthfulness, Mixtral-8x7B neglects it by 9.7% while GPT-4-turbo selects it by 9.4%. We also study the recent guidance released by OpenAI (ModelSpec), and Anthropic (Constitutional AI) to understand how their designated principles reflect their models' actual value prioritization when facing nuanced moral reasoning in daily-life settings. Finally, we find that end users cannot effectively steer such prioritization using system prompts.
Recent LLM-based agent frameworks have demonstrated impressive capabilities in task delegation and workflow orchestration, but face significant challenges in maintaining context awareness and ensuring planning consistency. This paper presents SagaLLM, a structured multi-agent framework that addresses four fundamental limitations in current LLM approaches: inadequate self-validation, context narrowing, lacking transaction properties, and insufficient inter-agent coordination. By implementing specialized context management agents and validation protocols, SagaLLM preserves critical constraints and state information throughout complex planning processes, enabling robust and consistent decision-making even during disruptions. We evaluate our approach using selected problems from the REALM benchmark, focusing on sequential and reactive planning scenarios that challenge both context retention and adaptive reasoning. Our experiments with state-of-the-art LLMs, Claude 3.7, DeepSeek R1, GPT-4o, and GPT-o1, demonstrate that while these models exhibit impressive reasoning capabilities, they struggle with maintaining global constraint awareness during complex planning tasks, particularly when adapting to unexpected changes. In contrast, the distributed cognitive architecture of SagaLLM shows significant improvements in planning consistency, constraint enforcement, and adaptation to disruptions in various scenarios.
The systems and software powered by Large Language Models (LLMs) and Multi-Modal LLMs (MLLMs) have played a critical role in numerous scenarios. However, current LLM systems are vulnerable to prompt-based attacks, with jailbreaking attacks enabling the LLM system to generate harmful content, while hijacking attacks manipulate the LLM system to perform attacker-desired tasks, underscoring the necessity for detection tools. Unfortunately, existing detecting approaches are usually tailored to specific attacks, resulting in poor generalization in detecting various attacks across different modalities. To address it, we propose JailGuard, a universal detection framework deployed on top of LLM systems for prompt-based attacks across text and image modalities. JailGuard operates on the principle that attacks are inherently less robust than benign ones. Specifically, JailGuard mutates untrusted inputs to generate variants and leverages the discrepancy of the variants' responses on the target model to distinguish attack samples from benign samples. We implement 18 mutators for text and image inputs and design a mutator combination policy to further improve detection generalization. The evaluation on the dataset containing 15 known attack types suggests that JailGuard achieves the best detection accuracy of 86.14%/82.90% on text and image inputs, outperforming state-of-the-art methods by 11.81%-25.73% and 12.20%-21.40%.
As open-source large language models (LLMs) like Llama3 become more capable, it is crucial to develop watermarking techniques to detect their potential misuse. Existing watermarking methods either add watermarks during LLM inference, which is unsuitable for open-source LLMs, or primarily target classification LLMs rather than recent generative LLMs. Adapting these watermarks to open-source LLMs for misuse detection remains an open challenge. This work defines two misuse scenarios for open-source LLMs: intellectual property (IP) violation and LLM Usage Violation. Then, we explore the application of inference-time watermark distillation and backdoor watermarking in these contexts. We propose comprehensive evaluation methods to assess the impact of various real-world further fine-tuning scenarios on watermarks and the effect of these watermarks on LLM performance. Our experiments reveal that backdoor watermarking could effectively detect IP Violation, while inference-time watermark distillation is applicable in both scenarios but less robust to further fine-tuning and has a more significant impact on LLM performance compared to backdoor watermarking. Exploring more advanced watermarking methods for open-source LLMs to detect their misuse should be an important future direction.
Quantization is a critical step to enable efficient LLM serving under limited resource. However, previous research observes that certain weights in the LLM, known as outliers, are significantly sensitive to quantization noises. Existing quantization methods leave these outliers as floating points or higher precisions to retain performance, posting challenges on the efficient hardware deployment of the mixed-precision model. This work investigates an alternative way to tame the sensitive weights' impact on the quantization error, by reducing the loss Hessian trace with respect to outliers through an efficient fine-tuning process. We propose Noise Perturbation Fine-tuning (NPFT), which identifies outlier weights and add random weight perturbations on the outliers as the model going through a PEFT optimization. NPFT tames the sensitivity of outlier weights so that the quantized model performance can be improved without special treatment to the outliers. When applied to OPT and LLaMA models, our NPFT method achieves stable performance improvements for both uniform and non-uniform quantizers, while also offering better inference efficiency. Notably, the simplest RTN can achieve performance on par with GPTQ using our NPFT on LLaMA2-7B-4bits benchmark.
This article explores the gaps that can manifest when using a large language model (LLM) to obtain simplified interpretations of data practices from a complex privacy policy. We exemplify these gaps to showcase issues in accuracy, completeness, clarity and representation, while advocating for continued research to realize an LLM's true potential in revolutionizing privacy management through personal assistants and automated compliance checking.
Fully Homomorphic Encryption over the torus (TFHE) enables computation on encrypted data without decryption, making it a cornerstone of secure and confidential computing. Despite its potential in privacy preserving machine learning, secure multi party computation, private blockchain transactions, and secure medical diagnostics, its adoption remains limited due to cryptographic complexity and usability challenges. While various TFHE libraries and compilers exist, practical code generation remains a hurdle. We propose a compiler integrated framework to evaluate LLM inference and agentic optimization for TFHE code generation, focusing on logic gates and ReLU activation. Our methodology assesses error rates, compilability, and structural similarity across open and closedsource LLMs. Results highlight significant limitations in off-the-shelf models, while agentic optimizations such as retrieval augmented generation (RAG) and few-shot prompting reduce errors and enhance code fidelity. This work establishes the first benchmark for TFHE code generation, demonstrating how LLMs, when augmented with domain-specific feedback, can bridge the expertise gap in FHE code generation.
Large Language Model (LLM) services such as ChatGPT, DALLE, and Cursor have quickly become essential for society, businesses, and individuals, empowering applications such as chatbots, image generation, and code assistance. The complexity of LLM systems makes them prone to failures and affects their reliability and availability, yet their failure patterns are not fully understood, making it an emerging problem. However, there are limited datasets and studies in this area, particularly lacking an open-access tool for analyzing LLM service failures based on incident reports. Addressing these problems, in this work we propose FAILS, the first open-sourced framework for incident reports collection and analysis on different LLM services and providers. FAILS provides comprehensive data collection, analysis, and visualization capabilities, including:(1) It can automatically collect, clean, and update incident data through its data scraper and processing components;(2) It provides 17 types of failure analysis, allowing users to explore temporal trends of incidents, analyze service reliability metrics, such as Mean Time to Recovery (MTTR) and Mean Time Between Failures (MTBF);(3) It leverages advanced LLM tools to assist in data analysis and interpretation, enabling users to gain observations and insights efficiently. All functions are integrated in the backend, allowing users to easily access them through a web-based frontend interface. FAILS supports researchers, engineers, and general users to understand failure patterns and further mitigate operational incidents and outages in LLM services. The framework is publicly available on https://github.com/atlarge-research/FAILS.
Large Language Models (LLMs) have demonstrated tremendous potential as the next-generation ranking-based recommendation system. Many recent works have shown that LLMs can significantly outperform conventional click-through-rate (CTR) prediction approaches. Despite such promising results, the computational inefficiency inherent in the current training paradigm makes it particularly challenging to train LLMs for ranking-based recommendation tasks on large datasets. To train LLMs for CTR prediction, most existing studies adopt the prevalent ''sliding-window'' paradigm. Given a sequence of $m$ user interactions, a unique training prompt is constructed for each interaction by designating it as the prediction target along with its preceding $n$ interactions serving as context. In turn, the sliding-window paradigm results in an overall complexity of $O(mn^2)$ that scales linearly with the length of user interactions. Consequently, a direct adoption to train LLMs with such strategy can result in prohibitively high training costs as the length of interactions grows. To alleviate the computational inefficiency, we propose a novel training paradigm, namely Dynamic Target Isolation (DTI), that structurally parallelizes the training of $k$ (where $k >> 1$) target interactions. Furthermore, we identify two major bottlenecks - hidden-state leakage and positional bias overfitting - that limit DTI to only scale up to a small value of $k$ (e.g., 5) then propose a computationally light solution to effectively tackle each. Through extensive experiments on three widely adopted public CTR datasets, we empirically show that DTI reduces training time by an average of $\textbf{92%}$ (e.g., from $70.5$ hrs to $5.31$ hrs), without compromising CTR prediction performance.
Recent studies in prompting large language model (LLM) for document-level machine translation (DMT) primarily focus on the inter-sentence context by flatting the source document into a long sequence. This approach relies solely on the sequence of sentences within the document. However, the complexity of document-level sequences is greater than that of shorter sentence-level sequences, which may limit LLM's ability in DMT when only this single-source knowledge is used. In this paper, we propose an enhanced approach by incorporating multiple sources of knowledge, including both the document summarization and entity translation, to enhance the performance of LLM-based DMT. Given a source document, we first obtain its summarization and translation of entities via LLM as the additional knowledge. We then utilize LLMs to generate two translations of the source document by fusing these two single knowledge sources, respectively. Finally, recognizing that different sources of knowledge may aid or hinder the translation of different sentences, we refine and rank the translations by leveraging a multi-knowledge fusion strategy to ensure the best results. Experimental results in eight document-level translation tasks show that our approach achieves an average improvement of 0.8, 0.6, and 0.4 COMET scores over the baseline without extra knowledge for LLaMA3-8B-Instruct, Mistral-Nemo-Instruct, and GPT-4o-mini, respectively.
Process reward models (PRMs) have shown success in complex reasoning tasks for large language models (LLMs). However, their application to machine translation (MT) remains underexplored due to the lack of systematic methodologies and evaluation benchmarks. To address this gap, we introduce \textbf{MT-RewardTree}, a comprehensive framework for constructing, evaluating, and deploying process reward models in MT. Unlike traditional vanilla preference pair construction, we propose a novel method for automatically generating token-level preference pairs using approximate Monte Carlo Tree Search (MCTS), which mitigates the prohibitive cost of human annotation for fine-grained steps. Then, we establish the first MT-specific reward model benchmark and provide a systematic comparison of different reward modeling architectures, revealing that token-level supervision effectively captures fine-grained preferences. Experimental results demonstrate that our MT-PRM-Qwen-2.5-3B achieves state-of-the-art performance in both token-level and sequence-level evaluation given the same input prefix. Furthermore, we showcase practical applications where PRMs enable test-time alignment for LLMs without additional alignment training and significantly improve performance in hypothesis ensembling. Our work provides valuable insights into the role of reward models in MT research. Our code and data are released in \href{https://sabijun.github.io/MT_RewardTreePage/}{https://sabijun.github.io/MT\_RewardTreePage}.
Automated audio captioning is a task that generates textual descriptions for audio content, and recent studies have explored using visual information to enhance captioning quality. However, current methods often fail to effectively fuse audio and visual data, missing important semantic cues from each modality. To address this, we introduce LAVCap, a large language model (LLM)-based audio-visual captioning framework that effectively integrates visual information with audio to improve audio captioning performance. LAVCap employs an optimal transport-based alignment loss to bridge the modality gap between audio and visual features, enabling more effective semantic extraction. Additionally, we propose an optimal transport attention module that enhances audio-visual fusion using an optimal transport assignment map. Combined with the optimal training strategy, experimental results demonstrate that each component of our framework is effective. LAVCap outperforms existing state-of-the-art methods on the AudioCaps dataset, without relying on large datasets or post-processing. Code is available at https://github.com/NAVER-INTEL-Co-Lab/gaudi-lavcap.
In real-world scenarios, most of the data obtained from information retrieval (IR) system is unstructured. Converting natural language sentences into structured Knowledge Graphs (KGs) remains a critical challenge. The quality of constructed KGs may also impact the performance of some KG-dependent domains like GraphRAG systems and recommendation systems. Recently, Large Language Models (LLMs) have demonstrated impressive capabilities in addressing a wide range of natural language processing tasks. However, there are still challenges when utilizing LLMs to address the task of generating structured KGs. And we have identified three limitations with respect to existing KG construction methods. (1)There is a large amount of information and excessive noise in real-world documents, which could result in extracting messy information. (2)Native LLMs struggle to effectively extract accuracy knowledge from some domain-specific documents. (3)Hallucinations phenomenon cannot be overlooked when utilizing LLMs directly as an unsupervised method for constructing KGs. In this paper, we propose GraphJudger, a knowledge graph construction framework to address the aforementioned challenges. We introduce three innovative modules in our method, which are entity-centric iterative text denoising, knowledge aware instruction tuning and graph judgement, respectively. We seek to utilize the capacity of LLMs to function as a graph judger, a capability superior to their role only as a predictor for KG construction problems. Experiments conducted on two general text-graph pair datasets and one domain-specific text-graph pair dataset show superior performances compared to baseline methods. The code of our proposed method is available at https://github.com/hhy-huang/GraphJudger.
Large Language Models (LLMs) are increasingly being used to automate programming tasks. Yet, LLMs' capabilities in reasoning about program semantics are still inadequately studied, leaving significant potential for further exploration. This paper introduces FormalBench, a comprehensive benchmark designed to evaluate LLMs' reasoning abilities on program semantics, particularly via the task of synthesizing formal program specifications to assist verifying program correctness. This task requires both comprehensive reasoning over all possible program executions and the generation of precise, syntactically correct expressions that adhere to formal syntax and semantics. Using this benchmark, we evaluated the ability of LLMs in synthesizing consistent and complete specifications. Our findings show that LLMs perform well with simple control flows but struggle with more complex structures, especially loops, even with advanced prompting. Additionally, LLMs exhibit limited robustness against semantic-preserving transformations. We also highlight common failure patterns and design self-repair prompts, improving success rates by 25%.
Scientific problem-solving involves synthesizing information while applying expert knowledge. We introduce CURIE, a scientific long-Context Understanding,Reasoning and Information Extraction benchmark to measure the potential of Large Language Models (LLMs) in scientific problem-solving and assisting scientists in realistic workflows. This benchmark introduces ten challenging tasks with a total of 580 problems and solution pairs curated by experts in six disciplines - materials science, condensed matter physics, quantum computing, geospatial analysis, biodiversity, and proteins - covering both experimental and theoretical work-flows in science. We evaluate a range of closed and open LLMs on tasks in CURIE which requires domain expertise, comprehension of long in-context information,and multi-step reasoning. While Gemini Flash 2.0 and Claude-3 show consistent high comprehension across domains, the popular GPT-4o and command-R+ fail dramatically on protein sequencing tasks. With the best performance at 32% there is much room for improvement for all models. We hope that insights gained from CURIE can guide the future development of LLMs in sciences. Evaluation code and data are in https://github.com/google/curie
This paper presents RAG-KG-IL, a novel multi-agent hybrid framework designed to enhance the reasoning capabilities of Large Language Models (LLMs) by integrating Retrieval-Augmented Generation (RAG) and Knowledge Graphs (KGs) with an Incremental Learning (IL) approach. Despite recent advancements, LLMs still face significant challenges in reasoning with structured data, handling dynamic knowledge evolution, and mitigating hallucinations, particularly in mission-critical domains. Our proposed RAG-KG-IL framework addresses these limitations by employing a multi-agent architecture that enables continuous knowledge updates, integrates structured knowledge, and incorporates autonomous agents for enhanced explainability and reasoning. The framework utilizes RAG to ensure the generated responses are grounded in verifiable information, while KGs provide structured domain knowledge for improved consistency and depth of understanding. The Incremental Learning approach allows for dynamic updates to the knowledge base without full retraining, significantly reducing computational overhead and improving the model's adaptability. We evaluate the framework using real-world case studies involving health-related queries, comparing it to state-of-the-art models like GPT-4o and a RAG-only baseline. Experimental results demonstrate that our approach significantly reduces hallucination rates and improves answer completeness and reasoning accuracy. The results underscore the potential of combining RAG, KGs, and multi-agent systems to create intelligent, adaptable systems capable of real-time knowledge integration and reasoning in complex domains.
The rise of large language models (LLMs) has revolutionized natural language processing (NLP), yet the influence of prompt sentiment, a latent affective characteristic of input text, remains underexplored. This study systematically examines how sentiment variations in prompts affect LLM-generated outputs in terms of coherence, factuality, and bias. Leveraging both lexicon-based and transformer-based sentiment analysis methods, we categorize prompts and evaluate responses from five leading LLMs: Claude, DeepSeek, GPT-4, Gemini, and LLaMA. Our analysis spans six AI-driven applications, including content generation, conversational AI, legal and financial analysis, healthcare AI, creative writing, and technical documentation. By transforming prompts, we assess their impact on output quality. Our findings reveal that prompt sentiment significantly influences model responses, with negative prompts often reducing factual accuracy and amplifying bias, while positive prompts tend to increase verbosity and sentiment propagation. These results highlight the importance of sentiment-aware prompt engineering for ensuring fair and reliable AI-generated content.
Analysis and comprehension of assembly code are crucial in various applications, such as reverse engineering. However, the low information density and lack of explicit syntactic structures in assembly code pose significant challenges. Pioneering approaches with masked language modeling (MLM)-based methods have been limited by facilitating natural language interaction. While recent methods based on decoder-focused large language models (LLMs) have significantly enhanced semantic representation, they still struggle to capture the nuanced and sparse semantics in assembly code. In this paper, we propose Assembly Augmented Tuning (ASMA-Tune), an end-to-end structural-semantic instruction-tuning framework. Our approach synergizes encoder architectures with decoder-based LLMs through projector modules to enable comprehensive code understanding. Experiments show that ASMA-Tune outperforms existing benchmarks, significantly enhancing assembly code comprehension and instruction-following abilities. Our model and dataset are public at https://github.com/wxy3596/ASMA-Tune.
LLMs are often claimed to be capable of Natural Language Inference (NLI), which is widely regarded as a cornerstone of more complex forms of reasoning. However, recent works show that LLMs still suffer from hallucinations in NLI due to attestation bias, where LLMs overly rely on propositional memory to build shortcuts. To solve the issue, we design an unsupervised framework to construct counterfactual reasoning data and fine-tune LLMs to reduce attestation bias. To measure bias reduction, we build bias-adversarial variants of NLI datasets with randomly replaced predicates in premises while keeping hypotheses unchanged. Extensive evaluations show that our framework can significantly reduce hallucinations from attestation bias. Then, we further evaluate LLMs fine-tuned with our framework on original NLI datasets and their bias-neutralized versions, where original entities are replaced with randomly sampled ones. Extensive results show that our framework consistently improves inferential performance on both original and bias-neutralized NLI datasets.
Large language models (LLMs) are used in chatbots or AI assistants to hold conversations with a human user. In such applications, the quality (e.g., user engagement, safety) of a conversation is important and can only be exactly known at the end of the conversation. To maximize its expected quality, conversation planning reasons about the stochastic transitions within a conversation to select the optimal LLM response at each turn. Existing simulation-based conversation planning algorithms typically select the optimal response by simulating future conversations with a large number of LLM queries at every turn. However, this process is extremely time-consuming and hence impractical for real-time conversations. This paper presents a novel approach called Semantic space COnversation Planning with improved Efficiency (SCOPE) that exploits the dense semantic representation of conversations to perform conversation planning efficiently. In particular, SCOPE models the stochastic transitions in conversation semantics and their associated rewards to plan entirely within the semantic space. This allows us to select the optimal LLM response at every conversation turn without needing additional LLM queries for simulation. As a result, SCOPE can perform conversation planning 70 times faster than conventional simulation-based planning algorithms when applied to a wide variety of conversation starters and two reward functions seen in the real world, yet achieving a higher reward within a practical planning budget. Our code can be found at: https://github.com/chenzhiliang94/convo-plan-SCOPE.
Large language models (LLMs) have shown outstanding performance across numerous real-world tasks. However, the autoregressive nature of these models makes the inference process slow and costly. Speculative decoding has emerged as a promising solution, leveraging a smaller auxiliary model to draft future tokens, which are then validated simultaneously by the larger model, achieving a speed-up of 1-2x. Although speculative decoding matches the same distribution as multinomial sampling, multinomial sampling itself is prone to suboptimal outputs, whereas beam sampling is widely recognized for producing higher-quality results by maintaining multiple candidate sequences at each step. This paper explores the novel integration of speculative decoding with beam sampling. However, there are four key challenges: (1) how to generate multiple sequences from the larger model's distribution given drafts sequences from the small model; (2) how to dynamically optimize the number of beams to balance efficiency and accuracy; (3) how to efficiently verify the multiple drafts in parallel; and (4) how to address the extra memory costs inherent in beam sampling. To address these challenges, we propose dynamic-width speculative beam decoding (DSBD). Specifically, we first introduce a novel draft and verification scheme that generates multiple sequences following the large model's distribution based on beam sampling trajectories from the small model. Then, we introduce an adaptive mechanism to dynamically tune the number of beams based on the context, optimizing efficiency and effectiveness. Besides, we extend tree-based parallel verification to handle multiple trees simultaneously, accelerating the verification process. Finally, we illustrate a simple modification to our algorithm to mitigate the memory overhead of beam sampling...
As large language models (LLMs) continue to demonstrate remarkable abilities across various domains, computer scientists are developing methods to understand their cognitive processes, particularly concerning how (and if) LLMs internally represent their beliefs about the world. However, this field currently lacks a unified theoretical foundation to underpin the study of belief in LLMs. This article begins filling this gap by proposing adequacy conditions for a representation in an LLM to count as belief-like. We argue that, while the project of belief measurement in LLMs shares striking features with belief measurement as carried out in decision theory and formal epistemology, it also differs in ways that should change how we measure belief. Thus, drawing from insights in philosophy and contemporary practices of machine learning, we establish four criteria that balance theoretical considerations with practical constraints. Our proposed criteria include accuracy, coherence, uniformity, and use, which together help lay the groundwork for a comprehensive understanding of belief representation in LLMs. We draw on empirical work showing the limitations of using various criteria in isolation to identify belief representations.
Human processes video reasoning in a sequential spatio-temporal reasoning logic, we first identify the relevant frames ("when") and then analyse the spatial relationships ("where") between key objects, and finally leverage these relationships to draw inferences ("what"). However, can Video Large Language Models (Video-LLMs) also "reason through a sequential spatio-temporal logic" in videos? Existing Video-LLM benchmarks primarily focus on assessing object presence, neglecting relational reasoning. Consequently, it is difficult to measure whether a model truly comprehends object interactions (actions/events) in videos or merely relies on pre-trained "memory" of co-occurrences as biases in generating answers. In this work, we introduce a Video Spatio-Temporal Reasoning (V-STaR) benchmark to address these shortcomings. The key idea is to decompose video understanding into a Reverse Spatio-Temporal Reasoning (RSTR) task that simultaneously evaluates what objects are present, when events occur, and where they are located while capturing the underlying Chain-of-thought (CoT) logic. To support this evaluation, we construct a dataset to elicit the spatial-temporal reasoning process of Video-LLMs. It contains coarse-to-fine CoT questions generated by a semi-automated GPT-4-powered pipeline, embedding explicit reasoning chains to mimic human cognition. Experiments from 14 Video-LLMs on our V-STaR reveal significant gaps between current Video-LLMs and the needs for robust and consistent spatio-temporal reasoning.
In this work, a thorough mathematical framework for incorporating Large Language Models (LLMs) into gamified systems is presented with an emphasis on improving task dynamics, user engagement, and reward systems. Personalized feedback, adaptive learning, and dynamic content creation are all made possible by integrating LLMs and are crucial for improving user engagement and system performance. A simulated environment tests the framework's adaptability and demonstrates its potential for real-world applications in various industries, including business, healthcare, and education. The findings demonstrate how LLMs can offer customized experiences that raise system effectiveness and user retention. This study also examines the difficulties this framework aims to solve, highlighting its importance in maximizing involvement and encouraging sustained behavioral change in a range of sectors.
Recent advancements in instruction tuning for large language models (LLMs) suggest that a small, high-quality dataset can significantly equip LLMs with instruction-following capabilities, outperforming large datasets often burdened by quality and redundancy issues. However, the challenge lies in automatically identifying valuable subsets from large datasets to boost both the effectiveness and efficiency of instruction tuning. In this paper, we first establish data selection criteria based on three distinct aspects of data value: diversity, difficulty, and dependability, and then propose the D3 method comprising two key steps of scoring and selection. Specifically, in the scoring step, we define the diversity function to measure sample distinctiveness and introduce the uncertainty-based prediction difficulty to evaluate sample difficulty by mitigating the interference of context-oriented generation diversity. Additionally, we integrate an external LLM for dependability assessment. In the selection step, we formulate the D3 weighted coreset objective, which jointly optimizes three aspects of data value to solve for the most valuable subset. The two steps of D3 can iterate multiple rounds, incorporating feedback to refine the selection focus adaptively. Experiments on three datasets demonstrate the effectiveness of D3 in endowing LLMs with competitive or even superior instruction-following capabilities using less than 10% of the entire dataset.
Long-form video understanding is complicated by the high redundancy of video data and the abundance of query-irrelevant information. To tackle these challenges, we propose VideoTree, a training-free framework which builds a query-adaptive and hierarchical video representation for LLM reasoning over long-form videos. First, VideoTree extracts query-relevant information from the input video through an iterative process, progressively refining the selection of keyframes based on their relevance to the query. Furthermore, VideoTree leverages the inherent hierarchical structure of long video data, which is often overlooked by existing LLM-based methods. Specifically, we incorporate multi-granularity information into a tree-based representation, allowing VideoTree to extract query-relevant details from long videos in a coarse-to-fine manner. This enables the model to effectively handle a wide range of video queries with varying levels of detail. Finally, VideoTree aggregates the hierarchical query-relevant information within the tree structure and feeds it into an LLM reasoning model to answer the query. Our experiments show that our method improves both reasoning accuracy and efficiency. Specifically, VideoTree outperforms existing training-free approaches on EgoSchema and NExT-QA with less inference time, achieving 61.1% and 75.6% accuracy on the test set without additional video-specific training. Moreover, on the long split of Video-MME (average 44 minutes), VideoTree achieves better performance than GPT-4V and many other MLLMs that were extensively trained on video data.
Large Language Models (LLMs) have shown impressive reasoning capabilities in well-defined problems with clear solutions, such as mathematics and coding. However, they still struggle with complex real-world scenarios like business negotiations, which require strategic reasoning-an ability to navigate dynamic environments and align long-term goals amidst uncertainty. Existing methods for strategic reasoning face challenges in adaptability, scalability, and transferring strategies to new contexts. To address these issues, we propose explicit policy optimization (EPO) for strategic reasoning, featuring an LLM that provides strategies in open-ended action space and can be plugged into arbitrary LLM agents to motivate goal-directed behavior. To improve adaptability and policy transferability, we train the strategic reasoning model via multi-turn reinforcement learning (RL) using process rewards and iterative self-play, without supervised fine-tuning (SFT) as a preliminary step. Experiments across social and physical domains demonstrate EPO's ability of long-term goal alignment through enhanced strategic reasoning, achieving state-of-the-art performance on social dialogue and web navigation tasks. Our findings reveal various collaborative reasoning mechanisms emergent in EPO and its effectiveness in generating novel strategies, underscoring its potential for strategic reasoning in real-world applications.
MMS-LLaMA: Efficient LLM-based Audio-Visual Speech Recognition with Minimal Multimodal Speech Tokens
Audio-Visual Speech Recognition (AVSR) achieves robust speech recognition in noisy environments by combining auditory and visual information. However, recent Large Language Model (LLM) based AVSR systems incur high computational costs due to the high temporal resolution of audio-visual speech processed by LLMs. In this work, we introduce an efficient multimodal speech LLM framework that minimizes token length while preserving essential linguistic content. Our approach employs an early av-fusion module for streamlined feature integration, an audio-visual speech Q-Former that dynamically allocates tokens based on input duration, and a refined query allocation strategy with a speech rate predictor to adjust token allocation according to speaking speed of each audio sample. Extensive experiments on the LRS3 dataset show that our method achieves state-of-the-art performance with a WER of 0.74% while using only 3.5 tokens per second. Moreover, our approach not only reduces token usage by 86% compared to the previous multimodal speech LLM framework, but also improves computational efficiency by reducing FLOPs by 35.7%.
As LLMs grow more powerful, their most profound achievement may be recognising when to say "I don't know". Existing studies on LLM self-knowledge have been largely constrained by human-defined notions of feasibility, often neglecting the reasons behind unanswerability by LLMs and failing to study deficient types of self-knowledge. This study aims to obtain intrinsic insights into different types of LLM self-knowledge with a novel methodology: allowing them the flexibility to set their own feasibility boundaries and then analysing the consistency of these limits. We find that even frontier models like GPT-4o and Mistral Large are not sure of their own capabilities more than 80% of the time, highlighting a significant lack of trustworthiness in responses. Our analysis of confidence balance in LLMs indicates that models swing between overconfidence and conservatism in feasibility boundaries depending on task categories and that the most significant self-knowledge weaknesses lie in temporal awareness and contextual understanding. These difficulties in contextual comprehension additionally lead models to question their operational boundaries, resulting in considerable confusion within the self-knowledge of LLMs. We make our code and results available publicly at https://github.com/knowledge-verse-ai/LLM-Self_Knowledge_Eval
This paper introduces UniTranslator, a visionary framework that re-imagines code translation as a collaborative endeavor among multiple, compact LLMs. By orchestrating the interaction of specialized agents, each focused on different aspects of the translation process and grounded in a deep understanding of programming concepts, UniTranslator achieves a level of accuracy and efficiency that rivals larger, monolithic models. Our preliminary evaluation demonstrates the potential of UniTranslator to overcome the limitations of existing approaches and unlock the power of smaller LLMs for complex code translation tasks. We explore the effectiveness of this dynamic multi-agent paradigm in handling diverse language pairs, including low-resource languages, and in mitigating common issues such as code artifacts and hallucinations through the use of Natural Language Inference (NLI) grounding and iterative feedback mechanisms
Large Language Models (LLMs) have demonstrated remarkable capabilities in natural language processing but also pose significant privacy risks by memorizing and leaking Personally Identifiable Information (PII). Existing mitigation strategies, such as differential privacy and neuron-level interventions, often degrade model utility or fail to effectively prevent leakage. To address this challenge, we introduce PrivacyScalpel, a novel privacy-preserving framework that leverages LLM interpretability techniques to identify and mitigate PII leakage while maintaining performance. PrivacyScalpel comprises three key steps: (1) Feature Probing, which identifies layers in the model that encode PII-rich representations, (2) Sparse Autoencoding, where a k-Sparse Autoencoder (k-SAE) disentangles and isolates privacy-sensitive features, and (3) Feature-Level Interventions, which employ targeted ablation and vector steering to suppress PII leakage. Our empirical evaluation on Gemma2-2b and Llama2-7b, fine-tuned on the Enron dataset, shows that PrivacyScalpel significantly reduces email leakage from 5.15\% to as low as 0.0\%, while maintaining over 99.4\% of the original model's utility. Notably, our method outperforms neuron-level interventions in privacy-utility trade-offs, demonstrating that acting on sparse, monosemantic features is more effective than manipulating polysemantic neurons. Beyond improving LLM privacy, our approach offers insights into the mechanisms underlying PII memorization, contributing to the broader field of model interpretability and secure AI deployment.
Chain-based reasoning methods like chain of thought (CoT) play a rising role in solving reasoning tasks for large language models (LLMs). However, the causal illusions between \textit{a step of reasoning} and \textit{corresponding state transitions} are becoming a significant obstacle to advancing LLMs' reasoning capabilities, especially in long-range reasoning tasks. This paper proposes a non-chain-based reasoning framework for simultaneous consideration of causal significance and consistency, i.e., the Causal Significance and Consistency Enhancer (CSCE). We customize LLM's loss function utilizing treatment effect assessments to enhance its reasoning ability from two aspects: causal significance and consistency. This ensures that the model captures essential causal relationships and maintains robust and consistent performance across various scenarios. Additionally, we transform the reasoning process from the cascading multiple one-step reasoning commonly used in Chain-Based methods, like CoT, to a causal-enhanced method that outputs the entire reasoning process in one go, further improving the model's reasoning efficiency. Extensive experiments show that our method improves both the reasoning success rate and speed. These improvements further demonstrate that non-chain-based methods can also aid LLMs in completing reasoning tasks.
Training-free video large language models (LLMs) leverage pretrained Image LLMs to process video content without the need for further training. A key challenge in such approaches is the difficulty of retaining essential visual and temporal information, constrained by the token limits in Image LLMs. To address this, we propose a two-stage method for selecting query-relevant tokens based on the LLM attention scores: compressing the video sequence and then expanding the sequence. However, during the compression stage, Image LLMs often exhibit a positional attention bias in video sequences, where attention is overly concentrated on later frames, causing early-frame information to be underutilized. To alleviate this attention bias during sequence compression, we propose Gridded Attention Pooling for preserving spatiotemporal structure. Additionally, we introduce Visual Summarization Tail to effectively utilize this bias, facilitating overall video understanding during sequence expansion. In this way, our method effectively Mitigates and Leverages attention Bias (LLaVA-MLB), enabling the frozen Image LLM for detailed video understanding. Experiments on several benchmarks demonstrate that our approach outperforms state-of-the-art methods, achieving superior performance in both efficiency and accuracy. Our code will be released.
N:M structured pruning is essential for large language models (LLMs) because it can remove less important network weights and reduce the memory and computation requirements. Existing pruning methods mainly focus on designing metrics to measure the importance of network components to guide pruning. Apart from the impact of these metrics, we observe that different layers have different sensitivities over the network performance. Thus, we propose an efficient method based on the trace of Fisher Information Matrix (FIM) to quantitatively measure and verify the different sensitivities across layers. Based on this, we propose Mixed Sparsity Pruning (MSP) which uses a pruning-oriented evolutionary algorithm (EA) to determine the optimal sparsity levels for different layers. To guarantee fast convergence and achieve promising performance, we utilize efficient FIM-inspired layer-wise sensitivity to initialize the population of EA. In addition, our MSP can work as a plug-and-play module, ready to be integrated into existing pruning methods. Extensive experiments on LLaMA and LLaMA-2 on language modeling and zero-shot tasks demonstrate our superior performance. In particular, in extreme pruning ratio (e.g. 75%), our method significantly outperforms existing methods in terms of perplexity (PPL) by orders of magnitude (Figure 1).
Recent research on Reasoning of Large Language Models (LLMs) has sought to further enhance their performance by integrating meta-thinking -- enabling models to monitor, evaluate, and control their reasoning processes for more adaptive and effective problem-solving. However, current single-agent work lacks a specialized design for acquiring meta-thinking, resulting in low efficacy. To address this challenge, we introduce Reinforced Meta-thinking Agents (ReMA), a novel framework that leverages Multi-Agent Reinforcement Learning (MARL) to elicit meta-thinking behaviors, encouraging LLMs to think about thinking. ReMA decouples the reasoning process into two hierarchical agents: a high-level meta-thinking agent responsible for generating strategic oversight and plans, and a low-level reasoning agent for detailed executions. Through iterative reinforcement learning with aligned objectives, these agents explore and learn collaboration, leading to improved generalization and robustness. Experimental results demonstrate that ReMA outperforms single-agent RL baselines on complex reasoning tasks, including competitive-level mathematical benchmarks and LLM-as-a-Judge benchmarks. Comprehensive ablation studies further illustrate the evolving dynamics of each distinct agent, providing valuable insights into how the meta-thinking reasoning process enhances the reasoning capabilities of LLMs.
Large Language Models (LLMs) have demonstrated remarkable performance in code completion. However, the training data used to develop these models often contain a significant amount of buggy code. Yet, it remains unclear to what extent these buggy instances influence LLMs' performance when tackling bug-prone code completion tasks. To fill this gap, this paper presents the first empirical study evaluating the performance of LLMs in completing bug-prone code. Through extensive experiments on 7 LLMs and the Defects4J dataset, we analyze LLMs' accuracy, robustness, and limitations in this challenging context. Our experimental results show that completing bug-prone code is significantly more challenging for LLMs than completing normal code. Notably, in bug-prone tasks, the likelihood of LLMs generating correct code is nearly the same as generating buggy code, and it is substantially lower than in normal code completion tasks (e.g., 12.27% vs. 29.85% for GPT-4). To our surprise, 44.44% of the bugs LLMs make are completely identical to the pre-fix version, indicating that LLMs have been seriously biased by historical bugs when completing code. Additionally, we investigate the effectiveness of existing post-processing techniques and find that while they can improve consistency, they do not significantly reduce error rates in bug-prone code scenarios. Our research highlights the limitations of current LLMs in handling bug-prone code and underscores the need for improved models and post-processing strategies to enhance code completion accuracy in real-world development environments.
Model editing aims to correct outdated or erroneous knowledge in large language models (LLMs) without the need for costly retraining. Lifelong model editing is the most challenging task that caters to the continuous editing requirements of LLMs. Prior works primarily focus on single or batch editing; nevertheless, these methods fall short in lifelong editing scenarios due to catastrophic knowledge forgetting and the degradation of model performance. Although retrieval-based methods alleviate these issues, they are impeded by slow and cumbersome processes of integrating the retrieved knowledge into the model. In this work, we introduce RECIPE, a RetriEval-augmented ContInuous Prompt lEarning method, to boost editing efficacy and inference efficiency in lifelong learning. RECIPE first converts knowledge statements into short and informative continuous prompts, prefixed to the LLM's input query embedding, to efficiently refine the response grounded on the knowledge. It further integrates the Knowledge Sentinel (KS) that acts as an intermediary to calculate a dynamic threshold, determining whether the retrieval repository contains relevant knowledge. Our retriever and prompt encoder are jointly trained to achieve editing properties, i.e., reliability, generality, and locality. In our experiments, RECIPE is assessed extensively across multiple LLMs and editing datasets, where it achieves superior editing performance. RECIPE also demonstrates its capability to maintain the overall performance of LLMs alongside showcasing fast editing and inference speed.
The emergence of large language models (LLMs) has revolutionized AI development, yet their training demands computational resources beyond a single cluster or even datacenter, limiting accessibility to large organizations. Decentralized training has emerged as a promising paradigm to leverage dispersed resources across clusters, datacenters, and global regions, democratizing LLM development for broader communities. As the first comprehensive exploration of this emerging field, we present decentralized LLM training as a resource-driven paradigm and categorize it into community-driven and organizational approaches. Furthermore, our in-depth analysis clarifies decentralized LLM training, including: (1) position with related domain concepts comparison, (2) decentralized resource development trends, and (3) recent advances with discussion under a novel taxonomy. We also provide up-to-date case studies and explore future directions, contributing to the evolution of decentralized LLM training research.
Software engineers use code-fluent large language models (LLMs) to help explain unfamiliar code, yet LLM explanations are not adapted to engineers' diverse problem-solving needs. We prompted an LLM to adapt to five problem-solving style types from an inclusive design method, the Gender Inclusiveness Magnifier (GenderMag). We ran a user study with software engineers to examine the impact of explanation adaptations on software engineers' perceptions, both for explanations which matched and mismatched engineers' problem-solving styles. We found that explanations were more frequently beneficial when they matched problem-solving style, but not every matching adaptation was equally beneficial; in some instances, diverse engineers found as much (or more) benefit from mismatched adaptations. Through an equity and inclusivity lens, our work highlights the benefits of having an LLM adapt its explanations to match engineers' diverse problem-solving style values, the potential harms when matched adaptations were not perceived well by engineers, and a comparison of how matching and mismatching LLM adaptations impacted diverse engineers.
Aligning large language models (LLMs) with diverse human preferences is critical for ensuring fairness and informed outcomes when deploying these models for decision-making. In this paper, we seek to uncover fundamental statistical limits concerning aligning LLMs with human preferences, with a focus on the probabilistic representation of human preferences and the preservation of diverse preferences in aligned LLMs. We first show that human preferences can be represented by a reward model if and only if the preference among LLM-generated responses is free of any Condorcet cycle. Moreover, we prove that Condorcet cycles exist with probability converging to one exponentially fast under a probabilistic preference model, thereby demonstrating the impossibility of fully aligning human preferences using reward-based approaches such as reinforcement learning from human feedback. Next, we explore the conditions under which LLMs would employ mixed strategies -- meaning they do not collapse to a single response -- when aligned in the limit using a non-reward-based approach, such as Nash learning from human feedback (NLHF). We identify a necessary and sufficient condition for mixed strategies: the absence of a response that is preferred over all others by a majority. As a blessing, we prove that this condition holds with high probability under the probabilistic preference model, thereby highlighting the statistical possibility of preserving minority preferences without explicit regularization in aligning LLMs. Finally, we leverage insights from our statistical results to design a novel, computationally efficient algorithm for finding Nash equilibria in aligning LLMs with NLHF. Our experiments show that Llama-3.2-1B, aligned with our algorithm, achieves a win rate of 60.55\% against the base model.
Large Language Models (LLMs) have shown notable potential in code generation for optimization algorithms, unlocking exciting new opportunities. This paper examines how LLMs, rather than creating algorithms from scratch, can improve existing ones without the need for specialized expertise. To explore this potential, we selected 10 baseline optimization algorithms from various domains (metaheuristics, reinforcement learning, deterministic, and exact methods) to solve the classic Travelling Salesman Problem. The results show that our simple methodology often results in LLM-generated algorithm variants that improve over the baseline algorithms in terms of solution quality, reduction in computational time, and simplification of code complexity, all without requiring specialized optimization knowledge or advanced algorithmic implementation skills.
With the increasing prevalence of mental health conditions worldwide, AI-powered chatbots and conversational agents have emerged as accessible tools to support mental health. However, deploying Large Language Models (LLMs) in mental healthcare applications raises significant privacy concerns, especially regarding regulations like HIPAA and GDPR. In this work, we propose FedMentalCare, a privacy-preserving framework that leverages Federated Learning (FL) combined with Low-Rank Adaptation (LoRA) to fine-tune LLMs for mental health analysis. We investigate the performance impact of varying client data volumes and model architectures (e.g., MobileBERT and MiniLM) in FL environments. Our framework demonstrates a scalable, privacy-aware approach for deploying LLMs in real-world mental healthcare scenarios, addressing data security and computational efficiency challenges.
We present a novel way to integrate flexible, context-dependent constraints into combinatorial optimization by leveraging Large Language Models (LLMs) alongside traditional algorithms. Although LLMs excel at interpreting nuanced, locally specified requirements, they struggle with enforcing global combinatorial feasibility. To bridge this gap, we propose an iterated fine-tuning framework where algorithmic feedback progressively refines the LLM's output distribution. Interpreting this as simulated annealing, we introduce a formal model based on a "coarse learnability" assumption, providing sample complexity bounds for convergence. Empirical evaluations on scheduling, graph connectivity, and clustering tasks demonstrate that our framework balances the flexibility of locally expressed constraints with rigorous global optimization more effectively compared to baseline sampling methods. Our results highlight a promising direction for hybrid AI-driven combinatorial reasoning.
As large language models (LLMs) rapidly advance, evaluating their performance is critical. LLMs are trained on multilingual data, but their reasoning abilities are mainly evaluated using English datasets. Hence, robust evaluation frameworks are needed using high-quality non-English datasets, especially low-resource languages (LRLs). This study evaluates eight state-of-the-art (SOTA) LLMs on Latvian and Giriama using a Massive Multitask Language Understanding (MMLU) subset curated with native speakers for linguistic and cultural relevance. Giriama is benchmarked for the first time. Our evaluation shows that OpenAI's o1 model outperforms others across all languages, scoring 92.8\% in English, 88.8\% in Latvian, and 70.8\% in Giriama on 0-shot tasks. Mistral-large (35.6\%) and Llama-70B IT (41\%) have weak performance, on both Latvian and Giriama. Our results underscore the need for localized benchmarks and human evaluations in advancing cultural AI contextualization.
Large Language Models (LLMs) have demonstrated remarkable adaptability in performing various tasks, including machine translation (MT), without explicit training. Models such as OpenAI's GPT-4 and Google's Gemini are frequently evaluated on translation benchmarks and utilized as translation tools due to their high performance. This paper examines Gemini's performance in translating an 18th-century Ottoman Turkish manuscript, Prisoner of the Infidels: The Memoirs of Osman Agha of Timisoara, into English. The manuscript recounts the experiences of Osman Agha, an Ottoman subject who spent 11 years as a prisoner of war in Austria, and includes his accounts of warfare and violence. Our analysis reveals that Gemini's safety mechanisms flagged between 14 and 23 percent of the manuscript as harmful, resulting in untranslated passages. These safety settings, while effective in mitigating potential harm, hinder the model's ability to provide complete and accurate translations of historical texts. Through real historical examples, this study highlights the inherent challenges and limitations of current LLM safety implementations in the handling of sensitive and context-rich materials. These real-world instances underscore potential failures of LLMs in contemporary translation scenarios, where accurate and comprehensive translations are crucial-for example, translating the accounts of modern victims of war for legal proceedings or humanitarian documentation.
In controlled text generation using large language models (LLMs), gaps arise between the language model's interpretation and human expectations. We look at the problem of controlling emotions in keyword-based sentence generation for both GPT-4 and LLaMA-3. We selected four emotion representations: Words, Valence-Arousal-Dominance (VAD) dimensions expressed in both Lexical and Numeric forms, and Emojis. Our human evaluation looked at the Human-LLM alignment for each representation, as well as the accuracy and realism of the generated sentences. While representations like VAD break emotions into easy-to-compute components, our findings show that people agree more with how LLMs generate when conditioned on English words (e.g., "angry") rather than VAD scales. This difference is especially visible when comparing Numeric VAD to words. However, we found that converting the originally-numeric VAD scales to Lexical scales (e.g., +4.0 becomes "High") dramatically improved agreement. Furthermore, the perception of how much a generated sentence conveys an emotion is highly dependent on the LLM, representation type, and which emotion it is.
Large Language Models (LLMs) have demonstrated great potential as evaluators of NLG systems, allowing for high-quality, reference-free, and multi-aspect assessments. However, existing LLM-based metrics suffer from two major drawbacks: reliance on proprietary models to generate training data or perform evaluations, and a lack of fine-grained, explanatory feedback. In this paper, we introduce OpeNLGauge, a fully open-source, reference-free NLG evaluation metric that provides accurate explanations based on error spans. OpeNLGauge is available as a two-stage ensemble of larger open-weight LLMs, or as a small fine-tuned evaluation model, with confirmed generalizability to unseen tasks, domains and aspects. Our extensive meta-evaluation shows that OpeNLGauge achieves competitive correlation with human judgments, outperforming state-of-the-art models on certain tasks while maintaining full reproducibility and providing explanations more than twice as accurate.
Multi-human multi-robot teams are increasingly recognized for their efficiency in executing large-scale, complex tasks by integrating heterogeneous yet potentially synergistic humans and robots. However, this inherent heterogeneity presents significant challenges in teaming, necessitating efficient initial task allocation (ITA) strategies that optimally form complementary human-robot pairs or collaborative chains and establish well-matched task distributions. While current learning-based methods demonstrate promising performance, they often incur high computational costs and lack the flexibility to incorporate user preferences in multi-objective optimization (MOO) or adapt to last-minute changes in dynamic real-world environments. To address these limitations, we propose REBEL, an LLM-based ITA framework that integrates rule-based and experience-enhanced learning to enhance LLM reasoning capabilities and improve in-context adaptability to MOO and situational changes. Extensive experiments validate the effectiveness of REBEL in both single-objective and multi-objective scenarios, demonstrating superior alignment with user preferences and enhanced situational awareness to handle unexpected team composition changes. Additionally, we show that REBEL can complement pre-trained ITA policies, further boosting situational adaptability and overall team performance. Website at https://sites.google.com/view/ita-rebel .
Closed large language models (LLMs) such as GPT-4 have set state-of-the-art results across a number of NLP tasks and have become central to NLP and machine learning (ML)-driven solutions. Closed LLMs' performance and wide adoption has sparked considerable debate about their accessibility in terms of availability, cost, and transparency. In this study, we perform a rigorous comparative analysis of nine leading LLMs, spanning closed, open, and open-source LLM ecosystems, across text assessment and generation tasks related to automated essay scoring. Our findings reveal that for few-shot learning-based assessment of human generated essays, open LLMs such as Llama 3 and Qwen2.5 perform comparably to GPT-4 in terms of predictive performance, with no significant differences in disparate impact scores when considering age- or race-related fairness. Moreover, Llama 3 offers a substantial cost advantage, being up to 37 times more cost-efficient than GPT-4. For generative tasks, we find that essays generated by top open LLMs are comparable to closed LLMs in terms of their semantic composition/embeddings and ML assessed scores. Our findings challenge the dominance of closed LLMs and highlight the democratizing potential of open LLMs, suggesting they can effectively bridge accessibility divides while maintaining competitive performance and fairness.
Large Language Model (LLM) agents have demonstrated remarkable capabilities in automating tasks and driving innovation across diverse educational applications. In this survey, we provide a systematic review of state-of-the-art research on LLM agents in education, categorizing them into two broad classes: (1) \emph{Pedagogical Agents}, which focus on automating complex pedagogical tasks to support both teachers and students; and (2) \emph{Domain-Specific Educational Agents}, which are tailored for specialized fields such as science education, language learning, and professional development. We comprehensively examine the technological advancements underlying these LLM agents, including key datasets, benchmarks, and algorithmic frameworks that drive their effectiveness. Furthermore, we discuss critical challenges such as privacy, bias and fairness concerns, hallucination mitigation, and integration with existing educational ecosystems. This survey aims to provide a comprehensive technological overview of LLM agents for education, fostering further research and collaboration to enhance their impact for the greater good of learners and educators alike.
Prompting LLMs offers an efficient way to guide output generation without explicit model training. In the e-commerce domain, prompting-based applications are widely used for tasks such as query understanding, recommender systems, and customer support. However, adapting LLMs to different tasks often requires extensive prompt engineering by domain experts, along with frequent updates to align with evolving business needs. Additionally, crafting fully unbiased natural language prompts remains a challenge for humans. To address these challenges, we propose a novel framework, Examples as the Prompt (EaP) which leverages labeled data to enhance prompts. Specifically, EaP automatically selects the most representative examples to maximize the few-shot capability of LLMs. It is efficient due to its unsupervised example selection and adaptive to potential data distribution shifts. We validate EaP on four real-world production use cases, demonstrating that it achieves comparable or even superior performance comparing to hand-crafted prompts designed by domain experts. Additionally, we introduce EaP_lite, which entirely replaces the natural language components of prompts with labeled examples. EaP_lite improves LLM inference speed by up to 70% without compromising performance. Latest online A/B test shows that using EaP and EaP_lite for data labeling can bring significant composite revenue gain by 0.06%.
Motivated by the remarkable success of artificial intelligence (AI) across diverse fields, the application of AI to solve scientific problems-often formulated as partial differential equations (PDEs)-has garnered increasing attention. While most existing research concentrates on theoretical properties (such as well-posedness, regularity, and continuity) of the solutions, alongside direct AI-driven methods for solving PDEs, the challenge of uncovering symbolic relationships within these equations remains largely unexplored. In this paper, we propose leveraging large language models (LLMs) to learn such symbolic relationships. Our results demonstrate that LLMs can effectively predict the operators involved in PDE solutions by utilizing the symbolic information in the PDEs. Furthermore, we show that discovering these symbolic relationships can substantially improve both the efficiency and accuracy of the finite expression method for finding analytical approximation of PDE solutions, delivering a fully interpretable solution pipeline. This work opens new avenues for understanding the symbolic structure of scientific problems and advancing their solution processes.
Reinforcement learning (RL)-based large language models (LLMs), such as ChatGPT, DeepSeek, and Grok-3, have gained significant attention for their exceptional capabilities in natural language processing and multimodal data understanding. Meanwhile, the rapid expansion of information services has driven the growing need for intelligence, efficient, and adaptable wireless networks. Wireless networks require the empowerment of RL-based LLMs while these models also benefit from wireless networks to broaden their application scenarios. Specifically, RL-based LLMs can enhance wireless communication systems through intelligent resource allocation, adaptive network optimization, and real-time decision-making. Conversely, wireless networks provide a vital infrastructure for the efficient training, deployment, and distributed inference of RL-based LLMs, especially in decentralized and edge computing environments. This mutual empowerment highlights the need for a deeper exploration of the interplay between these two domains. We first review recent advancements in wireless communications, highlighting the associated challenges and potential solutions. We then discuss the progress of RL-based LLMs, focusing on key technologies for LLM training, challenges, and potential solutions. Subsequently, we explore the mutual empowerment between these two fields, highlighting key motivations, open challenges, and potential solutions. Finally, we provide insights into future directions, applications, and their societal impact to further explore this intersection, paving the way for next-generation intelligent communication systems. Overall, this survey provides a comprehensive overview of the relationship between RL-based LLMs and wireless networks, offering a vision where these domains empower each other to drive innovations.
Recent Multimodal Large Language Models (MLLMs) have demonstrated significant progress in perceiving and reasoning over multimodal inquiries, ushering in a new research era for foundation models. However, vision-language misalignment in MLLMs has emerged as a critical challenge, where the textual responses generated by these models are not factually aligned with the given text-image inputs. Existing efforts to address vision-language misalignment have focused on developing specialized vision-language connectors or leveraging visual instruction tuning from diverse domains. In this paper, we tackle this issue from a fundamental yet unexplored perspective by revisiting the core architecture of MLLMs. Most MLLMs are typically built on decoder-only LLMs consisting of a causal attention mechanism, which limits the ability of the earlier modalities (e.g., images) to incorporate information from the latter modalities (e.g., text). To address this problem, we propose \MapleLeaf AKI, a novel MLLM that unlocks causal attention into modality-mutual attention (MMA) to enable image tokens to attend to text tokens. This simple yet effective design allows AKI to achieve superior performance in 12 multimodal understanding benchmarks (+7.2% on average) without introducing additional parameters and increasing training time. Our MMA design is intended to be generic, allowing for application across various modalities, and scalable to accommodate diverse multimodal scenarios. The code and model are publicly available at https://github.com/sony/aki to encourage further advancements in MLLMs across various directions.
Ensuring Large Language Models (LLMs) align with diverse human preferences while preserving privacy and fairness remains a challenge. Existing methods, such as Reinforcement Learning from Human Feedback (RLHF), rely on centralized data collection, making them computationally expensive and privacy-invasive. We introduce PluralLLM a federated learning-based approach that enables multiple user groups to collaboratively train a transformer-based preference predictor without sharing sensitive data, which can also serve as a reward model for aligning LLMs. Our method leverages Federated Averaging (FedAvg) to aggregate preference updates efficiently, achieving 46% faster convergence, a 4% improvement in alignment scores, and nearly the same group fairness measure as in centralized training. Evaluated on a Q/A preference alignment task, PluralLLM demonstrates that federated preference learning offers a scalable and privacy-preserving alternative for aligning LLMs with diverse human values.
The adoption of Large Language Models (LLMs) is rapidly expanding across various tasks that involve inherent graphical structures. Graphs are integral to a wide range of applications, including motion planning for autonomous vehicles, social networks, scene understanding, and knowledge graphs. Many problems, even those not initially perceived as graph-based, can be effectively addressed through graph theory. However, when applied to these tasks, LLMs often encounter challenges, such as hallucinations and mathematical inaccuracies. To overcome these limitations, we propose Graph-Grounded LLMs, a system that improves LLM performance on graph-related tasks by integrating a graph library through function calls. By grounding LLMs in this manner, we demonstrate significant reductions in hallucinations and improved mathematical accuracy in solving graph-based problems, as evidenced by the performance on the NLGraph benchmark. Finally, we showcase a disaster rescue application where the Graph-Grounded LLM acts as a decision-support system.
While Large Language Models (LLMs) have significantly advanced natural language processing, aligning them with human preferences remains an open challenge. Although current alignment methods rely primarily on explicit feedback, eye-tracking (ET) data offers insights into real-time cognitive processing during reading. In this paper, we present OASST-ETC, a novel eye-tracking corpus capturing reading patterns from 24 participants, while evaluating LLM-generated responses from the OASST1 dataset. Our analysis reveals distinct reading patterns between preferred and non-preferred responses, which we compare with synthetic eye-tracking data. Furthermore, we examine the correlation between human reading measures and attention patterns from various transformer-based models, discovering stronger correlations in preferred responses. This work introduces a unique resource for studying human cognitive processing in LLM evaluation and suggests promising directions for incorporating eye-tracking data into alignment methods. The dataset and analysis code are publicly available.
Synthetic data generation is integral to ML pipelines, e.g., to augment training data, replace sensitive information, and even to power advanced platforms like DeepSeek. While LLMs fine-tuned for synthetic data generation are gaining traction, synthetic table generation -- a critical data type in business and science -- remains under-explored compared to text and image synthesis. This paper shows that LLMs, whether used as-is or after traditional fine-tuning, are inadequate for generating synthetic tables. Their autoregressive nature, combined with random order permutation during fine-tuning, hampers the modeling of functional dependencies and prevents capturing conditional mixtures of distributions essential for real-world constraints. We demonstrate that making LLMs permutation-aware can mitigate these issues.
The proliferation of Large Language Models (LLMs) in recent years has realized many applications in various domains. Being trained with a huge of amount of data coming from various sources, LLMs can be deployed to solve different tasks, including those in Software Engineering (SE). Though they have been widely adopted, the potential of using LLMs cooperatively has not been thoroughly investigated. In this paper, we proposed Metagente as a novel approach to amplify the synergy of various LLMs. Metagente is a Multi-Agent framework based on a series of LLMs to self-optimize the system through evaluation, feedback, and cooperation among specialized agents. Such a framework creates an environment where multiple agents iteratively refine and optimize prompts from various perspectives. The results of these explorations are then reviewed and aggregated by a teacher agent. To study its performance, we evaluated Metagente with an SE task, i.e., summarization of README.MD files, and compared it with three well-established baselines, i.e., GitSum, LLaMA-2, and GPT-4o. The results show that our proposed approach works efficiently and effectively, consuming a small amount of data for fine-tuning but still getting a high accuracy, thus substantially outperforming the baselines. The performance gain compared to GitSum, the most relevant benchmark, ranges from 27.63% to 60.43%. More importantly, compared to using only one LLM, Metagente boots up the accuracy to multiple folds.
This study explores how recent large language models (LLMs) navigate relative clause attachment {ambiguity} and use world knowledge biases for disambiguation in six typologically diverse languages: English, Chinese, Japanese, Korean, Russian, and Spanish. We describe the process of creating a novel dataset -- MultiWho -- for fine-grained evaluation of relative clause attachment preferences in ambiguous and unambiguous contexts. Our experiments with three LLMs indicate that, contrary to humans, LLMs consistently exhibit a preference for local attachment, displaying limited responsiveness to syntactic variations or language-specific attachment patterns. Although LLMs performed well in unambiguous cases, they rigidly prioritized world knowledge biases, lacking the flexibility of human language processing. These findings highlight the need for more diverse, pragmatically nuanced multilingual training to improve LLMs' handling of complex structures and human-like comprehension.
Large Language Models (LLMs) are highly proficient in language-based tasks. Their language capabilities have positioned them at the forefront of the future AGI (Artificial General Intelligence) race. However, on closer inspection, Valmeekam et al. (2024); Zecevic et al. (2023); Wu et al. (2024) highlight a significant gap between their language proficiency and reasoning abilities. Reasoning in LLMs and Vision Language Models (VLMs) aims to bridge this gap by enabling these models to think and re-evaluate their actions and responses. Reasoning is an essential capability for complex problem-solving and a necessary step toward establishing trust in Artificial Intelligence (AI). This will make AI suitable for deployment in sensitive domains, such as healthcare, banking, law, defense, security etc. In recent times, with the advent of powerful reasoning models like OpenAI O1 and DeepSeek R1, reasoning endowment has become a critical research topic in LLMs. In this paper, we provide a detailed overview and comparison of existing reasoning techniques and present a systematic survey of reasoning-imbued language models. We also study current challenges and present our findings.
Integrating audio and visual data for training multimodal foundational models remains challenging. We present Audio-Video Vector Alignment (AVVA), which aligns audiovisual (AV) scene content beyond mere temporal synchronization via a Large Language Model (LLM)-based data curation pipeline. Specifically, AVVA scores and selects high-quality training clips using Whisper (speech-based audio foundation model) for audio and DINOv2 for video within a dual-encoder contrastive learning framework. Evaluations on AudioCaps, VALOR, and VGGSound demonstrate that this approach can achieve significant accuracy gains with substantially less curated data. For instance, AVVA yields a 7.6% improvement in top-1 accuracy for audio-to-video retrieval on VGGSound compared to ImageBind, despite training on only 192 hours of carefully filtered data (vs. 5800+ hours). Moreover, an ablation study highlights that trading data quantity for data quality improves performance, yielding respective top-3 accuracy increases of 47.8, 48.4, and 58.0 percentage points on AudioCaps, VALOR, and VGGSound over uncurated baselines. While these results underscore AVVA's data efficiency, we also discuss the overhead of LLM-driven curation and how it may be scaled or approximated in larger domains. Overall, AVVA provides a viable path toward more robust, text-free audiovisual learning with improved retrieval accuracy.
The linear growth of key-value (KV) cache memory and quadratic computational complexity pose significant bottlenecks for large language models (LLMs) in long-context processing. While existing KV cache optimization methods address these challenges through token pruning or feature merging, they often suffer from irreversible information loss or require costly parameter retraining. We propose ZeroMerge, a dynamic zero-shot compression framework that achieves efficient cache management through three key innovations: (1) Fine-grained memory allocation guided by multi-dimensional token importance metrics at head-level granularity, (2) A residual merging mechanism that preserves critical context through compensated attention scoring, and (3) Parameter-free adaptation compatible with diverse LLM architectures without retraining. Comprehensive evaluations across LLaMA-2 model demonstrate that ZeroMerge maintains full-cache performance at 5\% compression ratios while doubling inference throughput at 40K token lengths. The method effectively balances memory efficiency, generation quality, and deployment flexibility, advancing practical long-context LLM applications. The code is available at https://github.com/SusCom-Lab/ZeroMerge.
Our analysis of the NeurIPS 2023 large language model (LLM) fine-tuning competition revealed the following trend: top-performing models exhibit significant overfitting on benchmark datasets, mirroring the broader issue of benchmark overfitting on popular leaderboards and that data curation is essential in order to get a high performing LLM. The competition, which consisted of two stages - an open evaluation stage with publicly available tasks and a closed evaluation stage with unseen tasks - allowed us to assess the generalizability of fine-tuned LLMs. Our results highlight the limitations of current benchmark-based evaluation schemes for generative models and demonstrate the need for more robust evaluation methods. Notably, the winning submissions utilized standard open-source libraries and focused primarily on data curation. To facilitate further research and promote reproducibility, we release all competition entries, Docker files, and evaluation infrastructure, providing a valuable resource for the community to explore fine-tuning, overfitting, and reproducibility in LLMs.
Large language models (LLMs) have shown remarkable performance and generalization capabilities across multiple languages and tasks, making them very attractive targets for multi-modality integration (e.g., images or speech). In this work, we extend an existing LLM to the speech modality via speech discretization and continued pre-training. In particular, we are interested in multilingual LLMs, such as TOWER, as their pre-training setting allows us to treat discretized speech input as an additional translation language. The resulting open-source model, SPIRE, is able to transcribe and translate English speech input while maintaining TOWER's original performance on translation-related tasks, showcasing that discretized speech input integration as an additional language is feasible during LLM adaptation. We make our code and models available to the community.
Discourse understanding is essential for many NLP tasks, yet most existing work remains constrained by framework-dependent discourse representations. This work investigates whether large language models (LLMs) capture discourse knowledge that generalizes across languages and frameworks. We address this question along two dimensions: (1) developing a unified discourse relation label set to facilitate cross-lingual and cross-framework discourse analysis, and (2) probing LLMs to assess whether they encode generalizable discourse abstractions. Using multilingual discourse relation classification as a testbed, we examine a comprehensive set of 23 LLMs of varying sizes and multilingual capabilities. Our results show that LLMs, especially those with multilingual training corpora, can generalize discourse information across languages and frameworks. Further layer-wise analyses reveal that language generalization at the discourse level is most salient in the intermediate layers. Lastly, our error analysis provides an account of challenging relation classes.
LLM self-evaluation relies on the LLM's own ability to estimate response correctness, which can greatly improve its deployment reliability. In this research track, we propose the Chain-of-Embedding (CoE) in the latent space to enable LLMs to perform output-free self-evaluation. CoE consists of all progressive hidden states produced during the inference time, which can be treated as the latent thinking path of LLMs. We find that when LLMs respond correctly and incorrectly, their CoE features differ, these discrepancies assist us in estimating LLM response correctness. Experiments in four diverse domains and seven LLMs fully demonstrate the effectiveness of our method. Meanwhile, its label-free design intent without any training and millisecond-level computational cost ensures real-time feedback in large-scale scenarios. More importantly, we provide interesting insights into LLM response correctness from the perspective of hidden state changes inside LLMs.
Large Language Models (LLMs) are revolutionizing medical diagnostics by enhancing both disease classification and clinical decision-making. In this study, we evaluate the performance of two LLM- based diagnostic tools, DeepSeek R1 and O3 Mini, using a structured dataset of symptoms and diagnoses. We assessed their predictive accuracy at both the disease and category levels, as well as the reliability of their confidence scores. DeepSeek R1 achieved a disease-level accuracy of 76% and an overall accuracy of 82%, outperforming O3 Mini, which attained 72% and 75% respectively. Notably, DeepSeek R1 demonstrated exceptional performance in Mental Health, Neurological Disorders, and Oncology, where it reached 100% accuracy, while O3 Mini excelled in Autoimmune Disease classification with 100% accuracy. Both models, however, struggled with Respiratory Disease classification, recording accuracies of only 40% for DeepSeek R1 and 20% for O3 Mini. Additionally, the analysis of confidence scores revealed that DeepSeek R1 provided high-confidence predictions in 92% of cases, compared to 68% for O3 Mini. Ethical considerations regarding bias, model interpretability, and data privacy are also discussed to ensure the responsible integration of LLMs into clinical practice. Overall, our findings offer valuable insights into the strengths and limitations of LLM-based diagnostic systems and provide a roadmap for future enhancements in AI-driven healthcare.
In recent years, Large Language Models (LLMs) have exhibited remarkable capabilities, driving advancements in real-world applications. However, training LLMs on increasingly long input sequences imposes significant challenges due to high GPU memory and computational demands. Existing solutions face two key limitations: (1) memory reduction techniques, such as activation recomputation and CPU offloading, compromise training efficiency; (2) distributed parallelism strategies require excessive GPU resources, limiting the scalability of input sequence length. To address these gaps, we propose Adaptive Sequence Pipeline Parallel Offloading (SPPO), a novel LLM training framework that optimizes memory and computational resource efficiency for long-sequence training. SPPO introduces adaptive offloading, leveraging sequence-aware offloading, and two-level activation management to reduce GPU memory consumption without degrading the training efficiency. Additionally, SPPO develops an adaptive pipeline scheduling approach with a heuristic solver and multiplexed sequence partitioning to improve computational resource efficiency. Experimental results demonstrate that SPPO achieves up to 3.38x throughput improvement over Megatron-LM and DeepSpeed, realizing efficient training of a 7B LLM with sequence lengths of up to 4M tokens on only 128 A100 GPUs.
Due to the limited computational resources, most Large Language Models (LLMs) developers can only fine-tune Small Language Models (SLMs) on their own data. These private SLMs typically have limited effectiveness. To boost the performance of private SLMs, this paper proposes to ask general LLMs for help. The general LLMs can be APIs or larger LLMs whose inference cost the developers can afford. Specifically, we propose the G-Boost framework where a private SLM adaptively performs collaborative inference with a general LLM under the guide of process reward. Experiments demonstrate that our framework can significantly boost the performance of private SLMs.
Aligning with personalized preferences, which vary significantly across cultural, educational, and political differences, poses a significant challenge due to the computational costs and data demands of traditional alignment methods. In response, this paper presents Personalized Alignment at Decoding-time (PAD), a novel framework designed to align LLM outputs with diverse personalized preferences during the inference phase, eliminating the need for additional training. By introducing a unique personalized reward modeling strategy, this framework decouples the text generation process from personalized preferences, facilitating the generation of generalizable token-level personalized rewards. The PAD algorithm leverages these rewards to guide the decoding process, dynamically tailoring the base model's predictions to personalized preferences. Extensive experimental results demonstrate that PAD not only outperforms existing training-based alignment methods in terms of aligning with diverse preferences but also shows significant generalizability to preferences unseen during training and scalability across different base models. This work advances the capability of LLMs to meet user needs in real-time applications, presenting a substantial step forward in personalized LLM alignment.
Large language models (LLMs) have demonstrated remarkable capabilities in various complex tasks, yet they still suffer from hallucinations. Introducing external knowledge, such as knowledge graph, can enhance the LLMs' ability to provide factual answers. LLMs have the ability to interactively explore knowledge graphs. However, most approaches have been affected by insufficient internal knowledge excavation in LLMs, limited generation of trustworthy knowledge reasoning paths, and a vague integration between internal and external knowledge. Therefore, we propose KnowPath, a knowledge-enhanced large model framework driven by the collaboration of internal and external knowledge. It relies on the internal knowledge of the LLM to guide the exploration of interpretable directed subgraphs in external knowledge graphs, better integrating the two knowledge sources for more accurate reasoning. Extensive experiments on multiple real-world datasets confirm the superiority of KnowPath.
Speculative inference is a promising paradigm employing small speculative models (SSMs) as drafters to generate draft tokens, which are subsequently verified in parallel by the target large language model (LLM). This approach enhances the efficiency of inference serving by reducing LLM inference latency and costs while preserving generation quality. However, existing speculative methods face critical challenges, including inefficient resource utilization and limited draft acceptance, which constrain their scalability and overall effectiveness. To overcome these obstacles, we present CoSine, a novel speculative inference system that decouples sequential speculative decoding from parallel verification, enabling efficient collaboration among multiple nodes. Specifically, CoSine routes inference requests to specialized drafters based on their expertise and incorporates a confidence-based token fusion mechanism to synthesize outputs from cooperating drafters, ensuring high-quality draft generation. Additionally, CoSine dynamically orchestrates the execution of speculative decoding and verification in a pipelined manner, employing batch scheduling to selectively group requests and adaptive speculation control to minimize idle periods. By optimizing parallel workflows through heterogeneous node collaboration, CoSine balances draft generation and verification throughput in real-time, thereby maximizing resource utilization. Experimental results demonstrate that CoSine achieves superior performance compared to state-of-the-art speculative approaches. Notably, with equivalent resource costs, CoSine achieves up to a 23.2% decrease in latency and a 32.5% increase in throughput compared to baseline methods.
This work adapts and studies the gradient-based Membership Inference Test (gMINT) to the classification of text based on LLMs. MINT is a general approach intended to determine if given data was used for training machine learning models, and this work focuses on its application to the domain of Natural Language Processing. Using gradient-based analysis, the MINT model identifies whether particular data samples were included during the language model training phase, addressing growing concerns about data privacy in machine learning. The method was evaluated in seven Transformer-based models and six datasets comprising over 2.5 million sentences, focusing on text classification tasks. Experimental results demonstrate MINTs robustness, achieving AUC scores between 85% and 99%, depending on data size and model architecture. These findings highlight MINTs potential as a scalable and reliable tool for auditing machine learning models, ensuring transparency, safeguarding sensitive data, and fostering ethical compliance in the deployment of AI/NLP technologies.
Visual programming has the potential of providing novice programmers with a low-code experience to build customized processing pipelines. Existing systems typically require users to build pipelines from scratch, implying that novice users are expected to set up and link appropriate nodes from a blank workspace. In this paper, we introduce InstructPipe, an AI assistant for prototyping machine learning (ML) pipelines with text instructions. We contribute two large language model (LLM) modules and a code interpreter as part of our framework. The LLM modules generate pseudocode for a target pipeline, and the interpreter renders the pipeline in the node-graph editor for further human-AI collaboration. Both technical and user evaluation (N=16) shows that InstructPipe empowers users to streamline their ML pipeline workflow, reduce their learning curve, and leverage open-ended commands to spark innovative ideas.
We investigate the choice patterns of Large Language Models (LLMs) in the context of Decisions from Experience tasks that involve repeated choice and learning from feedback, and compare their behavior to human participants. We find that on the aggregate, LLMs appear to display behavioral biases similar to humans: both exhibit underweighting rare events and correlation effects. However, more nuanced analyses of the choice patterns reveal that this happens for very different reasons. LLMs exhibit strong recency biases, unlike humans, who appear to respond in more sophisticated ways. While these different processes may lead to similar behavior on average, choice patterns contingent on recent events differ vastly between the two groups. Specifically, phenomena such as ``surprise triggers change" and the ``wavy recency effect of rare events" are robustly observed in humans, but entirely absent in LLMs. Our findings provide insights into the limitations of using LLMs to simulate and predict humans in learning environments and highlight the need for refined analyses of their behavior when investigating whether they replicate human decision making tendencies.
Accurate identification of software vulnerabilities is crucial for system integrity. Vulnerability datasets, often derived from the National Vulnerability Database (NVD) or directly from GitHub, are essential for training machine learning models to detect these security flaws. However, these datasets frequently suffer from significant noise, typically 40% to 75%, due primarily to the automatic and indiscriminate labeling of all changes in vulnerability-fixing commits (VFCs) as vulnerability-related. This misclassification occurs because not all changes in a commit aimed at fixing vulnerabilities pertain to security threats; many are routine updates like bug fixes or test improvements. This paper introduces the first methodology that uses the Large Language Model (LLM) with a heuristic enhancement to automatically identify vulnerability-fixing changes from VFCs, achieving an F1-score of 0.82. VulSifter was applied to a large-scale study, where we conducted a crawl of 127,063 repositories on GitHub, resulting in the acquisition of 5,352,105 commits. VulSifter involves utilizing an LLM to comprehend code semantics and contextual information, while applying heuristics to filter out unrelated changes. We then developed CleanVul, a high-quality dataset comprising 8,203 functions using our LLM heuristic enhancement approach, demonstrating Correctness (90.6%) comparable to established datasets such as SVEN and PrimeVul. To evaluate the CleanVul dataset, we conducted experiments focusing on fine-tuning various LLMs on CleanVul and other high-quality datasets. Evaluation results reveal that LLMs fine-tuned on CleanVul not only exhibit enhanced accuracy but also superior generalization capabilities compared to those trained on uncleaned datasets. Specifically, models trained on CleanVul and tested on PrimeVul achieve accuracy higher than those trained and tested exclusively on PrimeVul.
Recent advancement of large language models (LLMs) has led to significant breakthroughs across various tasks, laying the foundation for the development of LLM-based speech translation systems. Existing methods primarily focus on aligning inputs and outputs across modalities while overlooking deeper semantic alignment within model representations. To address this limitation, we propose an Adaptive Inner Speech-Text Alignment (AI-STA) method to bridge the modality gap by explicitly aligning speech and text representations at selected layers within LLMs. To achieve this, we leverage the optimal transport (OT) theory to quantify fine-grained representation discrepancies between speech and text. Furthermore, we utilize the cross-modal retrieval technique to identify the layers that are best suited for alignment and perform joint training on these layers. Experimental results on speech translation (ST) tasks demonstrate that AI-STA significantly improves the translation performance of large speech-text models (LSMs), outperforming previous state-of-the-art approaches. Our findings highlight the importance of inner-layer speech-text alignment in LLMs and provide new insights into enhancing cross-modal learning.
Large language models (LLMs) exhibit strong reasoning abilities, often attributed to few-shot or zero-shot chain-of-thought (CoT) prompting. While effective, these methods require labor-intensive prompt engineering, raising the question of whether reasoning can be induced without reliance on explicit prompts. In this work, we unlock the reasoning capabilities of LLMs without explicit prompting. Inspired by zero-shot CoT and CoT-decoding, we propose a novel decoding strategy that systematically nudges LLMs to continue reasoning, thereby preventing immature reasoning processes. Specifically, we monitor the model's generation and inject a designated phrase whenever it is likely to conclude its response prematurely, before completing the reasoning process. Our experimental evaluations on diverse reasoning benchmarks demonstrate that our proposed strategy substantially improves LLM reasoning capabilities, highlighting the potential of decoding-based interventions as an alternative to traditional prompting techniques.
Generating accurate SQL from users' natural language questions (text-to-SQL) remains a long-standing challenge due to the complexities involved in user question understanding, database schema comprehension, and SQL generation. Traditional text-to-SQL systems, which combine human engineering and deep neural networks, have made significant progress. Subsequently, pre-trained language models (PLMs) have been developed for text-to-SQL tasks, achieving promising results. However, as modern databases and user questions grow more complex, PLMs with a limited parameter size often produce incorrect SQL. This necessitates more sophisticated and tailored optimization methods, which restricts the application of PLM-based systems. Recently, large language models (LLMs) have shown significant capabilities in natural language understanding as model scale increases. Thus, integrating LLM-based solutions can bring unique opportunities, improvements, and solutions to text-to-SQL research. In this survey, we provide a comprehensive review of existing LLM-based text-to-SQL studies. Specifically, we offer a brief overview of the technical challenges and evolutionary process of text-to-SQL. Next, we introduce the datasets and metrics designed to evaluate text-to-SQL systems. Subsequently, we present a systematic analysis of recent advances in LLM-based text-to-SQL. Finally, we make a summarization and discuss the remaining challenges in this field and suggest expectations for future research directions.
Large Language Models (LLMs) are increasingly being used to automate programming tasks. Yet, LLMs' capabilities in reasoning about program semantics are still inadequately studied, leaving significant potential for further exploration. This paper introduces FormalBench, a comprehensive benchmark designed to evaluate LLMs' reasoning abilities on program semantics, particularly via the task of synthesizing formal program specifications to assist verifying program correctness. This task requires both comprehensive reasoning over all possible program executions and the generation of precise, syntactically correct expressions that adhere to formal syntax and semantics. Using this benchmark, we evaluated the ability of LLMs in synthesizing consistent and complete specifications. Our findings show that LLMs perform well with simple control flows but struggle with more complex structures, especially loops, even with advanced prompting. Additionally, LLMs exhibit limited robustness against semantic-preserving transformations. We also highlight common failure patterns and design self-repair prompts, improving success rates by 25%.
Smart contract code summarization is crucial for efficient maintenance and vulnerability mitigation. While many studies use Large Language Models (LLMs) for summarization, their performance still falls short compared to fine-tuned models like CodeT5+ and CodeBERT. Some approaches combine LLMs with data flow analysis but fail to fully capture the hierarchy and control structures of the code, leading to information loss and degraded summarization quality. We propose SCLA, an LLM-based method that enhances summarization by integrating a Control Flow Graph (CFG) and semantic facts from the code's control flow into a semantically enriched prompt. SCLA uses a control flow extraction algorithm to derive control flows from semantic nodes in the Abstract Syntax Tree (AST) and constructs the corresponding CFG. Code semantic facts refer to both explicit and implicit information within the AST that is relevant to smart contracts. This method enables LLMs to better capture the structural and contextual dependencies of the code. We validate the effectiveness of SCLA through comprehensive experiments on a dataset of 40,000 real-world smart contracts. The experiment shows that SCLA significantly improves summarization quality, outperforming the SOTA baselines with improvements of 26.7%, 23.2%, 16.7%, and 14.7% in BLEU-4, METEOR, ROUGE-L, and BLEURT scores, respectively.
Despite the remarkable successes of Large Language Models (LLMs), their fundamental Transformer architecture possesses inherent theoretical limitations that restrict their capability to handle reasoning tasks with increasing computational complexity. Chain-of-Thought (CoT) prompting has emerged as a practical solution, supported by several theoretical studies. However, current CoT-based methods (including ToT, GoT, etc.) generally adopt a "one-prompt-fits-all" strategy, using fixed templates (e.g., "think step by step") across diverse reasoning tasks. This method forces models to navigate an extremely complex prompt space to identify effective reasoning paths. The current prompt designing research are also heavily relying on trial-and-error rather than theoretically informed guidance. In this paper, we provide a rigorous theoretical analysis of the complexity and interplay between two crucial spaces: the prompt space (the space of potential prompt structures) and the answer space (the space of reasoning solutions generated by LLMs) in CoT reasoning. We demonstrate how reliance on a single universal prompt (e.g. think step by step) can negatively impact the theoretical computability of LLMs, illustrating that prompt complexity directly influences the structure and effectiveness of the navigation in answer space. Our analysis highlights that sometimes human supervision is critical for efficiently navigating the prompt space. We theoretically and empirically show that task-specific prompting significantly outperforms unsupervised prompt generation, emphasizing the necessity of thoughtful human guidance in CoT prompting.
The combination of LLM agents with external tools enables models to solve complex tasks beyond their knowledge base. Human-designed tools are inflexible and restricted to solutions within the scope of pre-existing tools created by experts. To address this problem, we propose ATLASS, an advanced tool learning and selection system designed as a closed-loop framework. It enables the LLM to solve problems by dynamically generating external tools on demand. In this framework, agents play a crucial role in orchestrating tool selection, execution, and refinement, ensuring adaptive problem-solving capabilities. The operation of ATLASS follows three phases: The first phase, Understanding Tool Requirements, involves the Agents determining whether tools are required and specifying their functionality; the second phase, Tool Retrieval/Generation, involves the Agents retrieving or generating tools based on their availability; and the third phase, Task Solving, involves combining all the component tools necessary to complete the initial task. The Tool Dataset stores the generated tools, ensuring reusability and minimizing inference cost. Current LLM-based tool generation systems have difficulty creating complex tools that need APIs or external packages. In ATLASS, we solve the problem by automatically setting up the environment, fetching relevant API documentation online, and using a Python interpreter to create a reliable, versatile tool that works in a wider range of situations. OpenAI GPT-4.0 is used as the LLM agent, and safety and ethical concerns are handled through human feedback before executing generated code. By addressing the limitations of predefined toolsets and enhancing adaptability, ATLASS serves as a real-world solution that empowers users with dynamically generated tools for complex problem-solving.
Multi-agent systems (MAS) have shown great potential in executing complex tasks, but coordination and safety remain significant challenges. Multi-Agent Reinforcement Learning (MARL) offers a promising framework for agent collaboration, but it faces difficulties in handling complex tasks and designing reward functions. The introduction of Large Language Models (LLMs) has brought stronger reasoning and cognitive abilities to MAS, but existing LLM-based systems struggle to respond quickly and accurately in dynamic environments. To address these challenges, we propose LLM-based Graph Collaboration MARL (LGC-MARL), a framework that efficiently combines LLMs and MARL. This framework decomposes complex tasks into executable subtasks and achieves efficient collaboration among multiple agents through graph-based coordination. Specifically, LGC-MARL consists of two main components: an LLM planner and a graph-based collaboration meta policy. The LLM planner transforms complex task instructions into a series of executable subtasks, evaluates the rationality of these subtasks using a critic model, and generates an action dependency graph. The graph-based collaboration meta policy facilitates communication and collaboration among agents based on the action dependency graph, and adapts to new task environments through meta-learning. Experimental results on the AI2-THOR simulation platform demonstrate the superior performance and scalability of LGC-MARL in completing various complex tasks.
In recent years, the Ethereum platform has witnessed a proliferation of smart contracts, accompanied by exponential growth in total value locked (TVL). High-TVL smart contracts often require complex numerical computations, particularly in mathematical financial models used by many decentralized applications (DApps). Improper calculations can introduce numerical defects, posing potential security risks. Existing research primarily focuses on traditional numerical defects like integer overflow, and there is currently a lack of systematic research and effective detection methods targeting new types of numerical defects. In this paper, we identify five new types of numerical defects through the analysis of 1,199 audit reports by utilizing the open card method. Each defect is defined and illustrated with a code example to highlight its features and potential consequences. We also propose NumScout, a symbolic execution-based tool designed to detect these five defects. Specifically, the tool combines information from source code and bytecode, analyzing key operations such as comparisons and transfers, to effectively locate defects and report them based on predefined detection patterns. Furthermore, NumScout uses a large language model (LLM) to prune functions which are unrelated to numerical operations. This step allows symbolic execution to quickly enter the target function and improve runtime speed by 28.4%. We run NumScout on 6,617 real-world contracts and evaluated its performance based on manually labeled results. We find that 1,774 contracts contained at least one of the five defects, and the tool achieved an overall precision of 89.7%.
This work presents an interpretable decision-making framework for autonomous vehicles that integrates traffic regulations, norms, and safety guidelines comprehensively and enables seamless adaptation to different regions. While traditional rule-based methods struggle to incorporate the full scope of traffic rules, we develop a Traffic Regulation Retrieval (TRR) Agent based on Retrieval-Augmented Generation (RAG) to automatically retrieve relevant traffic rules and guidelines from extensive regulation documents and relevant records based on the ego vehicle's situation. Given the semantic complexity of the retrieved rules, we also design a reasoning module powered by a Large Language Model (LLM) to interpret these rules, differentiate between mandatory rules and safety guidelines, and assess actions on legal compliance and safety. Additionally, the reasoning is designed to be interpretable, enhancing both transparency and reliability. The framework demonstrates robust performance on both hypothesized and real-world cases across diverse scenarios, along with the ability to adapt to different regions with ease.
Integrating Large Language Models (LLMs) in healthcare diagnosis demands systematic frameworks that can handle complex medical scenarios while maintaining specialized expertise. We present KG4Diagnosis, a novel hierarchical multi-agent framework that combines LLMs with automated knowledge graph construction, encompassing 362 common diseases across medical specialties. Our framework mirrors real-world medical systems through a two-tier architecture: a general practitioner (GP) agent for initial assessment and triage, coordinating with specialized agents for in-depth diagnosis in specific domains. The core innovation lies in our end-to-end knowledge graph generation methodology, incorporating: (1) semantic-driven entity and relation extraction optimized for medical terminology, (2) multi-dimensional decision relationship reconstruction from unstructured medical texts, and (3) human-guided reasoning for knowledge expansion. KG4Diagnosis serves as an extensible foundation for specialized medical diagnosis systems, with capabilities to incorporate new diseases and medical knowledge. The framework's modular design enables seamless integration of domain-specific enhancements, making it valuable for developing targeted medical diagnosis systems. We provide architectural guidelines and protocols to facilitate adoption across medical contexts.
Video large language models have achieved remarkable performance in tasks such as video question answering, however, their temporal understanding remains suboptimal. To address this limitation, we curate a dedicated instruction fine-tuning dataset that focuses on enhancing temporal comprehension across five key dimensions. In order to reduce reliance on costly temporal annotations, we introduce a multi-task prompt fine-tuning approach that seamlessly integrates temporal-sensitive tasks into existing instruction datasets without requiring additional annotations. Furthermore, we develop a novel benchmark for temporal-sensitive video understanding that not only fills the gaps in dimension coverage left by existing benchmarks but also rigorously filters out potential shortcuts, ensuring a more accurate evaluation. Extensive experimental results demonstrate that our approach significantly enhances the temporal understanding of video-LLMs while avoiding reliance on shortcuts.
This paper introduces a methodology leveraging Large Language Models (LLMs) for sector-level portfolio allocation through systematic analysis of macroeconomic conditions and market sentiment. Our framework emphasizes top-down sector allocation by processing multiple data streams simultaneously, including policy documents, economic indicators, and sentiment patterns. Empirical results demonstrate superior risk-adjusted returns compared to traditional cross momentum strategies, achieving a Sharpe ratio of 2.51 and portfolio return of 8.79% versus -0.61 and -1.39% respectively. These results suggest that LLM-based systematic macro analysis presents a viable approach for enhancing automated portfolio allocation decisions at the sector level.
Cancer survivors face unique emotional challenges that impact their quality of life. Mobile diary entries-short text entries recording through their phone about their emotional experiences-provide a promising method for tracking these experiences in real time. Although emotion analysis tools show potential for recognizing emotions from text, current methods lack the contextual understanding necessary to accurately interpret the brief, personal narratives in mobile diaries. We propose CALLM, a context-aware emotion analysis framework that leverages Large Language Models (LLMs) with Retrieval-Augmented Generation (RAG), to analyze mobile diary entries from cancer survivors to predict their emotional states. The framework enhances prediction accuracy beyond existing methods by (1) integrating retrieved peer experiences as contextual examples and (2) incorporating individuals' temporal emotional trajectories from their mobile diary entries. We collected a large-scale dataset (N=407) of cancer survivors' mobile ecological momentary assessments (EMAs), which assessed positive and negative affect, desire to regulate emotions, social interaction quality, and availability for interventions, alongside daily mobile diary entries in an open response format regarding what was driving their current emotional experience. Results demonstrate strong performance of CALLM, with balanced accuracies reaching 72.96% for positive and 73.29% for negative affect, and 73.72% for predicting individual's desire to regulate emotions. Post-hoc analysis reveals that leveraging model confidence, encouraging longer diary entries, and incorporating personal ground truth, further enhance predictive outcomes. Our findings support the feasibility of deploying LLM-powered emotion analysis in chronic health populations and suggest promising directions for personalized interventions for cancer survivors.
Recent advances in large language models (LLMs) have led to a growing interest in developing LLM-based agents for automating web tasks. However, these agents often struggle with even simple tasks on real-world websites due to their limited capability to understand and process complex web page structures. In this work, we introduce LCoW, a framework for Learning language models to Contextualize complex Web pages into a more comprehensible form, thereby enhancing decision making by LLM agents. LCoW decouples web page understanding from decision making by training a separate contextualization module to transform complex web pages into comprehensible format, which are then utilized by the decision-making agent. We demonstrate that our contextualization module effectively integrates with LLM agents of various scales to significantly enhance their decision-making capabilities in web automation tasks. Notably, LCoW improves the success rates of closed-source LLMs (e.g., Gemini-1.5-flash, GPT-4o, Claude-3.5-Sonnet) by an average of 15.6%, and demonstrates a 23.7% average improvement in success rates for open-source LMs (e.g., Llama-3.1-8B, Llama-3.1-70B) on the WorkArena benchmark. Moreover, the Gemini-1.5-flash agent with LCoW achieves state-of-the-art results on the WebShop benchmark, outperforming human experts. The relevant code materials are available at our project page: https://lcowiclr2025.github.io.