Recent advances in Large Language Models (LLMs) have propelled intelligent agents from reactive responses to proactive support. While promising, existing proactive agents either rely exclusively on observations from enclosed environments (e.g., desktop UIs) with direct LLM inference or employ rule-based proactive notifications, leading to suboptimal user intent understanding and limited functionality for proactive service. In this paper, we introduce ContextAgent, the first context-aware proactive agent that incorporates extensive sensory contexts to enhance the proactive capabilities of LLM agents. ContextAgent first extracts multi-dimensional contexts from massive sensory perceptions on wearables (e.g., video and audio) to understand user intentions. ContextAgent then leverages the sensory contexts and the persona contexts from historical data to predict the necessity for proactive services. When proactive assistance is needed, ContextAgent further automatically calls the necessary tools to assist users unobtrusively. To evaluate this new task, we curate ContextAgentBench, the first benchmark for evaluating context-aware proactive LLM agents, covering 1,000 samples across nine daily scenarios and twenty tools. Experiments on ContextAgentBench show that ContextAgent outperforms baselines by achieving up to 8.5% and 6.0% higher accuracy in proactive predictions and tool calling, respectively. We hope our research can inspire the development of more advanced, human-centric, proactive AI assistants.
While humans naturally learn and adapt from past experiences, large language models (LLMs) and their agentic counterparts struggle to retain reasoning from previous tasks and apply them in future contexts. To address this limitation, we propose a novel framework, log-augmented generation (LAG) that directly reuses prior computation and reasoning from past logs at test time to enhance model's ability to learn from previous tasks and perform better on new, unseen challenges, all while keeping the system efficient and scalable. Specifically, our system represents task logs using key-value (KV) caches, encoding the full reasoning context of prior tasks while storing KV caches for only a selected subset of tokens. When a new task arises, LAG retrieves the KV values from relevant logs to augment generation. Our approach differs from reflection-based memory mechanisms by directly reusing prior reasoning and computations without requiring additional steps for knowledge extraction or distillation. Our method also goes beyond existing KV caching techniques, which primarily target efficiency gains rather than improving accuracy. Experiments on knowledge- and reasoning-intensive datasets demonstrate that our method significantly outperforms standard agentic systems that do not utilize logs, as well as existing solutions based on reflection and KV cache techniques.
Causal world models are systems that can answer counterfactual questions about an environment of interest, i.e. predict how it would have evolved if an arbitrary subset of events had been realized differently. It requires understanding the underlying causes behind chains of events and conducting causal inference for arbitrary unseen distributions. So far, this task eludes foundation models, notably large language models (LLMs), which do not have demonstrated causal reasoning capabilities beyond the memorization of existing causal relationships. Furthermore, evaluating counterfactuals in real-world applications is challenging since only the factual world is observed, limiting evaluation to synthetic datasets. We address these problems by explicitly extracting and modeling causal relationships and propose the Causal Cartographer framework. First, we introduce a graph retrieval-augmented generation agent tasked to retrieve causal relationships from data. This approach allows us to construct a large network of real-world causal relationships that can serve as a repository of causal knowledge and build real-world counterfactuals. In addition, we create a counterfactual reasoning agent constrained by causal relationships to perform reliable step-by-step causal inference. We show that our approach can extract causal knowledge and improve the robustness of LLMs for causal reasoning tasks while reducing inference costs and spurious correlations.
Task-oriented dialogue systems based on Large Language Models (LLMs) have gained increasing attention across various industries and achieved significant results. Current approaches condense complex procedural workflows into a single agent to achieve satisfactory performance on large-scale LLMs. However, these approaches face challenges to achieve comparable performance on fine-tuned lightweight LLMs, due to their limited capabilities in handling multiple complex logic. In this work, we design a Domain-Independent Multi-Agent Framework (DIMF), which contains Intent Classification Agent, Slot Filling Agent and Response Agent. This approach simplifies the learning complexity and enhances the generalization ability by separating the tasks into domain-independent components. In this framework, we enhance the capabilities in contextual understanding using the Direct Preference Optimisation (DPO) method, and propose a simple and effective Data Distribution Adaptation (DDA) method to mitigate degradation issues during DPO training. Experiments conducted on the MultiWOZ datasets show that our proposed method achieves a better average performance among all the baselines. Extensive analysis also demonstrates that our proposed framework exhibits excellent generalizability and zero-shot capability.
As retrieval-augmented AI agents become more embedded in society, their safety properties and ethical behavior remain insufficiently understood. In particular, the growing integration of LLMs and AI agents raises critical questions about how they engage with and are influenced by their environments. This study investigates how expanding retrieval access, from no external sources to Wikipedia-based retrieval and open web search, affects model reliability, bias propagation, and harmful content generation. Through extensive benchmarking of censored and uncensored LLMs and AI Agents, our findings reveal a consistent degradation in refusal rates, bias sensitivity, and harmfulness safeguards as models gain broader access to external sources, culminating in a phenomenon we term safety devolution. Notably, retrieval-augmented agents built on aligned LLMs often behave more unsafely than uncensored models without retrieval. This effect persists even under strong retrieval accuracy and prompt-based mitigation, suggesting that the mere presence of retrieved content reshapes model behavior in structurally unsafe ways. These findings underscore the need for robust mitigation strategies to ensure fairness and reliability in retrieval-augmented and increasingly autonomous AI systems.
Large language model (LLM) agents have shown promising performance in generating code for solving complex data science problems. Recent studies primarily focus on enhancing in-context learning through improved search, sampling, and planning techniques, while overlooking the importance of the order in which problems are tackled during inference. In this work, we develop a novel inference-time optimization framework, referred to as DSMentor, which leverages curriculum learning -- a strategy that introduces simpler task first and progressively moves to more complex ones as the learner improves -- to enhance LLM agent performance in challenging data science tasks. Our mentor-guided framework organizes data science tasks in order of increasing difficulty and incorporates a growing long-term memory to retain prior experiences, guiding the agent's learning progression and enabling more effective utilization of accumulated knowledge. We evaluate DSMentor through extensive experiments on DSEval and QRData benchmarks. Experiments show that DSMentor using Claude-3.5-Sonnet improves the pass rate by up to 5.2% on DSEval and QRData compared to baseline agents. Furthermore, DSMentor demonstrates stronger causal reasoning ability, improving the pass rate by 8.8% on the causality problems compared to GPT-4 using Program-of-Thoughts prompts. Our work underscores the importance of developing effective strategies for accumulating and utilizing knowledge during inference, mirroring the human learning process and opening new avenues for improving LLM performance through curriculum-based inference optimization.
Mathematical modeling is a cornerstone of scientific discovery and engineering practice, enabling the translation of real-world problems into formal systems across domains such as physics, biology, and economics. Unlike mathematical reasoning, which assumes a predefined formulation, modeling requires open-ended problem analysis, abstraction, and principled formalization. While Large Language Models (LLMs) have shown strong reasoning capabilities, they fall short in rigorous model construction, limiting their utility in real-world problem-solving. To this end, we formalize the task of LLM-powered real-world mathematical modeling, where agents must analyze problems, construct domain-appropriate formulations, and generate complete end-to-end solutions. We introduce MM-Bench, a curated benchmark of 111 problems from the Mathematical Contest in Modeling (MCM/ICM), spanning the years 2000 to 2025 and across ten diverse domains such as physics, biology, and economics. To tackle this task, we propose MM-Agent, an expert-inspired framework that decomposes mathematical modeling into four stages: open-ended problem analysis, structured model formulation, computational problem solving, and report generation. Experiments on MM-Bench show that MM-Agent significantly outperforms baseline agents, achieving an 11.88\% improvement over human expert solutions while requiring only 15 minutes and \$0.88 per task using GPT-4o. Furthermore, under official MCM/ICM protocols, MM-Agent assisted two undergraduate teams in winning the Finalist Award (\textbf{top 2.0\% among 27,456 teams}) in MCM/ICM 2025, demonstrating its practical effectiveness as a modeling copilot. Our code is available at https://github.com/usail-hkust/LLM-MM-Agent
Retrieval-augmented generation (RAG) systems empower large language models (LLMs) to access external knowledge during inference. Recent advances have enabled LLMs to act as search agents via reinforcement learning (RL), improving information acquisition through multi-turn interactions with retrieval engines. However, existing approaches either optimize retrieval using search-only metrics (e.g., NDCG) that ignore downstream utility or fine-tune the entire LLM to jointly reason and retrieve-entangling retrieval with generation and limiting the real search utility and compatibility with frozen or proprietary models. In this work, we propose s3, a lightweight, model-agnostic framework that decouples the searcher from the generator and trains the searcher using a Gain Beyond RAG reward: the improvement in generation accuracy over naive RAG. s3 requires only 2.4k training samples to outperform baselines trained on over 70x more data, consistently delivering stronger downstream performance across six general QA and five medical QA benchmarks.
Mobile GUI agents execute user commands by directly interacting with the graphical user interface (GUI) of mobile devices, demonstrating significant potential to enhance user convenience. However, these agents face considerable challenges in task planning, as they must continuously analyze the GUI and generate operation instructions step by step. This process often leads to difficulties in making accurate task plans, as GUI agents lack a deep understanding of how to effectively use the target applications, which can cause them to become "lost" during task execution. To address the task planning issue, we propose SPlanner, a plug-and-play planning module to generate execution plans that guide vision language model(VLMs) in executing tasks. The proposed planning module utilizes extended finite state machines (EFSMs) to model the control logits and configurations of mobile applications. It then decomposes a user instruction into a sequence of primary function modeled in EFSMs, and generate the execution path by traversing the EFSMs. We further refine the execution path into a natural language plan using an LLM. The final plan is concise and actionable, and effectively guides VLMs to generate interactive GUI actions to accomplish user tasks. SPlanner demonstrates strong performance on dynamic benchmarks reflecting real-world mobile usage. On the AndroidWorld benchmark, SPlanner achieves a 63.8% task success rate when paired with Qwen2.5-VL-72B as the VLM executor, yielding a 28.8 percentage point improvement compared to using Qwen2.5-VL-72B without planning assistance.
Knowledge Base Question Answering (KBQA) aims to answer natural language questions using structured knowledge from KBs. While LLM-only approaches offer generalization, they suffer from outdated knowledge, hallucinations, and lack of transparency. Chain-based KG-RAG methods address these issues by incorporating external KBs, but are limited to simple chain-structured questions due to the absence of planning and logical structuring. Inspired by semantic parsing methods, we propose PDRR: a four-stage framework consisting of Predict, Decompose, Retrieve, and Reason. Our method first predicts the question type and decomposes the question into structured triples. Then retrieves relevant information from KBs and guides the LLM as an agent to reason over and complete the decomposed triples. Experimental results demonstrate that PDRR consistently outperforms existing methods across various LLM backbones and achieves superior performance on both chain-structured and non-chain complex questions.
Large language model (LLM) based agents have shown great potential in following human instructions and automatically completing various tasks. To complete a task, the agent needs to decompose it into easily executed steps by planning. Existing studies mainly conduct the planning by inferring what steps should be executed next starting from the agent's initial state. However, this forward reasoning paradigm doesn't work well for complex tasks. We propose to study this issue in Minecraft, a virtual environment that simulates complex tasks based on real-world scenarios. We believe that the failure of forward reasoning is caused by the big perception gap between the agent's initial state and task goal. To this end, we leverage backward reasoning and make the planning starting from the terminal state, which can directly achieve the task goal in one step. Specifically, we design a BAckward Reasoning based agent (BAR). It is equipped with a recursive goal decomposition module, a state consistency maintaining module and a stage memory module to make robust, consistent, and efficient planning starting from the terminal state. Experimental results demonstrate the superiority of BAR over existing methods and the effectiveness of proposed modules.
Retrieval-augmented generation (RAG) enhances the text generation capabilities of large language models (LLMs) by integrating external knowledge and up-to-date information. However, traditional RAG systems are limited by static workflows and lack the adaptability required for multistep reasoning and complex task management. To address these limitations, agentic RAG systems (e.g., DeepResearch) have been proposed, enabling dynamic retrieval strategies, iterative context refinement, and adaptive workflows for handling complex search queries beyond the capabilities of conventional RAG. Recent advances, such as Search-R1, have demonstrated promising gains using outcome-based reinforcement learning, where the correctness of the final answer serves as the reward signal. Nevertheless, such outcome-supervised agentic RAG methods face challenges including low exploration efficiency, gradient conflict, and sparse reward signals. To overcome these challenges, we propose to utilize fine-grained, process-level rewards to improve training stability, reduce computational costs, and enhance efficiency. Specifically, we introduce a novel method ReasonRAG that automatically constructs RAG-ProGuide, a high-quality dataset providing process-level rewards for (i) query generation, (ii) evidence extraction, and (iii) answer generation, thereby enhancing model inherent capabilities via process-supervised reinforcement learning. With the process-level policy optimization, the proposed framework empowers LLMs to autonomously invoke search, generate queries, extract relevant evidence, and produce final answers. Compared to existing approaches such as Search-R1 and traditional RAG systems, ReasonRAG, leveraging RAG-ProGuide, achieves superior performance on five benchmark datasets using only 5k training instances, significantly fewer than the 90k training instances required by Search-R1.
Retrieval-Augmented Generation (RAG) systems empower large language models (LLMs) with external knowledge, yet struggle with efficiency-accuracy trade-offs when scaling to large knowledge graphs. Existing approaches often rely on monolithic graph retrieval, incurring unnecessary latency for simple queries and fragmented reasoning for complex multi-hop questions. To address these challenges, this paper propose SPLIT-RAG, a multi-agent RAG framework that addresses these limitations with question-driven semantic graph partitioning and collaborative subgraph retrieval. The innovative framework first create Semantic Partitioning of Linked Information, then use the Type-Specialized knowledge base to achieve Multi-Agent RAG. The attribute-aware graph segmentation manages to divide knowledge graphs into semantically coherent subgraphs, ensuring subgraphs align with different query types, while lightweight LLM agents are assigned to partitioned subgraphs, and only relevant partitions are activated during retrieval, thus reduce search space while enhancing efficiency. Finally, a hierarchical merging module resolves inconsistencies across subgraph-derived answers through logical verifications. Extensive experimental validation demonstrates considerable improvements compared to existing approaches.
Existing AutoML systems have advanced the automation of machine learning (ML); however, they still require substantial manual configuration and expert input, particularly when handling multimodal data. We introduce MLZero, a novel multi-agent framework powered by Large Language Models (LLMs) that enables end-to-end ML automation across diverse data modalities with minimal human intervention. A cognitive perception module is first employed, transforming raw multimodal inputs into perceptual context that effectively guides the subsequent workflow. To address key limitations of LLMs, such as hallucinated code generation and outdated API knowledge, we enhance the iterative code generation process with semantic and episodic memory. MLZero demonstrates superior performance on MLE-Bench Lite, outperforming all competitors in both success rate and solution quality, securing six gold medals. Additionally, when evaluated on our Multimodal AutoML Agent Benchmark, which includes 25 more challenging tasks spanning diverse data modalities, MLZero outperforms the competing methods by a large margin with a success rate of 0.92 (+263.6\%) and an average rank of 2.28. Our approach maintains its robust effectiveness even with a compact 8B LLM, outperforming full-size systems from existing solutions.
In the field of AI4Science, large-scale language models (LLMs) show great potential to parse complex scientific semantics, integrate cross-disciplinary knowledge, and assist critical task research. However, in the field of drug discovery, despite the optimization through professional data pre-training, context window expansion, and internet search, the existing LLMs are still facing challenges such as massive multi-modal and heterogeneous data processing, domain knowledge dynamic updating delay, and insufficient confidence in predicting the results of complex computational tasks. To address these challenges, we propose the DrugPilot, an LLM-based agent with parameterized reasoning for drug discovery. DrugPilot addresses key limitations of traditional end-to-end LLM prediction approaches through its parametric inference architecture. This agent system supports major phases of the drug discovery pipeline, facilitating automated planning and execution of multi-stage research tasks. To address the critical challenge of multi-modal drug data analysis (incorporating both public datasets and user-submitted data), we developed an interactive parameterized memory pool. This innovative component standardizes real-world drug data into parametric representations, simultaneously enabling efficient knowledge retrieval in multi-turn dialogue while mitigating the information loss inherent in text-based data transmission. Additionally, we created a drug instruct dataset across 8 essential drug discovery tasks for model fine-tuning and evaluation. Based on the Berkeley function calling evaluation framework, DrugPilot demonstrated the most advanced tool calling capabilities on our drug discovery tool instruction dataset, outperforming existing agents (e.g., ReAct, LoT). Specifically, it achieves task completion rates of 98.0%, 93.5%, and 64.0% on simple, multiple, and multi-turn tasks, respectively.
We introduce ${\rm C{\small LEVER}}$, a high-quality, curated benchmark of 161 problems for end-to-end verified code generation in Lean. Each problem consists of (1) the task of generating a specification that matches a held-out ground-truth specification, and (2) the task of generating a Lean implementation that provably satisfies this specification. Unlike prior benchmarks, ${\rm C{\small LEVER}}$ avoids test-case supervision, LLM-generated annotations, and specifications that leak implementation logic or allow vacuous solutions. All outputs are verified post-hoc using Lean's type checker to ensure machine-checkable correctness. We use ${\rm C{\small LEVER}}$ to evaluate several few-shot and agentic approaches based on state-of-the-art language models. These methods all struggle to achieve full verification, establishing it as a challenging frontier benchmark for program synthesis and formal reasoning. Our benchmark can be found on GitHub(https://github.com/trishullab/clever) as well as HuggingFace(https://huggingface.co/datasets/amitayusht/clever). All our evaluation code is also available online(https://github.com/trishullab/clever-prover).
Large language models (LLMs) have achieved remarkable capabilities but remain vulnerable to adversarial prompts known as jailbreaks, which can bypass safety alignment and elicit harmful outputs. Despite growing efforts in LLM safety research, existing evaluations are often fragmented, focused on isolated attack or defense techniques, and lack systematic, reproducible analysis. In this work, we introduce PandaGuard, a unified and modular framework that models LLM jailbreak safety as a multi-agent system comprising attackers, defenders, and judges. Our framework implements 19 attack methods and 12 defense mechanisms, along with multiple judgment strategies, all within a flexible plugin architecture supporting diverse LLM interfaces, multiple interaction modes, and configuration-driven experimentation that enhances reproducibility and practical deployment. Built on this framework, we develop PandaBench, a comprehensive benchmark that evaluates the interactions between these attack/defense methods across 49 LLMs and various judgment approaches, requiring over 3 billion tokens to execute. Our extensive evaluation reveals key insights into model vulnerabilities, defense cost-performance trade-offs, and judge consistency. We find that no single defense is optimal across all dimensions and that judge disagreement introduces nontrivial variance in safety assessments. We release the code, configurations, and evaluation results to support transparent and reproducible research in LLM safety.
Large language models (LLMs) exhibit strong capabilities as decision-making agents by interleaving reasoning and actions, as seen in ReAct-style frameworks. Yet, their practical deployment is constrained by high inference costs and large model sizes. We propose Structured Agent Distillation, a framework that compresses large LLM-based agents into smaller student models while preserving both reasoning fidelity and action consistency. Unlike standard token-level distillation, our method segments trajectories into {[REASON]} and {[ACT]} spans, applying segment-specific losses to align each component with the teacher's behavior. This structure-aware supervision enables compact agents to better replicate the teacher's decision process. Experiments on ALFWorld, HotPotQA-ReAct, and WebShop show that our approach consistently outperforms token-level and imitation learning baselines, achieving significant compression with minimal performance drop. Scaling and ablation results further highlight the importance of span-level alignment for efficient and deployable agents.
Metaheuristic algorithms are widely used for solving complex optimization problems, yet their effectiveness is often constrained by fixed structures and the need for extensive tuning. The Polymorphic Metaheuristic Framework (PMF) addresses this limitation by introducing a self-adaptive metaheuristic switching mechanism driven by real-time performance feedback and dynamic algorithmic selection. PMF leverages the Polymorphic Metaheuristic Agent (PMA) and the Polymorphic Metaheuristic Selection Agent (PMSA) to dynamically select and transition between metaheuristic algorithms based on key performance indicators, ensuring continuous adaptation. This approach enhances convergence speed, adaptability, and solution quality, outperforming traditional metaheuristics in high-dimensional, dynamic, and multimodal environments. Experimental results on benchmark functions demonstrate that PMF significantly improves optimization efficiency by mitigating stagnation and balancing exploration-exploitation strategies across various problem landscapes. By integrating AI-driven decision-making and self-correcting mechanisms, PMF paves the way for scalable, intelligent, and autonomous optimization frameworks, with promising applications in engineering, logistics, and complex decision-making systems.
Simulations, although powerful in accurately replicating real-world systems, often remain inaccessible to non-technical users due to their complexity. Conversely, large language models (LLMs) provide intuitive, language-based interactions but can lack the structured, causal understanding required to reliably model complex real-world dynamics. We introduce our simulation agent framework, a novel approach that integrates the strengths of both simulation models and LLMs. This framework helps empower users by leveraging the conversational capabilities of LLMs to interact seamlessly with sophisticated simulation systems, while simultaneously utilizing the simulations to ground the LLMs in accurate and structured representations of real-world phenomena. This integrated approach helps provide a robust and generalizable foundation for empirical validation and offers broad applicability across diverse domains.
Large language models (LLMs) have recently achieved remarkable results in complex multi-step tasks, such as mathematical reasoning and agentic software engineering. However, they often struggle to maintain consistent performance across multiple solution attempts. One effective approach to narrow the gap between average-case and best-case performance is guided test-time search, which explores multiple solution paths to identify the most promising one. Unfortunately, effective search techniques (e.g. MCTS) are often unsuitable for non-serializable RL environments, such as Docker containers, where intermediate environment states cannot be easily saved and restored. We investigate two complementary search strategies applicable to such environments: 1-step lookahead and trajectory selection, both guided by a learned action-value function estimator. On the SWE-bench Verified benchmark, a key testbed for agentic software engineering, we find these methods to double the average success rate of a fine-tuned Qwen-72B model, achieving 40.8%, the new state-of-the-art for open-weights models. Additionally, we show that these techniques are transferable to more advanced closed models, yielding similar improvements with GPT-4o.
Large Language Models (LLMs) have demonstrated strong generative capabilities but remain prone to inconsistencies and hallucinations. We introduce Peer Elicitation Games (PEG), a training-free, game-theoretic framework for aligning LLMs through a peer elicitation mechanism involving a generator and multiple discriminators instantiated from distinct base models. Discriminators interact in a peer evaluation setting, where rewards are computed using a determinant-based mutual information score that provably incentivizes truthful reporting without requiring ground-truth labels. We establish theoretical guarantees showing that each agent, via online learning, achieves sublinear regret in the sense their cumulative performance approaches that of the best fixed truthful strategy in hindsight. Moreover, we prove last-iterate convergence to a truthful Nash equilibrium, ensuring that the actual policies used by agents converge to stable and truthful behavior over time. Empirical evaluations across multiple benchmarks demonstrate significant improvements in factual accuracy. These results position PEG as a practical approach for eliciting truthful behavior from LLMs without supervision or fine-tuning.
Existing benchmarks that assess Language Models (LMs) as Language Agents (LAs) for tool use primarily focus on stateless, single-turn interactions or partial evaluations, such as tool selection in a single turn, overlooking the inherent stateful nature of interactions in multi-turn applications. To fulfill this gap, we propose \texttt{DialogTool}, a multi-turn dialogue dataset with stateful tool interactions considering the whole life cycle of tool use, across six key tasks in three stages: 1) \textit{tool creation}; 2) \textit{tool utilization}: tool awareness, tool selection, tool execution; and 3) \textit{role-consistent response}: response generation and role play. Furthermore, we build \texttt{VirtualMobile} -- an embodied virtual mobile evaluation environment to simulate API calls and assess the robustness of the created APIs\footnote{We will use tools and APIs alternatively, there are no significant differences between them in this paper.}. Taking advantage of these artifacts, we conduct comprehensive evaluation on 13 distinct open- and closed-source LLMs and provide detailed analysis at each stage, revealing that the existing state-of-the-art LLMs still cannot perform well to use tools over long horizons.
We introduce TimeSeriesGym, a scalable benchmarking framework for evaluating Artificial Intelligence (AI) agents on time series machine learning engineering challenges. Existing benchmarks lack scalability, focus narrowly on model building in well-defined settings, and evaluate only a limited set of research artifacts (e.g., CSV submission files). To make AI agent benchmarking more relevant to the practice of machine learning engineering, our framework scales along two critical dimensions. First, recognizing that effective ML engineering requires a range of diverse skills, TimeSeriesGym incorporates challenges from diverse sources spanning multiple domains and tasks. We design challenges to evaluate both isolated capabilities (including data handling, understanding research repositories, and code translation) and their combinations, and rather than addressing each challenge independently, we develop tools that support designing multiple challenges at scale. Second, we implement evaluation mechanisms for multiple research artifacts, including submission files, code, and models, using both precise numeric measures and more flexible LLM-based evaluation approaches. This dual strategy balances objective assessment with contextual judgment. Although our initial focus is on time series applications, our framework can be readily extended to other data modalities, broadly enhancing the comprehensiveness and practical utility of agentic AI evaluation. We open-source our benchmarking framework to facilitate future research on the ML engineering capabilities of AI agents.
Modular construction, involving off-site prefabrication and on-site assembly, offers significant advantages but presents complex coordination challenges for robotic automation. Effective task allocation is critical for leveraging multi-agent systems (MAS) in these structured environments. This paper introduces the Hybrid Voting-Based Task Assignment (HVBTA) framework, a novel approach to optimizing collaboration between heterogeneous multi-agent construction teams. Inspired by human reasoning in task delegation, HVBTA uniquely integrates multiple voting mechanisms with the capabilities of a Large Language Model (LLM) for nuanced suitability assessment between agent capabilities and task requirements. The framework operates by assigning Capability Profiles to agents and detailed requirement lists called Task Descriptions to construction tasks, subsequently generating a quantitative Suitability Matrix. Six distinct voting methods, augmented by a pre-trained LLM, analyze this matrix to robustly identify the optimal agent for each task. Conflict-Based Search (CBS) is integrated for decentralized, collision-free path planning, ensuring efficient and safe spatio-temporal coordination of the robotic team during assembly operations. HVBTA enables efficient, conflict-free assignment and coordination, facilitating potentially faster and more accurate modular assembly. Current work is evaluating HVBTA's performance across various simulated construction scenarios involving diverse robotic platforms and task complexities. While designed as a generalizable framework for any domain with clearly definable tasks and capabilities, HVBTA will be particularly effective for addressing the demanding coordination requirements of multi-agent collaborative robotics in modular construction due to the predetermined construction planning involved.
Large Language Models (LLMs) are catalyzing a paradigm shift in scientific discovery, evolving from task-specific automation tools into increasingly autonomous agents and fundamentally redefining research processes and human-AI collaboration. This survey systematically charts this burgeoning field, placing a central focus on the changing roles and escalating capabilities of LLMs in science. Through the lens of the scientific method, we introduce a foundational three-level taxonomy-Tool, Analyst, and Scientist-to delineate their escalating autonomy and evolving responsibilities within the research lifecycle. We further identify pivotal challenges and future research trajectories such as robotic automation, self-improvement, and ethical governance. Overall, this survey provides a conceptual architecture and strategic foresight to navigate and shape the future of AI-driven scientific discovery, fostering both rapid innovation and responsible advancement. Github Repository: https://github.com/HKUST-KnowComp/Awesome-LLM-Scientific-Discovery.
Retrieval-Augmented Generation (RAG) has significantly improved the performance of large language models (LLMs) on knowledge-intensive domains. However, although RAG achieved successes across distinct domains, there are still some unsolved challenges: 1) Effectiveness. Existing research mainly focuses on developing more powerful RAG retrievers, but how to enhance the generator's (LLM's) ability to utilize the retrieved information for reasoning and generation? 2) Transparency. Most RAG methods ignore which retrieved content actually contributes to the reasoning process, resulting in a lack of interpretability and visibility. To address this, we propose ARENA (Adaptive-Rewarded Evidence Navigation Agent), a transparent RAG generator framework trained via reinforcement learning (RL) with our proposed rewards. Based on the structured generation and adaptive reward calculation, our RL-based training enables the model to identify key evidence, perform structured reasoning, and generate answers with interpretable decision traces. Applied to Qwen2.5-7B-Instruct and Llama3.1-8B-Instruct, abundant experiments with various RAG baselines demonstrate that our model achieves 10-30% improvements on all multi-hop QA datasets, which is comparable with the SOTA Commercially-developed LLMs (e.g., OpenAI-o1, DeepSeek-R1). Further analyses show that ARENA has strong flexibility to be adopted on new datasets without extra training. Our models and codes are publicly released.
The exponential growth of scientific literature presents significant challenges for researchers navigating the complex knowledge landscape. We propose "Agentic Publications", a novel LLM-driven framework complementing traditional publishing by transforming papers into interactive knowledge systems. Our architecture integrates structured data with unstructured content through retrieval-augmented generation and multi-agent verification. The framework offers interfaces for both humans and machines, combining narrative explanations with machine-readable outputs while addressing ethical considerations through automated validation and transparent governance. Key features include continuous knowledge updates, automatic integration of new findings, and customizable detail levels. Our proof-of-concept demonstrates multilingual interaction, API accessibility, and structured knowledge representation through vector databases, knowledge graphs, and verification agents. This approach enhances scientific communication across disciplines, improving efficiency and collaboration while preserving traditional publishing pathways, particularly valuable for interdisciplinary fields where knowledge integration remains challenging.
As Large Language Models (LLMs) become increasingly integrated into real-world decision-making systems, understanding their behavioural vulnerabilities remains a critical challenge for AI safety and alignment. While existing evaluation metrics focus primarily on reasoning accuracy or factual correctness, they often overlook whether LLMs are robust to adversarial manipulation or capable of using adaptive strategy in dynamic environments. This paper introduces an adversarial evaluation framework designed to systematically stress-test the decision-making processes of LLMs under interactive and adversarial conditions. Drawing on methodologies from cognitive psychology and game theory, our framework probes how models respond in two canonical tasks: the two-armed bandit task and the Multi-Round Trust Task. These tasks capture key aspects of exploration-exploitation trade-offs, social cooperation, and strategic flexibility. We apply this framework to several state-of-the-art LLMs, including GPT-3.5, GPT-4, Gemini-1.5, and DeepSeek-V3, revealing model-specific susceptibilities to manipulation and rigidity in strategy adaptation. Our findings highlight distinct behavioral patterns across models and emphasize the importance of adaptability and fairness recognition for trustworthy AI deployment. Rather than offering a performance benchmark, this work proposes a methodology for diagnosing decision-making weaknesses in LLM-based agents, providing actionable insights for alignment and safety research.
The rapid advancement of bug-finding techniques has led to the discovery of more vulnerabilities than developers can reasonably fix, creating an urgent need for effective Automated Program Repair (APR) methods. However, the complexity of modern bugs often makes precise root cause analysis difficult and unreliable. To address this challenge, we propose crash-site repair to simplify the repair task while still mitigating the risk of exploitation. In addition, we introduce a template-guided patch generation approach that significantly reduces the token cost of Large Language Models (LLMs) while maintaining both efficiency and effectiveness. We implement our prototype system, WILLIAMT, and evaluate it against state-of-the-art APR tools. Our results show that, when combined with the top-performing agent CodeRover-S, WILLIAMT reduces token cost by 45.9% and increases the bug-fixing rate to 73.5% (+29.6%) on ARVO, a ground-truth open source software vulnerabilities benchmark. Furthermore, we demonstrate that WILLIAMT can function effectively even without access to frontier LLMs: even a local model running on a Mac M4 Mini achieves a reasonable repair rate. These findings highlight the broad applicability and scalability of WILLIAMT.