A large amount of work has been done in Multi-Agent Systems (MAS) for modeling and solving problems with multiple interacting agents. However, most LLMs are pretrained independently and not specifically optimized for coordination. Existing LLM fine-tuning frameworks rely on individual rewards, which require complex reward designs for each agent to encourage collaboration. To address these challenges, we model LLM collaboration as a cooperative Multi-Agent Reinforcement Learning (MARL) problem. We develop a multi-agent, multi-turn algorithm, Multi-Agent Group Relative Policy Optimization (MAGRPO), to solve it, building on current RL approaches for LLMs as well as MARL techniques. Our experiments on LLM writing and coding collaboration demonstrate that fine-tuning MAS with MAGRPO enables agents to generate high-quality responses efficiently through effective cooperation. Our approach opens the door to using other MARL methods for LLMs and highlights the associated challenges.
Simulating how team members collaborate within complex environments using Agentic AI is a promising approach to explore hypotheses grounded in social science theories and study team behaviors. We introduce VirtLab, a user-friendly, customizable, multi-agent, and scalable team simulation system that enables testing teams with LLM-based agents in spatial and temporal settings. This system addresses the current frameworks' design and technical limitations that do not consider flexible simulation scenarios and spatial settings. VirtLab contains a simulation engine and a web interface that enables both technical and non-technical users to formulate, run, and analyze team simulations without programming. We demonstrate the system's utility by comparing ground truth data with simulated scenarios.
The advent of Large Language Models (LLMs) is transforming search engines into conversational AI search products, primarily using Retrieval-Augmented Generation (RAG) on web corpora. However, this paradigm has significant industrial limitations. Traditional RAG approaches struggle with real-time needs and structured queries that require accessing dynamically generated content like ticket availability or inventory. Limited to indexing static pages, search engines cannot perform the interactive queries needed for such time-sensitive data. Academic research has focused on optimizing RAG for static content, overlooking complex intents and the need for dynamic sources like databases and real-time APIs. To bridge this gap, we introduce TURA (Tool-Augmented Unified Retrieval Agent for AI Search), a novel three-stage framework that combines RAG with agentic tool-use to access both static content and dynamic, real-time information. TURA has three key components: an Intent-Aware Retrieval module to decompose queries and retrieve information sources encapsulated as Model Context Protocol (MCP) Servers, a DAG-based Task Planner that models task dependencies as a Directed Acyclic Graph (DAG) for optimal parallel execution, and a lightweight Distilled Agent Executor for efficient tool calling. TURA is the first architecture to systematically bridge the gap between static RAG and dynamic information sources for a world-class AI search product. Serving tens of millions of users, it leverages an agentic framework to deliver robust, real-time answers while meeting the low-latency demands of a large-scale industrial system.
While LLMs exhibit impressive fluency and factual recall, they struggle with robust causal reasoning, often relying on spurious correlations and brittle patterns. Similarly, traditional Reinforcement Learning agents also lack causal understanding, optimizing for rewards without modeling why actions lead to outcomes. We introduce Causal Reflection, a framework that explicitly models causality as a dynamic function over state, action, time, and perturbation, enabling agents to reason about delayed and nonlinear effects. Additionally, we define a formal Reflect mechanism that identifies mismatches between predicted and observed outcomes and generates causal hypotheses to revise the agent's internal model. In this architecture, LLMs serve not as black-box reasoners, but as structured inference engines translating formal causal outputs into natural language explanations and counterfactuals. Our framework lays the theoretical groundwork for Causal Reflective agents that can adapt, self-correct, and communicate causal understanding in evolving environments.
The dream to create AI assistants as capable and versatile as the fictional J.A.R.V.I.S from Iron Man has long captivated imaginations. With the evolution of (multi-modal) large language models ((M)LLMs), this dream is closer to reality, as (M)LLM-based Agents using computing devices (e.g., computers and mobile phones) by operating within the environments and interfaces (e.g., Graphical User Interface (GUI)) provided by operating systems (OS) to automate tasks have significantly advanced. This paper presents a comprehensive survey of these advanced agents, designated as OS Agents. We begin by elucidating the fundamentals of OS Agents, exploring their key components including the environment, observation space, and action space, and outlining essential capabilities such as understanding, planning, and grounding. We then examine methodologies for constructing OS Agents, focusing on domain-specific foundation models and agent frameworks. A detailed review of evaluation protocols and benchmarks highlights how OS Agents are assessed across diverse tasks. Finally, we discuss current challenges and identify promising directions for future research, including safety and privacy, personalization and self-evolution. This survey aims to consolidate the state of OS Agents research, providing insights to guide both academic inquiry and industrial development. An open-source GitHub repository is maintained as a dynamic resource to foster further innovation in this field. We present a 9-page version of our work, accepted by ACL 2025, to provide a concise overview to the domain.
Recent advances in large language models (LLMs) have unlocked powerful reasoning and decision-making capabilities. However, their inherent dependence on static parametric memory fundamentally limits their adaptability, factual accuracy, and interpretability in knowledge-intensive scenarios. Knowledge graphs (KGs), as structured repositories of explicit relational knowledge, offer a promising approach for augmenting LLMs with external, interpretable memory. Nevertheless, most existing methods that combine LLMs with KGs treat reasoning and knowledge updating as separate processes, resulting in suboptimal utilization of new information and hindering real-time updates. In this work, we propose TRAIL: a novel, unified framework for Thinking, Reasoning, And Incremental Learning that couples joint inference and dynamic KG refinement with large language models. TRAIL enables LLM agents to iteratively explore, update, and refine knowledge graphs during the reasoning process, employing a confidence-driven mechanism for the generation, validation, and pruning of new facts. This plug-and-play architecture facilitates seamless integration with various LLMs, supporting continual adaptation without the need for retraining. Extensive experiments on multiple benchmarks demonstrate that TRAIL outperforms existing KG-augmented and retrieval-augmented LLM baselines by 3% to 13%. More importantly, these results represent a significant step toward developing adaptive, memory-augmented language models capable of continual learning and reliable, transparent reasoning.
Red teaming is critical for identifying vulnerabilities and building trust in current LLMs. However, current automated methods for Large Language Models (LLMs) rely on brittle prompt templates or single-turn attacks, failing to capture the complex, interactive nature of real-world adversarial dialogues. We propose a novel paradigm: training an AI to strategically `break' another AI. By formalizing red teaming as a Markov Decision Process (MDP) and employing a hierarchical Reinforcement Learning (RL) framework, we effectively address the inherent sparse reward and long-horizon challenges. Our generative agent learns coherent, multi-turn attack strategies through a fine-grained, token-level harm reward, enabling it to uncover subtle vulnerabilities missed by existing baselines. This approach sets a new state-of-the-art, fundamentally reframing LLM red teaming as a dynamic, trajectory-based process (rather than a one-step test) essential for robust AI deployment.
Effective customer support requires not only accurate problem solving but also structured and empathetic communication aligned with professional standards. However, existing dialogue datasets often lack strategic guidance, and real-world service data is difficult to access and annotate. To address this, we introduce the task of Customer Support Conversation (CSC), aimed at training customer service agents to respond using well-defined support strategies. We propose a structured CSC framework grounded in COPC guidelines, defining five conversational stages and twelve strategies to guide high-quality interactions. Based on this, we construct CSConv, an evaluation dataset of 1,855 real-world customer-agent conversations rewritten using LLMs to reflect deliberate strategy use, and annotated accordingly. Additionally, we develop a role-playing approach that simulates strategy-rich conversations using LLM-powered roles aligned with the CSC framework, resulting in the training dataset RoleCS. Experiments show that fine-tuning strong LLMs on RoleCS significantly improves their ability to generate high-quality, strategy-aligned responses on CSConv. Human evaluations further confirm gains in problem resolution. All code and data will be made publicly available at https://github.com/aliyun/qwen-dianjin.
Frontier LLMs only recently enabled serviceable, autonomous web agents. At that, a model poses as an instantaneous domain model backend. Ought to suggest interaction, it is consulted with a web-based task and respective application state. The key problem lies in application state serialisation $\unicode{x2013}$ referred to as snapshot. State-of-the-art web agents are premised on grounded GUI snapshots, i.e., screenshots enhanced with visual cues. Not least to resemble human perception, but for images representing relatively cheap means of model input. LLM vision still lag behind code interpretation capabilities. DOM snapshots, which structurally resemble HTML, impose a desired alternative. Vast model input token size, however, disables reliable implementation with web agents to date. We propose D2Snap, a first-of-its-kind DOM downsampling algorithm. Based on a GPT-4o backend, we evaluate D2Snap on tasks sampled from the Online-Mind2Web dataset. The success rate of D2Snap-downsampled DOM snapshots (67%) matches a grounded GUI snapshot baseline (65%) $\unicode{x2013}$ within the same input token order of magnitude (1e3). Our best evaluated configurations $\unicode{x2013}$ one token order above, but within the model's context window $\unicode{x2013}$ outperform this baseline by 8%. Our evaluation, moreover, yields that DOM-inherent hierarchy embodies a strong UI feature for LLMs.
Literature reviews play an important role in scientific research. Recent advances in large language models (LLMs) have boosted the development of automated systems for the entire literature review workflow, from retrieval to manuscript drafting. However, a key challenge is that mistakes made in early stages can propagate and amplify in subsequent steps, leading to compounding errors that undermine the faithfulness of the final review. To tackle this issue, we propose the Multi-Agent Taskforce Collaboration (MATC) framework, which consists of a manager agent and four executor agents for literature searching, outline generation, fact localization, and manuscript drafting. We propose three novel collaboration paradigms, forming exploration, exploitation, and experience taskforces, to effectively organize agents and mitigate compounding errors both between and within executor agents. Experimental results show that MATC achieves state-of-the-art performance on existing benchmarks. We further propose a new benchmark dataset featuring more diverse topics for faithful literature review generation.
Interactive multimodal agents must convert raw visual observations into coherent sequences of language-conditioned actions -- a capability that current vision-language models (VLMs) still lack. Earlier reinforcement-learning (RL) efforts could, in principle, endow VLMs with such skills, but they have seldom tested whether the learned behaviours generalize beyond their training simulators, and they depend either on brittle hyperparameter tuning or on dense-reward environments with low state variability. We introduce Vision-Language Decoupled Actor-Critic (VL-DAC), a lightweight, hyperparameter-free RL algorithm. VL-DAC applies PPO updates to action tokens while learning value only at the environment-step level: an arrangement, to our knowledge, not previously explored for large VLMs or LLMs. This simple decoupling removes unstable weighting terms and yields faster, more reliable convergence. Training a single VLM with VL-DAC in one inexpensive simulator at a time (MiniWorld, Gym-Cards, ALFWorld, or WebShop) already produces policies that generalize widely: +50\% relative on BALROG (game-centric agentic control), +5\% relative on the hardest part of VSI-Bench (spatial planning), and +2\% on VisualWebBench (web navigation), all without degrading general image understanding accuracy. These results provide the first evidence that a simple RL algorithm can train VLMs entirely in cheap synthetic worlds while delivering measurable gains on real-image agentic, spatial-reasoning, and web-navigation benchmarks.
Existing benchmarks in e-commerce primarily focus on basic user intents, such as finding or purchasing products. However, real-world users often pursue more complex goals, such as applying vouchers, managing budgets, and finding multi-products seller. To bridge this gap, we propose ShoppingBench, a novel end-to-end shopping benchmark designed to encompass increasingly challenging levels of grounded intent. Specifically, we propose a scalable framework to simulate user instructions based on various intents derived from sampled real-world products. To facilitate consistent and reliable evaluations, we provide a large-scale shopping sandbox that serves as an interactive simulated environment, incorporating over 2.5 million real-world products. Experimental results demonstrate that even state-of-the-art language agents (such as GPT-4.1) achieve absolute success rates under 50% on our benchmark tasks, highlighting the significant challenges posed by our ShoppingBench. In addition, we propose a trajectory distillation strategy and leverage supervised fine-tuning, along with reinforcement learning on synthetic trajectories, to distill the capabilities of a large language agent into a smaller one. As a result, our trained agent achieves competitive performance compared to GPT-4.1.
Large Language Model (LLM) powered agents have emerged as effective planners for Automated Machine Learning (AutoML) systems. While most existing AutoML approaches focus on automating feature engineering and model architecture search, recent studies in time series forecasting suggest that lightweight models can often achieve state-of-the-art performance. This observation led us to explore improving data quality, rather than model architecture, as a potentially fruitful direction for AutoML on time series data. We propose DCATS, a Data-Centric Agent for Time Series. DCATS leverages metadata accompanying time series to clean data while optimizing forecasting performance. We evaluated DCATS using four time series forecasting models on a large-scale traffic volume forecasting dataset. Results demonstrate that DCATS achieves an average 6% error reduction across all tested models and time horizons, highlighting the potential of data-centric approaches in AutoML for time series forecasting.
Prior work synthesizes tool-use LLM datasets by first generating a user query, followed by complex tool-use annotations like DFS. This leads to inevitable annotation failures and low efficiency in data generation. We introduce ToolGrad, an agentic framework that inverts this paradigm. ToolGrad first constructs valid tool-use chains through an iterative process guided by textual "gradients", and then synthesizes corresponding user queries. This "answer-first" approach led to ToolGrad-5k, a dataset generated with more complex tool use, lower cost, and 100% pass rate. Experiments show that models trained on ToolGrad-5k outperform those on expensive baseline datasets and proprietary LLMs, even on OOD benchmarks.
Recent studies have extended the application of large language models (LLMs) to geographic problems, revealing surprising geospatial competence even without explicit spatial supervision. However, LLMs still face challenges in spatial consistency, multi-hop reasoning, and geographic bias. To address these issues, we propose GeoSR, a self-refining agentic reasoning framework that embeds core geographic principles -- most notably Tobler's First Law of Geography -- into an iterative prediction loop. In GeoSR, the reasoning process is decomposed into three collaborating agents: (1) a variable-selection agent that selects relevant covariates from the same location; (2) a point-selection agent that chooses reference predictions at nearby locations generated by the LLM in previous rounds; and (3) a refine agent that coordinates the iterative refinement process by evaluating prediction quality and triggering further rounds when necessary. This agentic loop progressively improves prediction quality by leveraging both spatial dependencies and inter-variable relationships. We validate GeoSR on tasks ranging from physical-world property estimation to socioeconomic prediction. Experimental results show consistent improvements over standard prompting strategies, demonstrating that incorporating geostatistical priors and spatially structured reasoning into LLMs leads to more accurate and equitable geospatial predictions. The code of GeoSR is available at https://github.com/JinfanTang/GeoSR.
Motion sensor time-series are central to human activity recognition (HAR), with applications in health, sports, and smart devices. However, existing methods are trained for fixed activity sets and require costly retraining when new behaviours or sensor setups appear. Recent attempts to use large language models (LLMs) for HAR, typically by converting signals into text or images, suffer from limited accuracy and lack verifiable interpretability. We propose ZARA, the first agent-based framework for zero-shot, explainable HAR directly from raw motion time-series. ZARA integrates an automatically derived pair-wise feature knowledge base that captures discriminative statistics for every activity pair, a multi-sensor retrieval module that surfaces relevant evidence, and a hierarchical agent pipeline that guides the LLM to iteratively select features, draw on this evidence, and produce both activity predictions and natural-language explanations. ZARA enables flexible and interpretable HAR without any fine-tuning or task-specific classifiers. Extensive experiments on 8 HAR benchmarks show that ZARA achieves SOTA zero-shot performance, delivering clear reasoning while exceeding the strongest baselines by 2.53x in macro F1. Ablation studies further confirm the necessity of each module, marking ZARA as a promising step toward trustworthy, plug-and-play motion time-series analysis. Our codes are available at https://github.com/zechenli03/ZARA.
As large language models (LLMs) demonstrate increasingly powerful reasoning and orchestration capabilities, LLM-based agents are rapidly proliferating for complex data-related tasks. Despite this progress, the current design of how LLMs interact with databases exhibits critical limitations in usability, security, privilege management, and data transmission efficiency. To resolve these challenges, we introduce BridgeScope, a universal toolkit bridging LLMs and databases through three key innovations. First, it modularizes SQL operations into fine-grained tools for context retrieval, CRUD execution, and ACID-compliant transaction management, enabling more precise and LLM-friendly functionality controls. Second, it aligns tool implementations with both database privileges and user security policies to steer LLMs away from unsafe or unauthorized operations, improving task execution efficiency while safeguarding database security. Third, it introduces a proxy mechanism for seamless inter-tool data transfer, bypassing LLM transmission bottlenecks. All of these designs are database-agnostic and can be transparently integrated with existing agent architectures. We also release an open-source implementation of BridgeScope for PostgreSQL. Evaluations on two novel benchmarks demonstrate that BridgeScope enables LLM agents to operate databases more effectively, reduces token usage by up to 80% through improved security awareness, and uniquely supports data-intensive workflows beyond existing toolkits, establishing BridgeScope as a robust foundation for next-generation intelligent data automation.
Intelligent personal assistants (IPAs) such as Siri and Google Assistant are designed to enhance human capabilities and perform tasks on behalf of users. The emergence of LLM agents brings new opportunities for the development of IPAs. While responsive capabilities have been widely studied, proactive behaviors remain underexplored. Designing an IPA that is proactive, privacy-preserving, and capable of self-evolution remains a significant challenge. Designing such IPAs relies on the cognitive architecture of LLM agents. This work proposes Cognition Forest, a semantic structure designed to align cognitive modeling with system-level design. We unify cognitive architecture and system design into a self-reinforcing loop instead of treating them separately. Based on this principle, we present Galaxy, a framework that supports multidimensional interactions and personalized capability generation. Two cooperative agents are implemented based on Galaxy: KoRa, a cognition-enhanced generative agent that supports both responsive and proactive skills; and Kernel, a meta-cognition-based meta-agent that enables Galaxy's self-evolution and privacy preservation. Experimental results show that Galaxy outperforms multiple state-of-the-art benchmarks. Ablation studies and real-world interaction cases validate the effectiveness of Galaxy.
Designing effective algorithmic components remains a fundamental obstacle in tackling NP-hard combinatorial optimization problems (COPs), where solvers often rely on carefully hand-crafted strategies. Despite recent advances in using large language models (LLMs) to synthesize high-quality components, most approaches restrict the search to a single element - commonly a heuristic scoring function - thus missing broader opportunities for innovation. In this paper, we introduce a broader formulation of solver design as a multi-strategy optimization problem, which seeks to jointly improve a set of interdependent components under a unified objective. To address this, we propose Multi-strategy Optimization via Turn-based Interactive Framework (MOTIF) - a novel framework based on Monte Carlo Tree Search that facilitates turn-based optimization between two LLM agents. At each turn, an agent improves one component by leveraging the history of both its own and its opponent's prior updates, promoting both competitive pressure and emergent cooperation. This structured interaction broadens the search landscape and encourages the discovery of diverse, high-performing solutions. Experiments across multiple COP domains show that MOTIF consistently outperforms state-of-the-art methods, highlighting the promise of turn-based, multi-agent prompting for fully automated solver design.
Social intelligence has become a critical capability for large language models (LLMs), enabling them to engage effectively in real-world social tasks such as accommodation, persuasion, collaboration, and negotiation. Reinforcement learning (RL) is a natural fit for training socially intelligent agents because it allows models to learn sophisticated strategies directly through social interactions. However, social interactions have two key characteristics that set barriers for RL training: (1) partial observability, where utterances have indirect and delayed effects that complicate credit assignment, and (2) multi-dimensionality, where behaviors such as rapport-building or knowledge-seeking contribute indirectly to goal achievement. These characteristics make Markov decision process (MDP)-based RL with single-dimensional episode-level rewards inefficient and unstable. To address these challenges, we propose Sotopia-RL, a novel framework that refines coarse episode-level feedback into utterance-level, multi-dimensional rewards. Utterance-level credit assignment mitigates partial observability by attributing outcomes to individual utterances, while multi-dimensional rewards capture the full richness of social interactions and reduce reward hacking. Experiments in Sotopia, an open-ended social learning environment, demonstrate that Sotopia-RL achieves state-of-the-art social goal completion scores (7.17 on Sotopia-hard and 8.31 on Sotopia-full), significantly outperforming existing approaches. Ablation studies confirm the necessity of both utterance-level credit assignment and multi-dimensional reward design for RL training. Our implementation is publicly available at: https://github.com/sotopia-lab/sotopia-rl.
Large Language Models (LLMs) are trained on vast and diverse internet corpora that often include inaccurate or misleading content. Consequently, LLMs can generate misinformation, making robust fact-checking essential. This review systematically analyzes how LLM-generated content is evaluated for factual accuracy by exploring key challenges such as hallucinations, dataset limitations, and the reliability of evaluation metrics. The review emphasizes the need for strong fact-checking frameworks that integrate advanced prompting strategies, domain-specific fine-tuning, and retrieval-augmented generation (RAG) methods. It proposes five research questions that guide the analysis of the recent literature from 2020 to 2025, focusing on evaluation methods and mitigation techniques. The review also discusses the role of instruction tuning, multi-agent reasoning, and external knowledge access via RAG frameworks. Key findings highlight the limitations of current metrics, the value of grounding outputs with validated external evidence, and the importance of domain-specific customization to improve factual consistency. Overall, the review underlines the importance of building LLMs that are not only accurate and explainable but also tailored for domain-specific fact-checking. These insights contribute to the advancement of research toward more trustworthy and context-aware language models.
Long-context large language models (LLMs), such as Gemini-2.5-Pro and Claude-Sonnet-4, are increasingly used to empower advanced AI systems, including retrieval-augmented generation (RAG) pipelines and autonomous agents. In these systems, an LLM receives an instruction along with a context--often consisting of texts retrieved from a knowledge database or memory--and generates a response that is contextually grounded by following the instruction. Recent studies have designed solutions to trace back to a subset of texts in the context that contributes most to the response generated by the LLM. These solutions have numerous real-world applications, including performing post-attack forensic analysis and improving the interpretability and trustworthiness of LLM outputs. While significant efforts have been made, state-of-the-art solutions such as TracLLM often lead to a high computation cost, e.g., it takes TracLLM hundreds of seconds to perform traceback for a single response-context pair. In this work, we propose AttnTrace, a new context traceback method based on the attention weights produced by an LLM for a prompt. To effectively utilize attention weights, we introduce two techniques designed to enhance the effectiveness of AttnTrace, and we provide theoretical insights for our design choice. We also perform a systematic evaluation for AttnTrace. The results demonstrate that AttnTrace is more accurate and efficient than existing state-of-the-art context traceback methods. We also show that AttnTrace can improve state-of-the-art methods in detecting prompt injection under long contexts through the attribution-before-detection paradigm. As a real-world application, we demonstrate that AttnTrace can effectively pinpoint injected instructions in a paper designed to manipulate LLM-generated reviews. The code is at https://github.com/Wang-Yanting/AttnTrace.
We present Agent Lightning, a flexible and extensible framework that enables Reinforcement Learning (RL)-based training of Large Language Models (LLMs) for any AI agent. Unlike existing methods that tightly couple RL training with agent or rely on sequence concatenation with masking, Agent Lightning achieves complete decoupling between agent execution and training, allowing seamless integration with existing agents developed via diverse ways (e.g., using frameworks like LangChain, OpenAI Agents SDK, AutoGen, and building from scratch) with almost ZERO code modifications. By formulating agent execution as Markov decision process, we define an unified data interface and propose a hierarchical RL algorithm, LightningRL, which contains a credit assignment module, allowing us to decompose trajectories generated by ANY agents into training transition. This enables RL to handle complex interaction logic, such as multi-agent scenarios and dynamic workflows. For the system design, we introduce a Training-Agent Disaggregation architecture, and brings agent observability frameworks into agent runtime, providing a standardized agent finetuning interface. Experiments across text-to-SQL, retrieval-augmented generation, and math tool-use tasks demonstrate stable, continuous improvements, showcasing the framework's potential for real-world agent training and deployment.
Generative models, particularly Large Language Models (LLMs), produce fluent outputs yet lack verifiable guarantees. We adapt Design by Contract (DbC) and type-theoretic principles to introduce a contract layer that mediates every LLM call. Contracts stipulate semantic and type requirements on inputs and outputs, coupled with probabilistic remediation to steer generation toward compliance. The layer exposes the dual view of LLMs as semantic parsers and probabilistic black-box components. Contract satisfaction is probabilistic and semantic validation is operationally defined through programmer-specified conditions on well-typed data structures. More broadly, this work postulates that any two agents satisfying the same contracts are \emph{functionally equivalent} with respect to those contracts.
Research on applications of Reinforcement Learning (RL) to Large Language Models (LLMs) has mostly been focused on single-turn problems, such as mathematical reasoning or single-shot code generation. While these problems can be viewed as token-level multi-turn MDPs, this view corresponds to a degenerate case of multi-turn interaction where the environment provides no feedback. This contrasts with many real-world domains, such as software engineering (SWE), which require rich multi-turn interactions with a stateful environment that responds to each action with a non-trivial observation. To bridge this gap, we demonstrate the successful application of RL to this general regime. Using a modified Decoupled Advantage Policy Optimization (DAPO) algorithm, we train an agent based on Qwen2.5-72B-Instruct to solve real-world software engineering tasks. Our approach increases the agent's success rate on the SWE-bench Verified benchmark from a 20% rejection fine-tuned baseline to 39%, without relying on any teacher models. On SWE-rebench, our agent matches or outperforms leading open-weight models such as DeepSeek-V3-0324 and Qwen3-235B-A22B using an identical scaffolding, offering a viable path toward building more capable autonomous agents for complex real-world problems based on open models.
Drug discovery frequently loses momentum when data, expertise, and tools are scattered, slowing design cycles. To shorten this loop we built a hierarchical, tool using agent framework that automates molecular optimisation. A Principal Researcher defines each objective, a Database agent retrieves target information, an AI Expert generates de novo scaffolds with a sequence to molecule deep learning model, a Medicinal Chemist edits them while invoking a docking tool, a Ranking agent scores the candidates, and a Scientific Critic polices the logic. Each tool call is summarised and stored causing the full reasoning path to remain inspectable. The agents communicate through concise provenance records that capture molecular lineage, to build auditable, molecule centered reasoning trajectories and reuse successful transformations via in context learning. Three cycle research loops were run against AKT1 protein using five large language models. After ranking the models by mean docking score, we ran 20 independent scale ups on the two top performers. We then compared the leading LLMs' binding affinity results across three configurations, LLM only, single agent, and multi agent. Our results reveal an architectural trade off, the multi agent setting excelled at focused binding optimization, improving average predicted binding affinity by 31%. In contrast, single agent runs generated molecules with superior drug like properties at the cost of less potent binding scores. Unguided LLM runs finished fastest, yet their lack of transparent tool signals left the validity of their reasoning paths unverified. These results show that test time scaling, focused feedback loops and provenance convert general purpose LLMs into auditable systems for molecular design, and suggest that extending the toolset to ADMET and selectivity predictors could push research workflows further along the discovery pipeline.
The rapid expansion of publicly-available medical data presents a challenge for clinicians and researchers alike, increasing the gap between the volume of scientific literature and its applications. The steady growth of studies and findings overwhelms medical professionals at large, hindering their ability to systematically review and understand the latest knowledge. This paper presents an approach to information extraction and automatic knowledge graph (KG) generation to identify and connect biomedical knowledge. Through a pipeline of large language model (LLM) agents, the system decomposes 44 PubMed abstracts into semantically meaningful proposition sentences and extracts KG triples from these sentences. The triples are enhanced using a combination of open domain and ontology-based information extraction methodologies to incorporate ontological categories. On top of this, a context variable is included during extraction to allow the triple to stand on its own - thereby becoming `quadruples'. The extraction accuracy of the LLM is validated by comparing natural language sentences generated from the enhanced triples to the original propositions, achieving an average cosine similarity of 0.874. The similarity for generated sentences of enhanced triples were compared with generated sentences of ordinary triples showing an increase as a result of the context variable. Furthermore, this research explores the ability for LLMs to infer new relationships and connect clusters in the knowledge base of the knowledge graph. This approach leads the way to provide medical practitioners with a centralised, updated in real-time, and sustainable knowledge source, and may be the foundation of similar gains in a wide variety of fields.
In real-world routing problems, users often propose conflicting or unreasonable requirements, which result in infeasible optimization models due to overly restrictive or contradictory constraints, leading to an empty feasible solution set. Existing Large Language Model (LLM)-based methods attempt to diagnose infeasible models, but modifying such models often involves multiple potential adjustments that these methods do not consider. To fill this gap, we introduce Multi-Objective Infeasibility Diagnosis (MOID), which combines LLM agents and multi-objective optimization within an automatic routing solver, to provide a set of representative actionable suggestions. Specifically, MOID employs multi-objective optimization to consider both path cost and constraint violation, generating a set of trade-off solutions, each encompassing varying degrees of model adjustments. To extract practical insights from these solutions, MOID utilizes LLM agents to generate a solution analysis function for the infeasible model. This function analyzes these distinct solutions to diagnose the original infeasible model, providing users with diverse diagnostic insights and suggestions. Finally, we compare MOID with several LLM-based methods on 50 types of infeasible routing problems. The results indicate that MOID automatically generates multiple diagnostic suggestions in a single run, providing more practical insights for restoring model feasibility and decision-making compared to existing methods.
The core challenge in automotive exterior design is balancing subjective aesthetics with objective aerodynamic performance while dramatically accelerating the development cycle. To address this, we propose a novel, LLM-driven multi-agent framework that automates the end-to-end workflow from ambiguous requirements to 3D concept model performance validation. The workflow is structured in two stages: conceptual generation and performance validation. In the first stage, agents collaborate to interpret fuzzy design requirements, generate concept sketches, and produce photorealistic renderings using diffusion models. In the second stage, the renderings are converted to 3D point clouds, where a Drag Prediction Agent, built upon a lightweight surrogate model, provides near-instantaneous predictions of the drag coefficient and pressure fields, replacing time-consuming CFD simulations. The primary contribution of this work is the seamless integration of creative generation with a rapid engineering validation loop within a unified, automated system, which provides a new paradigm for efficiently balancing creative exploration with engineering constraints in the earliest stages of design.
The Board Game Arena library provides a framework for evaluating the decision making abilities of large language models (LLMs) through strategic board games implemented in Google OpenSpiel library. The framework enables systematic comparisons between LLM based agents and other agents (random, human, reinforcement learning agents, etc.) in various game scenarios by wrapping multiple board and matrix games and supporting different agent types. It integrates API access to models via LiteLLM, local model deployment via vLLM, and offers distributed execution through Ray. Additionally it provides extensive analysis tools for the LLM reasoning traces. This paper summarizes the structure, key characteristics, and motivation of the repository, highlighting how it contributes to the empirical evaluation of the reasoning of LLM and game-theoretic behavior