Large language models excel at function- and file-level code generation, yet generating complete repositories from scratch remains a fundamental challenge. This process demands coherent and reliable planning across proposal- and implementation-level stages, while natural language, due to its ambiguity and verbosity, is ill-suited for faithfully representing complex software structures. To address this, we introduce the Repository Planning Graph (RPG), a persistent representation that unifies proposal- and implementation-level planning by encoding capabilities, file structures, data flows, and functions in one graph. RPG replaces ambiguous natural language with an explicit blueprint, enabling long-horizon planning and scalable repository generation. Building on RPG, we develop ZeroRepo, a graph-driven framework for repository generation from scratch. It operates in three stages: proposal-level planning and implementation-level refinement to construct the graph, followed by graph-guided code generation with test validation. To evaluate this setting, we construct RepoCraft, a benchmark of six real-world projects with 1,052 tasks. On RepoCraft, ZeroRepo produces repositories averaging nearly 36K LOC, roughly 3.9$\times$ the strongest baseline (Claude Code) and about 64$\times$ other baselines. It attains 81.5% functional coverage and a 69.7% pass rate, exceeding Claude Code by 27.3 and 35.8 percentage points, respectively. Further analysis shows that RPG models complex dependencies, enables progressively more sophisticated planning through near-linear scaling, and enhances LLM understanding of repositories, thereby accelerating agent localization.
Code translation transforms source code from one programming language (PL) to another. Validating the functional equivalence of translation and repairing, if necessary, are critical steps in code translation. Existing automated validation and repair approaches struggle to generalize to many PLs due to high engineering overhead, and they rely on existing and often inadequate test suites, which results in false claims of equivalence and ineffective translation repair. We develop MatchFixAgent, a large language model (LLM)-based, PL-agnostic framework for equivalence validation and repair of translations. MatchFixAgent features a multi-agent architecture that divides equivalence validation into several sub-tasks to ensure thorough and consistent semantic analysis of the translation. Then it feeds this analysis to test agent to write and execute tests. Upon observing a test failure, the repair agent attempts to fix the translation bug. The final (in)equivalence decision is made by the verdict agent, considering semantic analyses and test execution results. We compare MatchFixAgent's validation and repair results with four repository-level code translation techniques. We use 2,219 translation pairs from their artifacts, which cover 6 PL pairs, and are collected from 24 GitHub projects totaling over 900K lines of code. Our results demonstrate that MatchFixAgent produces (in)equivalence verdicts for 99.2% of translation pairs, with the same equivalence validation result as prior work on 72.8% of them. When MatchFixAgent's result disagrees with prior work, we find that 60.7% of the time MatchFixAgent's result is actually correct. In addition, we show that MatchFixAgent can repair 50.6% of inequivalent translation, compared to prior work's 18.5%. This demonstrates that MatchFixAgent is far more adaptable to many PL pairs than prior work, while producing highly accurate validation results.
We present Agentic Aerial Cinematography: From Dialogue Cues to Cinematic Trajectories (ACDC), an autonomous drone cinematography system driven by natural language communication between human directors and drones. The main limitation of previous drone cinematography workflows is that they require manual selection of waypoints and view angles based on predefined human intent, which is labor-intensive and yields inconsistent performance. In this paper, we propose employing large language models (LLMs) and vision foundation models (VFMs) to convert free-form natural language prompts directly into executable indoor UAV video tours. Specifically, our method comprises a vision-language retrieval pipeline for initial waypoint selection, a preference-based Bayesian optimization framework that refines poses using aesthetic feedback, and a motion planner that generates safe quadrotor trajectories. We validate ACDC through both simulation and hardware-in-the-loop experiments, demonstrating that it robustly produces professional-quality footage across diverse indoor scenes without requiring expertise in robotics or cinematography. These results highlight the potential of embodied AI agents to close the loop from open-vocabulary dialogue to real-world autonomous aerial cinematography.
Language-guided supervision, which utilizes a frozen semantic target from a Pretrained Language Model (PLM), has emerged as a promising paradigm for visual Continual Learning (CL). However, relying on a single target introduces two critical limitations: 1) semantic ambiguity, where a polysemous category name results in conflicting visual representations, and 2) intra-class visual diversity, where a single prototype fails to capture the rich variety of visual appearances within a class. To this end, we propose MuproCL, a novel framework that replaces the single target with multiple, context-aware prototypes. Specifically, we employ a lightweight LLM agent to perform category disambiguation and visual-modal expansion to generate a robust set of semantic prototypes. A LogSumExp aggregation mechanism allows the vision model to adaptively align with the most relevant prototype for a given image. Extensive experiments across various CL baselines demonstrate that MuproCL consistently enhances performance and robustness, establishing a more effective path for language-guided continual learning.
Background: Large language models (LLMs) show promise in medicine, but their deployment in hospitals is limited by restricted access to electronic health record (EHR) systems. The Model Context Protocol (MCP) enables integration between LLMs and external tools. Objective: To evaluate whether an LLM connected to an EHR database via MCP can autonomously retrieve clinically relevant information in a real hospital setting. Methods: We developed EHR-MCP, a framework of custom MCP tools integrated with the hospital EHR database, and used GPT-4.1 through a LangGraph ReAct agent to interact with it. Six tasks were tested, derived from use cases of the infection control team (ICT). Eight patients discussed at ICT conferences were retrospectively analyzed. Agreement with physician-generated gold standards was measured. Results: The LLM consistently selected and executed the correct MCP tools. Except for two tasks, all tasks achieved near-perfect accuracy. Performance was lower in the complex task requiring time-dependent calculations. Most errors arose from incorrect arguments or misinterpretation of tool results. Responses from EHR-MCP were reliable, though long and repetitive data risked exceeding the context window. Conclusions: LLMs can retrieve clinical data from an EHR via MCP tools in a real hospital setting, achieving near-perfect performance in simple tasks while highlighting challenges in complex ones. EHR-MCP provides an infrastructure for secure, consistent data access and may serve as a foundation for hospital AI agents. Future work should extend beyond retrieval to reasoning, generation, and clinical impact assessment, paving the way for effective integration of generative AI into clinical practice.
While reinforcement learning from scratch has shown impressive results in solving sequential decision-making tasks with efficient simulators, real-world applications with expensive interactions require more sample-efficient agents. Foundation models (FMs) are natural candidates to improve sample efficiency as they possess broad knowledge and reasoning capabilities, but it is yet unclear how to effectively integrate them into the reinforcement learning framework. In this paper, we anticipate and, most importantly, evaluate two promising strategies. First, we consider the use of foundation world models (FWMs) that exploit the prior knowledge of FMs to enable training and evaluating agents with simulated interactions. Second, we consider the use of foundation agents (FAs) that exploit the reasoning capabilities of FMs for decision-making. We evaluate both approaches empirically in a family of grid-world environments that are suitable for the current generation of large language models (LLMs). Our results suggest that improvements in LLMs already translate into better FWMs and FAs; that FAs based on current LLMs can already provide excellent policies for sufficiently simple environments; and that the coupling of FWMs and reinforcement learning agents is highly promising for more complex settings with partial observability and stochastic elements.
This paper presents MicroRCA-Agent, an innovative solution for microservice root cause analysis based on large language model agents, which constructs an intelligent fault root cause localization system with multimodal data fusion. The technical innovations are embodied in three key aspects: First, we combine the pre-trained Drain log parsing algorithm with multi-level data filtering mechanism to efficiently compress massive logs into high-quality fault features. Second, we employ a dual anomaly detection approach that integrates Isolation Forest unsupervised learning algorithms with status code validation to achieve comprehensive trace anomaly identification. Third, we design a statistical symmetry ratio filtering mechanism coupled with a two-stage LLM analysis strategy to enable full-stack phenomenon summarization across node-service-pod hierarchies. The multimodal root cause analysis module leverages carefully designed cross-modal prompts to deeply integrate multimodal anomaly information, fully exploiting the cross-modal understanding and logical reasoning capabilities of large language models to generate structured analysis results encompassing fault components, root cause descriptions, and reasoning trace. Comprehensive ablation studies validate the complementary value of each modal data and the effectiveness of the system architecture. The proposed solution demonstrates superior performance in complex microservice fault scenarios, achieving a final score of 50.71. The code has been released at: https://github.com/tangpan360/MicroRCA-Agent.
Modern AI-powered Integrated Development Environments (AI-IDEs) are increasingly defined by an Agent-centric architecture, where an LLM-powered Agent is deeply integrated to autonomously execute complex tasks. This tight integration, however, also introduces a new and critical attack surface. Attackers can exploit these components by injecting malicious instructions into untrusted external sources, effectively hijacking the Agent to perform harmful operations beyond the user's intention or awareness. This emerging threat has quickly attracted research attention, leading to various proposed attack vectors, such as hijacking Model Context Protocol (MCP) Servers to access private data. However, most existing approaches lack stealth and persistence, limiting their practical impact. We propose the Cuckoo Attack, a novel attack that achieves stealthy and persistent command execution by embedding malicious payloads into configuration files. These files, commonly used in AI-IDEs, execute system commands during routine operations, without displaying execution details to the user. Once configured, such files are rarely revisited unless an obvious runtime error occurs, creating a blind spot for attackers to exploit. We formalize our attack paradigm into two stages, including initial infection and persistence. Based on these stages, we analyze the practicality of the attack execution process and identify the relevant exploitation techniques. Furthermore, we analyze the impact of Cuckoo Attack, which can not only invade the developer's local computer but also achieve supply chain attacks through the spread of configuration files. We contribute seven actionable checkpoints for vendors to evaluate their product security. The critical need for these checks is demonstrated by our end-to-end Proof of Concept, which validated the proposed attack across nine mainstream Agent and AI-IDE pairs.
High-quality long-context data is essential for training large language models (LLMs) capable of processing extensive documents, yet existing synthesis approaches using relevance-based aggregation face challenges of computational efficiency. We present LiteLong, a resource-efficient method for synthesizing long-context data through structured topic organization and multi-agent debate. Our approach leverages the BISAC book classification system to provide a comprehensive hierarchical topic organization, and then employs a debate mechanism with multiple LLMs to generate diverse, high-quality topics within this structure. For each topic, we use lightweight BM25 retrieval to obtain relevant documents and concatenate them into 128K-token training samples. Experiments on HELMET and Ruler benchmarks demonstrate that LiteLong achieves competitive long-context performance and can seamlessly integrate with other long-dependency enhancement methods. LiteLong makes high-quality long-context data synthesis more accessible by reducing both computational and data engineering costs, facilitating further research in long-context language training.
Slang is a commonly used type of informal language that poses a daunting challenge to NLP systems. Recent advances in large language models (LLMs), however, have made the problem more approachable. While LLM agents are becoming more widely applied to intermediary tasks such as slang detection and slang interpretation, their generalizability and reliability are heavily dependent on whether these models have captured structural knowledge about slang that align well with human attested slang usages. To answer this question, we contribute a systematic comparison between human and machine-generated slang usages. Our evaluative framework focuses on three core aspects: 1) Characteristics of the usages that reflect systematic biases in how machines perceive slang, 2) Creativity reflected by both lexical coinages and word reuses employed by the slang usages, and 3) Informativeness of the slang usages when used as gold-standard examples for model distillation. By comparing human-attested slang usages from the Online Slang Dictionary (OSD) and slang generated by GPT-4o and Llama-3, we find significant biases in how LLMs perceive slang. Our results suggest that while LLMs have captured significant knowledge about the creative aspects of slang, such knowledge does not align with humans sufficiently to enable LLMs for extrapolative tasks such as linguistic analyses.
With the increasing popularity of large language models (LLMs) and LLM-based agents, reliable and effective code evaluation metrics (CEMs) have become crucial for progress across several software engineering tasks. While popular benchmarks often provide test cases to assess the correctness of generated code, crafting and executing test cases is expensive. Reference-based CEMs provide a cheaper alternative by scoring a candidate program based on its functional similarity to a reference. Although prior research has focused on reporting the weak correlation between these CEMs and functional correctness, the causes are only assumed, and plausible solutions remain unexplored. In this work, we critically evaluate four state-of-the-art reference-based CEMs, revealing their strong bias towards surface-level features rather than code functionality. Despite this surface bias, current evaluation datasets for these CEMs rarely include code pairs that are surface-similar yet functionally dissimilar, or functionally similar yet surface-dissimilar. To mitigate this gap, we propose LoCaL (Looks Can Lie), a CEM evaluation benchmark, with 3117 code pairs at both the method and program levels. Each pair is labeled with a functional similarity score and aims to target regions where CEMs are likely to perform poorly. The functional similarity scores are calculated through differential fuzzing, which eliminates the need for predefined test cases and, at the same time, improves the reliability of the scores by executing an order of magnitude more tests than prior work. We find that all four CEMs show significant performance degradation on LoCaL, compared to the baselines. Finally, based on our findings, we draw the implication that exposing CEMs to LoCaL-like data might facilitate the development of metrics that are robust to surface bias.
The rapid evolution of neural architectures - from multilayer perceptrons to large-scale Transformer-based models - has enabled language models (LLMs) to exhibit emergent agentic behaviours when equipped with memory, planning, and external tool use. However, their inherent stochasticity and multi-step decision processes render classical evaluation methods inadequate for diagnosing agentic performance. This work introduces a diagnostic framework for expert systems that not only evaluates but also facilitates the transfer of expert behaviour into LLM-powered agents. The framework integrates (i) curated golden datasets of expert annotations, (ii) silver datasets generated through controlled behavioural mutation, and (iii) an LLM-based Agent Judge that scores and prescribes targeted improvements. These prescriptions are embedded into a vectorized recommendation map, allowing expert interventions to propagate as reusable improvement trajectories across multiple system instances. We demonstrate the framework on a multi-agent recruiter-assistant system, showing that it uncovers latent cognitive failures - such as biased phrasing, extraction drift, and tool misrouting - while simultaneously steering agents toward expert-level reasoning and style. The results establish a foundation for standardized, reproducible expert behaviour transfer in stochastic, tool-augmented LLM agents, moving beyond static evaluation to active expert system refinement.
Medical decision-making often involves integrating knowledge from multiple clinical specialties, typically achieved through multidisciplinary teams. Inspired by this collaborative process, recent work has leveraged large language models (LLMs) in multi-agent collaboration frameworks to emulate expert teamwork. While these approaches improve reasoning through agent interaction, they are limited by static, pre-assigned roles, which hinder adaptability and dynamic knowledge integration. To address these limitations, we propose KAMAC, a Knowledge-driven Adaptive Multi-Agent Collaboration framework that enables LLM agents to dynamically form and expand expert teams based on the evolving diagnostic context. KAMAC begins with one or more expert agents and then conducts a knowledge-driven discussion to identify and fill knowledge gaps by recruiting additional specialists as needed. This supports flexible, scalable collaboration in complex clinical scenarios, with decisions finalized through reviewing updated agent comments. Experiments on two real-world medical benchmarks demonstrate that KAMAC significantly outperforms both single-agent and advanced multi-agent methods, particularly in complex clinical scenarios (i.e., cancer prognosis) requiring dynamic, cross-specialty expertise. Our code is publicly available at: https://github.com/XiaoXiao-Woo/KAMAC.
This paper proposes a novel architectural framework aimed at enhancing security and reliability in multi-agent systems (MAS). A central component of this framework is a network of Sentinel Agents, functioning as a distributed security layer that integrates techniques such as semantic analysis via large language models (LLMs), behavioral analytics, retrieval-augmented verification, and cross-agent anomaly detection. Such agents can potentially oversee inter-agent communications, identify potential threats, enforce privacy and access controls, and maintain comprehensive audit records. Complementary to the idea of Sentinel Agents is the use of a Coordinator Agent. The Coordinator Agent supervises policy implementation, and manages agent participation. In addition, the Coordinator also ingests alerts from Sentinel Agents. Based on these alerts, it can adapt policies, isolate or quarantine misbehaving agents, and contain threats to maintain the integrity of the MAS ecosystem. This dual-layered security approach, combining the continuous monitoring of Sentinel Agents with the governance functions of Coordinator Agents, supports dynamic and adaptive defense mechanisms against a range of threats, including prompt injection, collusive agent behavior, hallucinations generated by LLMs, privacy breaches, and coordinated multi-agent attacks. In addition to the architectural design, we present a simulation study where 162 synthetic attacks of different families (prompt injection, hallucination, and data exfiltration) were injected into a multi-agent conversational environment. The Sentinel Agents successfully detected the attack attempts, confirming the practical feasibility of the proposed monitoring approach. The framework also offers enhanced system observability, supports regulatory compliance, and enables policy evolution over time.
The emergence of large language models (LLMs) has brought a new paradigm to automated essay scoring (AES), a long-standing and practical application of natural language processing in education. However, achieving human-level multi-perspective understanding and judgment remains a challenge. In this work, we propose Roundtable Essay Scoring (RES), a multi-agent evaluation framework designed to perform precise and human-aligned scoring under a zero-shot setting. RES constructs evaluator agents based on LLMs, each tailored to a specific prompt and topic context. Each agent independently generates a trait-based rubric and conducts a multi-perspective evaluation. Then, by simulating a roundtable-style discussion, RES consolidates individual evaluations through a dialectical reasoning process to produce a final holistic score that more closely aligns with human evaluation. By enabling collaboration and consensus among agents with diverse evaluation perspectives, RES outperforms prior zero-shot AES approaches. Experiments on the ASAP dataset using ChatGPT and Claude show that RES achieves up to a 34.86% improvement in average QWK over straightforward prompting (Vanilla) methods.
In online learning environments, students often lack personalized peer interactions, which play a crucial role in supporting cognitive development and learning engagement. Although previous studies have utilized large language models (LLMs) to simulate interactive dynamic learning environments for students, these interactions remain limited to conversational exchanges, lacking insights and adaptations to the learners' individualized learning and cognitive states. As a result, students' interest in discussions with AI learning companions is low, and they struggle to gain inspiration from such interactions. To address this challenge, we propose OnlineMate, a multi-agent learning companion system driven by LLMs that integrates the Theory of Mind (ToM). OnlineMate is capable of simulating peer-like agent roles, adapting to learners' cognitive states during collaborative discussions, and inferring their psychological states, such as misunderstandings, confusion, or motivation. By incorporating Theory of Mind capabilities, the system can dynamically adjust its interaction strategies to support the development of higher-order thinking and cognition. Experimental results in simulated learning scenarios demonstrate that OnlineMate effectively fosters deep learning and discussions while enhancing cognitive engagement in online educational settings.
Health informatics research is characterized by diverse data modalities, rapid knowledge expansion, and the need to integrate insights across biomedical science, data analytics, and clinical practice. These characteristics make it particularly well-suited for agent-based approaches that can automate knowledge exploration, manage complex workflows, and generate clinically meaningful outputs. Recent progress in large language model (LLM)-based agents has demonstrated promising capabilities in literature synthesis, data analysis, and even end-to-end research execution. However, existing systems remain limited for health informatics because they lack mechanisms to interpret medical visualizations and often overlook domain-specific quality requirements. To address these gaps, we introduce OpenLens AI, a fully automated framework tailored to health informatics. OpenLens AI integrates specialized agents for literature review, data analysis, code generation, and manuscript preparation, enhanced by vision-language feedback for medical visualization and quality control for reproducibility. The framework automates the entire research pipeline, producing publication-ready LaTeX manuscripts with transparent and traceable workflows, thereby offering a domain-adapted solution for advancing health informatics research.
Retrieval-augmented Generation (RAG) is a prevalent approach for domain-specific LLMs, yet it is often plagued by "Retrieval Hallucinations"--a phenomenon where fine-tuned models fail to recognize and act upon poor-quality retrieved documents, thus undermining performance. To address this, we propose the Adversarial Collaboration RAG (AC-RAG) framework. AC-RAG employs two heterogeneous agents: a generalist Detector that identifies knowledge gaps, and a domain-specialized Resolver that provides precise solutions. Guided by a moderator, these agents engage in an adversarial collaboration, where the Detector's persistent questioning challenges the Resolver's expertise. This dynamic process allows for iterative problem dissection and refined knowledge retrieval. Extensive experiments show that AC-RAG significantly improves retrieval accuracy and outperforms state-of-the-art RAG methods across various vertical domains.
Large language models (LLMs) are increasingly being integrated into software development processes. The ability to generate code and submit pull requests with minimal human intervention, through the use of autonomous AI agents, is poised to become a standard practice. However, little is known about the practical usefulness of these pull requests and the extent to which their contributions are accepted in real-world projects. In this paper, we empirically study 567 GitHub pull requests (PRs) generated using Claude Code, an agentic coding tool, across 157 diverse open-source projects. Our analysis reveals that developers tend to rely on agents for tasks such as refactoring, documentation, and testing. The results indicate that 83.8% of these agent-assisted PRs are eventually accepted and merged by project maintainers, with 54.9% of the merged PRs are integrated without further modification. The remaining 45.1% require additional changes benefit from human revisions, especially for bug fixes, documentation, and adherence to project-specific standards. These findings suggest that while agent-assisted PRs are largely acceptable, they still benefit from human oversight and refinement.
Multi-agent reinforcement learning (MARL) holds substantial promise for intelligent decision-making in complex environments. However, it suffers from a coordination and scalability bottleneck as the number of agents increases. To address these issues, we propose the LLM-empowered expert demonstrations framework for multi-agent reinforcement learning (LEED). LEED consists of two components: a demonstration generation (DG) module and a policy optimization (PO) module. Specifically, the DG module leverages large language models to generate instructions for interacting with the environment, thereby producing high-quality demonstrations. The PO module adopts a decentralized training paradigm, where each agent utilizes the generated demonstrations to construct an expert policy loss, which is then integrated with its own policy loss. This enables each agent to effectively personalize and optimize its local policy based on both expert knowledge and individual experience. Experimental results show that LEED achieves superior sample efficiency, time efficiency, and robust scalability compared to state-of-the-art baselines.
Modeling semantic and structural information from tabular data remains a core challenge for effective table understanding. Existing Table-as-Text approaches flatten tables for large language models (LLMs), but lose crucial structural cues, while Table-as-Image methods preserve structure yet struggle with fine-grained semantics. Recent Table-as-Multimodality strategies attempt to combine textual and visual views, but they (1) statically process both modalities for every query-table pair within a large multimodal LLMs (MLLMs), inevitably introducing redundancy and even conflicts, and (2) depend on costly fine-tuning of MLLMs. In light of this, we propose TableDART, a training-efficient framework that integrates multimodal views by reusing pretrained single-modality models. TableDART introduces a lightweight 2.59M-parameter MLP gating network that dynamically selects the optimal path (either Text-only, Image-only, or Fusion) for each table-query pair, effectively reducing redundancy and conflicts from both modalities. In addition, we propose a novel agent to mediate cross-modal knowledge integration by analyzing outputs from text- and image-based models, either selecting the best result or synthesizing a new answer through reasoning. This design avoids the prohibitive costs of full MLLM fine-tuning. Extensive experiments on seven benchmarks show that TableDART establishes new state-of-the-art performance among open-source models, surpassing the strongest baseline by an average of 4.02%. The code is available at: https://anonymous.4open.science/r/TableDART-C52B
With the growing adoption of Large Language Models (LLMs) in automating complex, multi-agent workflows, organizations face mounting risks from errors, emergent behaviors, and systemic failures that current evaluation methods fail to capture. We present AgentCompass, the first evaluation framework designed specifically for post-deployment monitoring and debugging of agentic workflows. AgentCompass models the reasoning process of expert debuggers through a structured, multi-stage analytical pipeline: error identification and categorization, thematic clustering, quantitative scoring, and strategic summarization. The framework is further enhanced with a dual memory system-episodic and semantic-that enables continual learning across executions. Through collaborations with design partners, we demonstrate the framework's practical utility on real-world deployments, before establishing its efficacy against the publicly available TRAIL benchmark. AgentCompass achieves state-of-the-art results on key metrics, while uncovering critical issues missed in human annotations, underscoring its role as a robust, developer-centric tool for reliable monitoring and improvement of agentic systems in production.
Understanding and reasoning about entire software repositories is an essential capability for intelligent software engineering tools. While existing benchmarks such as CoSQA and CodeQA have advanced the field, they predominantly focus on small, self-contained code snippets. These setups fail to capture the complexity of real-world repositories, where effective understanding and reasoning often require navigating multiple files, understanding software architecture, and grounding answers in long-range code dependencies. In this paper, we present SWE-QA, a repository-level code question answering (QA) benchmark designed to facilitate research on automated QA systems in realistic code environments. SWE-QA involves 576 high-quality question-answer pairs spanning diverse categories, including intention understanding, cross-file reasoning, and multi-hop dependency analysis. To construct SWE-QA, we first crawled 77,100 GitHub issues from 11 popular repositories. Based on an analysis of naturally occurring developer questions extracted from these issues, we developed a two-level taxonomy of repository-level questions and constructed a set of seed questions for each category. For each category, we manually curated and validated questions and collected their corresponding answers. As a prototype application, we further develop SWE-QA-Agent, an agentic framework in which LLM agents reason and act to find answers automatically. We evaluate six advanced LLMs on SWE-QA under various context augmentation strategies. Experimental results highlight the promise of LLMs, particularly our SWE-QA-Agent framework, in addressing repository-level QA, while also revealing open challenges and pointing to future research directions.
Human conversation involves language, speech, and visual cues, with each medium providing complementary information. For instance, speech conveys a vibe or tone not fully captured by text alone. While multimodal LLMs focus on generating text responses from diverse inputs, less attention has been paid to generating natural and engaging speech. We propose a human-like agent that generates speech responses based on conversation mood and responsive style information. To achieve this, we build a novel MultiSensory Conversation dataset focused on speech to enable agents to generate natural speech. We then propose a multimodal LLM-based model for generating text responses and voice descriptions, which are used to generate speech covering paralinguistic information. Experimental results demonstrate the effectiveness of utilizing both visual and audio modalities in conversation to generate engaging speech. The source code is available in https://github.com/kimtaesu24/MSenC
Recent studies have shown that carefully designed workflows coordinating large language models(LLMs) significantly enhance task-solving capabilities compared to using a single model. While an increasing number of works focus on autonomous workflow construction, most existing approaches rely solely on historical experience, leading to limitations in efficiency and adaptability. We argue that while historical experience is valuable, workflow construction should also flexibly respond to the unique characteristics of each task. To this end, we propose an a priori dynamic framework for automated workflow construction. Our framework first leverages Q-table learning to optimize the decision space, guiding agent decisions and enabling effective use of historical experience. At the same time, agents evaluate the current task progress and make a priori decisions regarding the next executing agent, allowing the system to proactively select the more suitable workflow structure for each given task. Additionally, we incorporate mechanisms such as cold-start initialization, early stopping, and pruning to further improve system efficiency. Experimental evaluations on four benchmark datasets demonstrate the feasibility and effectiveness of our approach. Compared to state-of-the-art baselines, our method achieves an average improvement of 4.05%, while reducing workflow construction and inference costs to only 30.68%-48.31% of those required by existing methods.
Large language models (LLMs), a recent advance in deep learning and machine intelligence, have manifested astonishing capacities, now considered among the most promising for artificial general intelligence. With human-like capabilities, LLMs have been used to simulate humans and serve as AI assistants across many applications. As a result, great concern has arisen about whether and under what circumstances LLMs think and behave like real human agents. Rationality is among the most important concepts in assessing human behavior, both in thinking (i.e., theoretical rationality) and in taking action (i.e., practical rationality). In this work, we propose the first benchmark for evaluating the omnibus rationality of LLMs, covering a wide range of domains and LLMs. The benchmark includes an easy-to-use toolkit, extensive experimental results, and analysis that illuminates where LLMs converge and diverge from idealized human rationality. We believe the benchmark can serve as a foundational tool for both developers and users of LLMs.
Effort estimation is a crucial activity in agile software development, where teams collaboratively review, discuss, and estimate the effort required to complete user stories in a product backlog. Current practices in agile effort estimation heavily rely on subjective assessments, leading to inaccuracies and inconsistencies in the estimates. While recent machine learning-based methods show promising accuracy, they cannot explain or justify their estimates and lack the capability to interact with human team members. Our paper fills this significant gap by leveraging the powerful capabilities of Large Language Models (LLMs). We propose a novel LLM-based multi-agent framework for agile estimation that not only can produce estimates, but also can coordinate, communicate and discuss with human developers and other agents to reach a consensus. Evaluation results on a real-life dataset show that our approach outperforms state-of-the-art techniques across all evaluation metrics in the majority of the cases. Our human study with software development practitioners also demonstrates an overwhelmingly positive experience in collaborating with our agents in agile effort estimation.
Effective interactive tool use requires agents to master Tool Integrated Reasoning (TIR): a complex process involving multi-turn planning and long-context dialogue management. To train agents for this dynamic process, particularly in multi-modal contexts, we introduce a sandbox environment for reinforcement learning (RL) that supports interleaved speech-text rollouts. Our core strategy, Turn-level Adjudicated Reinforcement Learning (TARL), addresses the challenge of credit assignment in long-horizon tasks by employing a Large Language Model (LLM) as a judge to provide turn-level evaluation. To enhance exploration, we integrate a mixed-task training curriculum with mathematical reasoning problems. This unified approach boosts the task pass rate on the text-based $\tau$-bench by over 6% compared to strong RL baselines. Crucially, we demonstrate our framework's suitability for fine-tuning a multi-modal foundation model for agentic tasks. By training a base multi-modal LLM on interleaved speech-text rollouts, we equip it with tool-use abilities, paving the way for more natural, voice-driven interactive agents.
Large language models (LLMs) are increasingly deployed as task-oriented agents, where success depends on their ability to generate accurate function calls under realistic, multilingual conditions. However, existing agent evaluations largely overlook cultural and linguistic diversity, often relying on monolingual or naively translated benchmarks. We introduce Ticket-Bench, a benchmark for multilingual agent evaluation in task-oriented scenarios. Ticket-Bench simulates the domain of soccer ticket purchases across six major languages: Portuguese, English, Spanish, German, Italian, and French. Using localized teams, cities, and user profiles to provide a higher level of realism. We evaluate a wide range of commercial and open-source LLMs, measuring function-calling accuracy and consistency across languages. Results show that reasoning-oriented models (e.g., GPT-5, Qwen3-235B) dominate performance but still exhibit notable cross-lingual disparities. These findings underscore the need for culturally aware, multilingual benchmarks to guide the development of robust LLM agents.
We present GestOS, a gesture-based operating system for high-level control of heterogeneous robot teams. Unlike prior systems that map gestures to fixed commands or single-agent actions, GestOS interprets hand gestures semantically and dynamically distributes tasks across multiple robots based on their capabilities, current state, and supported instruction sets. The system combines lightweight visual perception with large language model (LLM) reasoning: hand poses are converted into structured textual descriptions, which the LLM uses to infer intent and generate robot-specific commands. A robot selection module ensures that each gesture-triggered task is matched to the most suitable agent in real time. This architecture enables context-aware, adaptive control without requiring explicit user specification of targets or commands. By advancing gesture interaction from recognition to intelligent orchestration, GestOS supports scalable, flexible, and user-friendly collaboration with robotic systems in dynamic environments.