ICLR 2026 论文列表

Automatic tool generation (ATG) has emerged as a key approach to enable the automatic adaptation across diverse tasks within a single generalist agent. Despite their potential, we argue that current ATG agents, often built on reactive paradigms, fail to effectively adapt to realistic environments requiring long-term reasoning and stateful interaction, particularly in office ecosystems. We empirically show that current ATG agents underperform by up to 27.43\%. This performance degradation stems from three fundamental limitations of prevailing agent paradigms: (1) a failure to build a coherent world model from long, partially observable contexts; (2) a memory-less execution model where stateless actions fail to track state evolution during iterative tasks; and (3) a static capability generation model focusing on one-shot tool generation for immediate needs, thereby forcing redundant regeneration for similar steps. To address these fundamental limitations, we propose ROGA, which instantiates a new agent paradigm for long-horizon, stateful environments. ROGA moves beyond simple reactive loops by introducing four foundational algorithmic innovations: (1) \textbf{Active World Modeling}, an iterative process where the agent actively probes the environment to construct its own world model; (2) a \textbf{Persistent Symbolic Memory} that explicitly tracks the state evolution for temporal reasoning; and (3) a \textbf{Dynamic Capability Evolution} model for long-term adaptation and meta-learning on the agent's own capabilities. Comprehensive experiments on widely used benchmarks show that ROGA consistently outperforms existing ATG agents by up to 13.64\%. These results underscore ROGA's potential to advance the ATG paradigm, delivering a practical pathway toward building sustainable generalist agents in realistic environments.

Multi-Agent Path Finding (MAPF) is a representative multi-agent coordination problem, where multiple agents are required to navigate to their respective goals without collisions. Solving MAPF optimally is known to be NP-hard, leading to the adoption of learning-based approaches to alleviate the online computational burden. Prevailing approaches, such as Graph Neural Networks (GNNs), are typically constrained to *pairwise* message passing between agents. However, this limitation leads to suboptimal behaviours and critical issues, such as attention dilution, particularly in dense environments where group (i.e. beyond just two agents) coordination is most critical. Despite the importance of such higher-order interactions, existing approaches have not been able to fully explore them. To address this representational bottleneck, we introduce HMAGAT (Hypergraph Multi-Agent Attention Network), a novel architecture that leverages attentional mechanisms over directed hypergraphs to explicitly capture group dynamics. Empirically, HMAGAT establishes a new state-of-the-art among learning-based MAPF solvers: e.g., despite having just 1M parameters and being trained on 100$\times$ less data, it outperforms the current SoTA 85M parameter model. Through detailed analysis of HMAGAT's attention values, we demonstrate how hypergraph representations mitigate the attention dilution inherent in GNNs and capture complex interactions where pairwise methods fail. Our results illustrate that appropriate inductive biases are often more critical than the training data size or sheer parameter count for multi-agent problems.

The proliferation of Large Language Models (LLMs) has created a diverse ecosystem of models with highly varying performance and costs, necessitating effective query routing to balance performance and expense. Current routing systems often rely on a centralized external router trained on a fixed set of LLMs, making them inflexible and prone to poor performance since the small router can not fully understand the knowledge boundaries of different LLMs. We introduce DiSRouter (Distributed Self-Router), a novel paradigm that shifts from centralized control to distributed routing. In DiSRouter, a query traverses a network of LLM agents, each independently deciding whether to answer or route to other agents based on its own self-awareness—its ability to judge its competence. This distributed design offers superior flexibility, scalability, and generalizability. To enable this, we propose a two-stage Self-Awareness Training pipeline that enhances each LLM's self-awareness. Extensive experiments demonstrate that DiSRouter significantly outperforms existing routing methods in utility across various scenarios, effectively distinguishes between easy and hard queries, and shows strong generalization to out-of-domain tasks. Our work validates that leveraging an LLM's intrinsic self-awareness is more effective than external assessment, paving the way for more modular and efficient multi-agent systems.

Recently, Large Language Models (LLMs) have been applied to scientific equation discovery, leveraging their embedded scientific knowledge for hypothesis generation. However, current methods typically confine LLMs to the role of an equation proposer within search algorithms like genetic programming. In this paper, we present SR-Scientist, a framework that elevates the LLM from a simple equation proposer to an autonomous AI scientist that writes code to analyze data, implements the equation as code, submits it for evaluation, and optimizes the equation based on experimental feedback. Specifically, we wrap the code interpreter into a set of tools for data analysis and equation evaluation. The agent is instructed to optimize the equation by utilizing these tools over a long horizon with minimal human-defined pipelines. Empirical results show that SR-Scientist outperforms baseline methods by an absolute margin of 6\% to 35\% on datasets covering four science disciplines. Additionally, we demonstrate our method's robustness to noise, the generalization of the discovered equations to out-of-domain data, and their symbolic accuracy. Furthermore, we develop an end-to-end reinforcement learning framework to enhance the agent's capabilities.

Academic posters are vital for scholarly communication, yet their manual creation is time-consuming. However, automated academic poster generation faces significant challenges in preserving intricate scientific details and achieving effective visual-textual integration. Existing approaches often struggle with semantic richness, structural nuances, and lack standardized benchmarks for evaluating generated academic posters comprehensively. To address these limitations, we introduce P2P, the first flexible, LLM-based multi-agent framework that generates high-quality, HTML-rendered academic posters directly from research papers. P2P employs three specialized agents—for visual element processing, content generation, and final poster assembly—each integrated with dedicated checker modules to enable iterative refinement and ensure output quality. To foster advancements and rigorous evaluation in this domain, we argue that generated posters must be assessed from two complementary perspectives: objective fidelity and subjective quality. So we establish P2Peval, a comprehensive benchmark featuring 1738 checklist items and a dual evaluation methodology (Fine-Grained and Universal). Our Fine-Grained Evaluation uses human-annotated checklists to objectively measure the faithful preservation of verifiable content from the source paper. Concurrently, our Universal Evaluation captures subjective, holistic quality by training a model to align with human aesthetic preferences across key design principles. We evaluate a total of 35 models. To power these advancements, we also release P2Pinstruct, the first large-scale instruction dataset comprising over 30,000 high-quality examples tailored for the academic paper-to-poster generation task. Furthermore, our contributions aim to streamline research dissemination while offering a principled blueprint for evaluating complex, creative AI-generated artifacts. The code is on the anonymous GitHub, https://anonymous.4open.science/r/c-163Canonymous/README.md.

For multi-stage recommenders in industry, a user request would first trigger a simple and efficient retriever module that selects and ranks a list of relevant items, then the recommender calls a slower but more sophisticated reranking model that refines the item list exposure to the user. To consistently optimize the two-stage retrieval reranking framework, most efforts have focused on learning reranker-aware retrievers. In contrast, there has been limited work on how to achieve a retriever-aware reranker. In this work, we provide evidence that the retriever scores from the previous stage are informative signals that have been underexplored. Specifically, we first empirically show that the reranking task under the two-stage framework is naturally a noise reduction problem on the retriever scores, and theoretically show the limitations of naive utilization techniques of the retriever scores. Following this notion, we derive an adversarial framework DNR that associates the denoising reranker with a carefully designed noise generation module. The resulting DNR solution extends the conventional score error minimization loss with three augmented objectives, including: 1) a denoising objective that aims to denoise the noisy retriever scores to align with the user feedback; 2) an adversarial retriever score generation objective that improves the exploration in the retriever score space; and 3) a distribution regularization term that aims to align the distribution of generated noisy retriever scores with the real ones. We conduct extensive experiments on three public datasets and an industrial recommender system, together with analytical support, to validate the effectiveness of the proposed DNR.

Large language model (LLM)-based multi-agent systems have shown strong capabilities in tasks such as code generation and collaborative reasoning. However, the effectiveness and robustness of these systems critically depend on their communication topology, which is often fixed or statically learned, ignoring real-world dynamics such as model upgrades, API (or tool) changes, or knowledge source variability. To address this limitation, we propose CARD (Conditional Agentic Graph Designer), a conditional graph-generation framework that instantiates AMACP, a protocol for adaptive multi-agent communication. CARD explicitly incorporates dynamic environmental signals into graph construction, enabling topology adaptation at both training and runtime. Through a conditional variational graph encoder and environment-aware optimization, CARD produces communication structures that are both effective and resilient to shifts in model capability or resource availability. Empirical results on HumanEval, MATH, and MMLU demonstrate that CARD consistently outperforms static and prompt-based baselines, achieving higher accuracy and robustness across diverse conditions. The source code is available at: \url{https://anonymous.4open.science/r/agentgraph-FF9A}.

A fundamental limitation of current AI agents is their inability to learn complex skills on the fly at test time, often behaving like “clever but clueless interns” in novel environments. This severely limits their practical utility. To systematically measure and drive progress on this challenge, we first introduce the Jericho Test-Time Learning (J-TTL) benchmark. J-TTL is a new evaluation setup where an agent must play the same game for several consecutive episodes, attempting to improve its performance from one episode to the next. On J-TTL, we find that existing adaptation methods like reflection, memory, or reinforcement learning struggle. To address the challenges posed by our benchmark, we present EvoTest, an evolutionary test-time learning framework that improves an agent without any fine-tuning or gradients—by evolving the entire agentic system after every episode. EvoTest has two roles: the Actor Agent, which plays the game, and the Evolver Agent, which analyzes the episode transcript to propose a revised configuration for the next run. This configuration rewrites the prompt, updates memory by logging effective state–action choices, tunes hyperparameters, and learns the tool-use routines. On our J-TTL benchmark, EvoTest consistently increases performance, outperforming not only reflection and memory-only baselines but also more complex online fine-tuning methods. Notably, our method is the only one capable of winning two games (Detective and Library), while all baselines fail to win any.

With the increasing demand for step-wise, cross-modal, and knowledge-grounded reasoning, multimodal large language models (MLLMs) are evolving beyond the traditional fixed retrieve-then-generate paradigm toward more sophisticated agentic multimodal retrieval-augmented generation (MM-RAG). Existing benchmarks, however, mainly focus on simplified QA with short retrieval chains, leaving adaptive planning and multimodal reasoning underexplored. We present MC-Search, the first benchmark for agentic MM-RAG with long, step-wise annotated reasoning chains spanning five representative reasoning structures. Each example specifies sub-questions, retrieval modalities, supporting facts, and intermediate answers, with fidelity ensured by HAVE (Hop-wise Attribution and Verification of Evidence), resulting in 3,333 high-quality examples averaging 3.7 hops. Beyond answer accuracy, MC-Search introduces new process-level metrics for reasoning quality, stepwise retrieval and planning accuracy. By developing a unified agentic MM-RAG pipeline, we benchmark six leading MLLMs and reveal systematic issues such as over- and under-retrieval and modality-misaligned planning. Finally, we introduce Search-Align, a process-supervised fine-tuning framework leveraging verified reasoning chains, showing that our data not only enables faithful evaluation but also improves planning and retrieval fidelity in open-source MLLMs.

CodeQL is a powerful static analysis engine that represents programs’ abstract syntax trees as databases that can be queried to detect security vulnerabilities. While CodeQL supports expressive interprocedural dataflow queries, the coverage and precision of its existing security queries remain limited, and writing new queries is challenging even for experts. Automatically synthesizing CodeQL queries from known vulnerabilities (CVEs) can provide fine-grained vulnerability signatures, enabling both improved detection and systematic variant analysis. We present QLCoder, an agentic framework for synthesizing CodeQL queries from known CVE descriptions. QLCoder leverages the Model Context Protocol (MCP) for agentic tool use, integrates abstract syntax tree guidance, and incorporates CodeQL’s language infrastructure and documentation into the synthesis loop. A key challenge is that state-of-the-art large language models hallucinate deprecated CodeQL syntax due to limited training data and outdated knowledge. QLCoder addresses this by combining contextual engineering, iterative query feedback, and structured tool interaction to reliably generate executable, up-to-date queries.

Code agents have gained widespread adoption due to their strong code generation capabilities and integration with code interpreters, enabling dynamic execution, debugging, and interactive programming capabilities. While these advancements have streamlined complex workflows, they have also introduced critical safety and security risks. Current static safety benchmarks and red-teaming tools are inadequate for identifying emerging real-world risky scenarios, as they fail to cover certain boundary conditions, such as the combined effects of different jailbreak tools. In this work, we propose RedCodeAgent, the first automated red-teaming agent designed to systematically uncover vulnerabilities in diverse code agents. With an adaptive memory module, RedCodeAgent can leverage existing jailbreak knowledge, dynamically select the most effective red-teaming tools and tool combinations in a tailored toolbox for a given input query, thus identifying vulnerabilities that might otherwise be overlooked. For reliable evaluation, we develop simulated sandbox environments to additionally evaluate the execution results of code agents, mitigating potential biases of LLM-based judges that only rely on static code. Through extensive evaluations across multiple state-of-the-art code agents, diverse risky scenarios, and various programming languages, RedCodeAgent consistently outperforms existing red-teaming methods, achieving higher attack success rates and lower rejection rates with high efficiency. We further validate RedCodeAgent on real-world code assistants, e.g., Cursor and Codeium, exposing previously unidentified security risks. By automating and optimizing red-teaming processes, RedCodeAgent enables scalable, adaptive, and effective safety assessments of code agents.

Model-based reinforcement learning (MBRL) has achieved remarkable success in robotics due to its high sample efficiency and planning capability. However, extending MBRL to physical multi-robot cooperation remains challenging due to the complexity of joint dynamics. To address this challenge, we propose the Sequential World Model (**SeqWM**), a novel framework that integrates the sequential paradigm into multi-robot MBRL. SeqWM employs independent, autoregressive agent-wise world models to represent joint dynamics, where each agent generates its future trajectory and plans its actions based on the predictions of its predecessors. This design lowers modeling complexity and enables the emergence of advanced cooperative behaviors through explicit intention sharing. Experiments on Bi-DexHands and Multi-Quadruped demonstrate that SeqWM outperforms existing state-of-the-art model-based and model-free baselines in both overall performance and sample efficiency, while exhibiting advanced cooperative behaviors such as predictive adaptation, temporal alignment, and role division. Furthermore, SeqWM has been successfully deployed on physical quadruped robots, validating its effectiveness in real-world multi-robot systems. Demos and code are available at: https://github.com/zhaozijie2022/seqwm

LLM-based web agents show immense promise for information seeking, yet their effectiveness on long-horizon tasks is hindered by a fundamental trade-off in context management. Prevailing ReAct-based agents suffer from context saturation as they accumulate noisy, raw histories, while methods that fixedly summarize the full history at each step risk the irreversible loss of critical details. Addressing these, we introduce AgentFold, a novel agent paradigm inspired by the human cognitive process of retrospective consolidation. AgentFold treats its context as a dynamic cognitive workspace to be actively sculpted, rather than a passive log to be filled. At each step, it learns to execute a folding operation, which manages its historical trajectory at multiple scales: it can perform granular condensations to preserve vital, fine-grained details, or deep consolidations to abstract away entire multi-step sub-tasks. The results on prominent benchmarks are striking: our AgentFold-30B-A3B agent achieves 36.2% on BrowseComp and 47.3% on BrowseComp-ZH. Notably, this performance not only surpasses or matches open-source models of a dramatically larger scale, such as the GLM-4.5-355B-A32B and the DeepSeek-V3.1-671B-A37B, but also surpasses leading proprietary agents like OpenAI's o4-mini. Model will be open-sourced.

Obtaining high-quality outputs from Large Language Models (LLMs) often depends upon the choice of a sampling-based decoding strategy to probabilistically choose the next token at each generation step. While a variety of such sampling methods have been proposed, their performance can be sensitive to the selection of hyperparameters which may require different settings depending upon the generation task and temperature configuration. In this work, we introduce $p\textrm{-less}$ sampling: an information-theoretic approach to sampling which dynamically sets a truncation threshold at each decoding step based on the entire token probability distribution. Unlike existing methods, $p\textrm{-less}$ sampling has no hyperparameters and consistently produces high-quality outputs as temperature increases. We provide theoretical perspectives on $p$-less sampling to ground our proposed method and conduct experiments to empirically validate its effectiveness across a range of math, logical reasoning, and creative writing tasks. Our results demonstrate how $p\textrm{-less}$ sampling consistently outperforms existing sampling approaches while exhibiting much less degradation in text quality at higher temperature values. We further show how $p$-less achieves greater inference-time efficiency than alternative methods through lower average token sampling times and shorter generation lengths, without sacrificing accuracy. Finally, we provide analyses to highlight the benefits of $p\textrm{-less}$ through qualitative examples, case studies, and diversity assessments.

One of the fundamental problems in digital agents is their lack of understanding of their environment. For instance, a web browsing agent may get lost in unfamiliar websites, uncertain what pages must be visited to achieve its goals. To address this, we propose Go-Browse, a method for automatically collecting diverse and realistic web agent data at scale through structured exploration of web environments. Go-Browse achieves efficient exploration by framing data collection as a graph search, enabling reuse of information across exploration episodes. We instantiate our method on the WebArena benchmark, collecting a dataset of 10K successful task-solving trajectories and 40K interaction steps across 100 URLs. Fine-tuning a 7B parameter language model on this dataset achieves a success rate of 21.7% on the WebArena benchmark, beating GPT-4o mini by 2.4% and exceeding current state-of-the-art results for sub-10B parameter models by 2.9%.

As artificial intelligence (AI) systems approach and surpass expert human performance across a broad range of tasks, obtaining high-quality human supervision for evaluation and training becomes increasingly challenging. Our focus is on tasks that require deep knowledge and skills of multiple domains. Unfortunately, even the best human experts are knowledgeable only in a single narrow area, and will not be able to evaluate the correctness of advanced AI systems on such superhuman tasks. However, based on their narrow expertise, humans may provide a weak signal, i.e., a *complementary label* indicating an option that is incorrect. For example, a cardiologist could state that ``this is not related to cardiology,'' even if they cannot identify the true disease. Based on this weak signal, we propose a scalable oversight framework that enables us to evaluate frontier AI systems without the need to prepare the ground truth. We derive an *unbiased* estimator of top-1 accuracy from complementary labels and quantify how many complementary labels are needed to match the variance of ordinary labels. We further introduce two estimators to combine scarce ordinary labels with abundant complementary labels. We provide finite-sample deviation guarantees for both complementary-only and the mixed estimators. Empirically, we show that we can evaluate the output of large language models without the ground truth, if we have complementary labels. We further show that we can train an AI system with such weak signals: we show how we can design an agentic AI system automatically that can perform better by these partitioned human supervision.

Multi-Agent System (MAS) powered by Visual Language Models (VLMs) enables challenging tasks but suffers from a novel failure term, multi-agent visual hallucination snowballing, where hallucinations are seeded in a single agent and amplified by following ones due to the over-reliance on textual flow to relay visual information. Through turn-, layer-, and token-wise attention analyses, we provide detailed insights into the essence of hallucination snowballing regarding the reduction of visual attention allocation. It leads us to identify a subset of vision tokens with a unimodal attention peak in middle layers that best preserve visual evidence but gradually diminish in deeper agent turns, resulting in the visual hallucination snowballing in MAS. Thus, we propose ViF, a lightweight, plug-and-play mitigation paradigm that relays inter-agent messages with Visual Flow powered by the selected visual relay tokens and applies attention reallocation to amplify this pattern. The experiment results demonstrate that our method markedly reduces hallucination snowballing, consistently improving the performance across eight benchmarks based on four common MAS structures and ten base models. The implementation source code will be made publicly available.

Multi-Agent System (MAS) and Reinforcement Learning (RL) are both widely adopted to improve large language model (LLM) agentic performance. MAS strengthens task-specialized performance via role-based orchestration; RL leverages environment rewards to train stronger policies, such as Group Relative Policy Optimization (GRPO)-style optimization. Yet applying on-policy RL training to MAS is underexplored. While promising, it poses several challenges. On the algorithm side, Standard GRPO grouping assumptions fail in MAS because prompts differ by role and turn. On the system side, the training system needs to support MAS-workflow-based rollouts and on-policy updates for both single and multiple policy models. To address these issues, we introduce AT-GRPO, consisting of (i) an Agent- and Turn-wise grouped RL algorithm tailored for MAS and (ii) a system to support both single-policy and multi-policy training. Across game, plan, coding, and math tasks, AT-GRPO demonstrates substantial performance gains across diverse domains. Especially on long-horizon planning tasks, AT-GRPO boosts accuracy from a 14.0–47.0% single-agent RL baseline to 96.0–99.5%. Furthermore, it improves reasoning performance, with an average gain of 3.87–7.62% on coding and 9.0-17.93% on math. The code are available at https://github.com/pettingllms-ai/PettingLLMs.

VLM-based mobile agents are increasingly popular due to their capabilities to interact with smartphone GUIs and XML-structured texts and to complete daily tasks. However, existing online benchmarks fail to obtain stable critical reward signals under dynamic environmental changes, and neglect the influence of noise components and interactive instructions. Offline benchmarks evaluate the agents through single-path trajectories, which stand in contrast to the inherently multi-solution characteristics of GUI tasks. To address these limitations, we introduce SMAN-Bench, a benchmark designed to evaluate agents under Single-path, Multi-path, Ambiguous, and Noisy task settings. We employ a slot-based instruction generation method to match templates with GUI trajectories from an existing, graph-structured, unlabeled mobile corpus. SMAN-Bench includes a common task split, with offline multi-path evaluation to assess the agent’s ability to obtain step rewards during task execution. It contains a noisy split based on pop-ups and ad apps, and a contaminated split named AITZ-Noise to simulate a realistic noisy environment. Furthermore, an ambiguous instruction split with preset Q&A interactions is released to evaluate the agent’s proactive interaction capabilities. Our evaluation covers mobile agent frameworks like AppAgent-v1, Mobile-Agent-v2, and Mobile-Agent-E, and includes both open-source and closed-source mobile fundamental models, as well as several multimodal reasoning models.