论文

本文提出 Toolformer，训练其决定调用哪些 API、何时调用、传入什么参数，以及如何将结果最佳地融入后续 token 预测，在多种下游任务上显著提升零样本性能。This paper introduces Toolformer, a model trained to decide which APIs to call, when to call them, what arguments to pass, and how to best incorporate the results into future token prediction, which achieves substantially improved zero-shot performance across a variety of downstream tasks.

本文提出 SWE-Pruner Pro，在 Agent 内部直接对工具输出进行剪枝，通过一个小型 head 将 Agent 自身的内部表征转化为针对每一行的 keep-or-prune 标签，并采用以每段工具输出行数为键的长度感知嵌入。SWE-Pruner Pro is proposed, which prunes tool outputs directly inside the agent, with a small head turns the agent's own internal representations into a keep-or-prune label for each line, with a length-aware embedding keyed to each tool output's line count.

实证结果表明，分层防御栈对大多数攻击单元有效，但仍存在一小部分残余攻击面在操作系统层面本质上不可区分，这表明针对新出现的自我状态攻击类别，需要重新审视操作系统级防御。The empirical results show that a layered defense stack is effective on most attack cells while a small residual attack surface remains structurally indistinguishable at the OS level, suggesting that against the newly established class of self-state attacks, OS-level defense needs to be reconsidered.

这些结果识别出关键决策支持的遗漏是 Qwen 主设定中的因果机制，并在多个模型族中得到印证，揭示了支持、损失和解码层面干预之间的克制-能力权衡关系。These results identify decision-critical support omission as a causal mechanism in the primary Qwen setting and support it across model families, tracing a restraint-capability trade-off among support-, loss-, and decoding-level interventions.

本文提出 DeepSearch-Evolve，一个面向 web agent 的自蒸馏框架，基于 DeepSearch-World——一个具备可复现搜索与页面读取工具的确定性、可验证环境——从而实现长程 web agent 的可扩展自演化。DeepSearch-Evolve is presented, a self-distillation framework for web agents built on DeepSearch-World, a deterministic and verifiable environment with reproducible search and page-reading tools that enables scalable self-evolution for long-horizon web agents.

DeepDebug 在两个测试的开源权重 backbone 上均取得了所评估方法中最高的严格归因准确率，在 qwen3.5-9b 上达到 28.8% 的精确 agent 与步骤准确率，而最强的单遍 baseline 为 21.7%。DeepDebug achieves the best strict attribution accuracy among the evaluated methods on both tested open-weight backbones, reaching 28.8 percent exact agent-and-step accuracy on qwen3.5-9b versus 21.7 percent for the strongest single-pass baseline.

论证了经济层面的依据：天真的上下文累积会使 token 成本随对话长度呈二次增长，粗糙的摘要以线性成本换取准确率的断崖式下降，唯有经过验证的压缩才能以线性成本保持保真度。The economic case is made: naive context accumulation grows token cost quadratically in conversation length, crude summarization buys linear cost at the price of an accuracy cliff, and only validated compaction achieves linear cost with preserved fidelity.

与经典强化学习基线相比，从试错经验中进行上下文学习并随后进行经验蒸馏（Experience Distillation），以至少 9.6× 更少的环境样本达到了相当的性能。Compared with classical reinforcement-learning baselines, in-context learning from trial-and-error experience followed by Experience Distillation matches their performance with at least \(9.6\times\) fewer environment samples.

ReOPD 将昂贵的 Agent-环境交互转化为可复用的离线资源，实现跨工具、任务和环境的可扩展蒸馏，在学生训练期间保持或提升 OPD 级别的准确率，零次工具调用，并且每次 rollout 至少比 OPD 快 4×。ReOPD turns expensive agent-environment interaction into a reusable offline resource, enabling scalable distillation across tools, tasks, and environments and preserves or improves OPD-level accuracy, uses zero tool calls during student training, and is at least 4$\times faster per rollout than OPD.

实证研究表明 AutoGen 框架在多个示例应用中有效，应用领域涵盖数学、编码、问答、运筹学、在线决策、娱乐等。Empirical studies demonstrate the effectiveness of the AutoGen framework in many example applications, with domains ranging from mathematics, coding, question answering, operations research, online decision-making, entertainment, etc.

该工作主张将研究构想视为多目标的联合问题，并将其建模为 Quality-Diversity (QD) 搜索；同时提出 IDEAgent，一个通过 lineage 管理思路演化的 multi-agent 框架。This work argues that research ideation should be treated as a conjunction of both objectives and framed as a Quality-Diversity (QD) search, and introduces IDEAgent, a multi-agent framework that manages the evolution of ideas through lineages.

Molt 是一个 PyTorch-native 的训练框架，旨在压低上述开销：代码库精简清晰，研究者能完整装进脑中，AI 编程助手也能完整阅读和推理，从而可以端到端地追溯和修改算法流程。Molt is a PyTorch-native training framework built to keep that cost small: a codebase compact and clean enough for a researcher to hold in their head, and for an AI coding assistant to read and reason about in its entirety, so the algorithm flow can be traced and changed end to end.

本文提出一种面向由一个或多个 agent 构建的自动化研究系统的词汇表，使其设计选择更易于描述与比较，从而将结构性设计问题——例如 agent 应在何时通信、获得或失去某项能力，或在多次运行间传递信息——转化为可测试的选择。A vocabulary for automated research systems built from one or more agents to make their design choices easier to describe and compare, which turns structural design questions, such as when agents should communicate, gain or lose a capability, or carry information across runs, into testable choices.

本文提出 RARG（Relevance-Aware RipGrep Search Agent），将相关性转化为 corpus 交互的执行先验，并证明相关性感知交互可带来更快且更可靠的搜索收敛。The Relevance-Aware RipGrep Search Agent (RARG) is introduced, which turns relevance into an execution prior for corpus interaction, and demonstrates that relevance-aware interaction enables faster and more reliable search convergence.

本文综述了该边界上的五类漏洞：多步攻击链、与沙箱边界冲突的目标、供应链与凭据暴露、持续性的 command-and-control，以及自动化行动的速度。This review synthesizes five vulnerability classes at that boundary: multi-step offensive chains, objectives that conflict with sandbox boundaries, supply-chain and credential exposure, persistent command-and-control, and the speed of automated action.

这些结果共同支持多视角的仓库上下文服务，在仓库上下文全生命周期中提供显式的、面向操作的有效性边界以及质量-成本前沿。Together, these results support multi-view repository-context serving with explicit, operation-specific validity boundaries with quality-cost frontiers across the repository-context lifecycle.

介绍 VisualPatchWorld，将世界动态表示为代码，先通过短时主动探查选择定性动力学形式，再通过最小化多步预测误差，从记录的状态-动作轨迹中拟合该形式的自由参数。VisualPatchWorld is introduced, which represents world dynamics as code and first selects a qualitative dynamical form with short active probes, then fits that form's free parameters from recorded state-action traces by minimizing multi-step prediction error.

本文提出 Metis，首个 memory foundation model 原型，赋予 foundation model 原生记忆能力，并表明原生记忆在架构、端到端优化和效率方面具有优势。This paper proposes Metis, the first prototype of memory foundation models, which empower foundation models with native memory capabilities and shows that native memory offers advantages in architecture, end-to-end optimization, and efficiency.

跨任务的测试时扩展实验表明，SkillRise 跨任务复用可迁移的 skill，而非受益于对同一任务的重复采样，在保持强性能的同时显著降低多阶段 skill 学习流水线的运行时开销。Scaling at test time across tasks suggests that SkillRise reuses transferable skills across tasks rather than benefiting from repeated sampling of the same task, and retains strong performance while substantially reducing the runtime overhead of skill learning pipelines with multiple stages.

提出 CAST（Credit Assignment from Solver Teachers），将游戏求解器状态价值的变化转换为求解器优势，并将其作为 turn 级信号注入 RLVR，在 ALFWorld 和 WebShop 上取得最高的平均 zero-shot 性能。CAST (Credit Assignment from Solver Teachers), which converts value changes in a game solver's state value into solver advantages and injects them into RLVR as turn-level signals and achieves the highest average zero-shot performance on ALFWorld and WebShop.

本文贡献包括：圣训学概念到多 agent 流水线的形式化映射、实现 claim 链和分级 narrator 注册表的关系模式、结合链等级与内容批评的决策矩阵，以及对真实物理教材中 20,000 条 claim 的评估。A formal mapping from hadith-science concepts to multi-agent pipelines, a relational schema implementing claim chains and a graded narrator registry, a decision matrix combining chain grade with content criticism, and an evaluation on 20,000 claims from real physics textbooks are contributed.

描述一个 memory store：agent 本地的 Neo4j 属性图，增强 HNSW 向量索引，并采用完整的双时态数据模型，支持时间点语义检索而无需物理覆盖历史。A memory store is described: an agent-local Neo4j property graph augmented with HNSW vector indexes and a full bitemporal data model that supports point-in-time semantic retrieval without physically overwriting history.

提出 MindForge，一个自动化 pipeline，将开源命令行程序转换为无源码环境（仅暴露编译后的可执行参考文件和文档），在全部 7 个未见软件工程 benchmark（涵盖长链路仓库生成与翻译）上一致优于基座模型。MindForge is introduced, an automated pipeline that converts open-source command-line programs into source-free environments that expose only a compiled reference executable and its documentation that consistently improves over the base model across all seven unseen software engineering benchmarks, spanning long-horizon repository generation and translation.

Voice Memory，一个面向 agentic 语音识别的纯推理方案：流式推理时，冻结 corrector 读取单一 per-domain memory，逐 utterance 决定是否作用于假设或弃权并保留 1-best，跨 corrector 族可迁移，推理路径不增加任何参数。Voice Memory, a inference-only scheme for agentic speech recognition: at stream time, a frozen corrector reads a single per-domain memory and decides per utterance whether to act on the hypothesis or abstain and keep the 1-best, and transfers across corrector families and adds zero parameters to the inference path.

提出 LedgerMind，配套三层 Grounding Protocol、一个按问题复杂度匹配推理深度的 Adaptive Dual-Path Dispatcher，以及一个具备形式化 provenance 非放大保证的事件触发验证与修复引擎，同时提升答案准确率与轨迹级忠实度。LedgerMind is introduced, augmented by a Three-Layer Grounding Protocol, an Adaptive Dual-Path Dispatcher that matches reasoning depth to question complexity, and an Event-Triggered Verification-and-Repair engine with a formal provenance non-amplification guarantee that improves both answer accuracy and trajectory-level faithfulness.

记忆是长时程 LLM 智能体的核心，但现有记忆系统主要保存交互内容，而未建模哪些智能体在何种条件下可信。这一局限在多智能体系统中尤为关键，因为中心模型可能无法直接验证来自对等方、看似合理或相关的响应。我们提出 Σ-Mem，一种在线可靠性记忆，记录单个对等方的历史能力证据以及跨对等集的对等关系证据。两种证据均以实对称状态形式维护，并基于后Memory is central to long-horizon LLM agents, yet existing memory systems primarily preserve interaction content rather than modeling which agents can be trusted and under what conditions. This limitation is particularly important in multi-agent systems, where a central model may be unable to directly verify plausible or correlated peer responses. We introduce Σ-Mem, an online reliability memory that records historical competence evidence for individual peers and peer relationship evidence across the peer set. Both forms of evidence are maintained as real symmetric states and updated from post

提出 MisKnow-Agent，一个受控评测框架，通过可控的权威线索与来源风格构造支撑人工审核结论的任务文档，并采用报告级 false-conclusion 采纳率（仅统计认可错误结论的报告），基于三种 backbone LLM 评估 DeerFlow 与 WebThinker。MisKnow-Agent is introduced, a controlled evaluation framework that constructs task-specific documents supporting manually audited false conclusions with controlled authority cues and source styles that evaluates DeerFlow and WebThinker with three backbone LLMs using a report-level false-conclusion adoption rate that counts only reports endorsing the false conclusion.

研究表明，领域特定的 orchestration 可使缺乏专业团队的实验室也能实现可审计的预处理，并为其他科学领域的 AI agent 提供了可借鉴的设计原则。The results indicate that domain-specific orchestration can bring auditable preprocessing within reach of laboratories lacking dedicated expertise, illustrating design principles applicable to AI agents in other scientific domains.

本工作识别出记忆溯源洗白现象：基于LLM的记忆整合过程中，外部观察可能被改写为看似用户历史或工作流支持的内容，在保留动作触发的同时，抹去本应限制其权威性的低可信度来源。This work identifies memory provenance laundering: during LLM-based memory consolidation, an external observation may be rewritten as apparent user history or workflow support, preserving an action trigger while erasing the low-trust source that should limit its authority.

提出QWorld，用分位数-分位数匹配目标替代EP，直接将投影后的潜在样本与秩匹配的高斯分位数对齐，从而在尾部保持有效的修正梯度。QWorld is proposed, which replaces EP with a quantile-quantile matching objective that directly aligns projected latent samples with rank-matched Gaussian quantiles, thereby maintaining effective corrective gradients in the tails.

世界模型为规划和行动提供了预测性基础，但现有建模方式仅回答物理层面的问题：它是什么/在哪里，以及将如何演变。然而，人类行为由隐藏的心理状态驱动（一个人相信什么、想要什么、意图做什么、感受如何，以及认为在社会上何为可接受），因此仅追踪物理场景而忽略每个智能体所知与所信内容的模型，会对看起来正确的场景预测出错误的行动。我们将心理世界建模（MWM）形式化为一个通用理论框架，将心理变量作为世界模型的核心组成部分。World models enable a predictive substrate for planning and action, yet existing formulations merely answer a physical question: what/where it is, and how will it evolve. Human behavior, however, is driven by hidden mental state (what a person believes, wants, intends, feels, and considers socially permissible), so a model that tracks the physical scene but not what each agent knows and believes about it predicts the wrong action for the right-looking scene. We formulate Mental World Modeling (MWM), a generic theoretical framework that makes mental variables core components of a world model ra

提出EMBL AI Librarian，一个升级Europe PMC接口的知识层，面向AI agent，提升多项任务表现：文献综合、claim验证、开放域问答，以及下游生物学任务如protocol问题与序列操作。EMBL AI Librarian is introduced, a knowledge layer that upgrades the Europe PMC interface for AI agents that improves performance across a range of tasks: literature synthesis, claim verification, open-domain question answering, and downstream biology tasks such as protocol questions and sequence manipulation.

将长视野执行重新表述为任务状态管理问题，提出LongHorizon-Harness，在执行外部显式维护任务状态，并仅用从环境中独立验证的事实更新它。This work reformulate long-horizon execution as a task-state management problem and proposes LongHorizon-Harness, which maintains the task state explicitly outside execution and updates it only with facts independently verified from the environment.

本文将技能生成建模为序列编辑过程，把技能构建分解为可单独评估的编辑，并提出一种新颖的回滚奖励，通过在锚定查询上对比原始技能与编辑后技能的下游执行效果来评估每次编辑。This work forms skill generation as a sequential editing process that decomposes skill construction into individually evaluable edits, and introduces a novel rollback reward that evaluates each edit by comparing downstream execution under the original and edited skills on an anchored query.

结果表明结构化 agent 记忆无需生成过去的中间表示，Zero-Mem 在消除记忆操作中 LLM 调用与 LLM-token 消耗的同时取得具有竞争力的性能The results show that structured agent memory need not generate an intermediate representation of the past, and Zero-Mem achieves competitive performance while eliminating LLM calls and LLM-token consumption from memory operations.

结果是面向稀疏 event-KV 服务的记忆契约：写入什么、落在何处、源消失后什么得以保留The result is a memory contract for sparse event-KV serving: what to write, where it lands, and what survives once the source is gone.