今日必看

MaD-RSI: Model-as-Data for Recursive Self-Improvement of Agent Harnesses

arXiv:2610.28201v1 Announce Type: new Abstract: Recursive self-improvement (RSI) of agent harnesses has emerged as an effective approach to enhancing the capabilities of large language model (LLM) agents. By searching the harness space for a specific LLM on a given dataset, RSI c

Navalia: Advancing Cowork Agents through Verifiable Task Synthesis and Long-Horizon Post-Training

arXiv:2610.26887v1 Announce Type: new Abstract: Large language model (LLM) agents are evolving from conversational assistants into Cowork Agents capable of autonomous planning, tool use, and complex task execution. Financial analysis requires these agents to integrate cross-docum

REM: What to Store, When to Recall, What to Forget in Experience Memory for Coding Agents

arXiv:2610.27377v1 Announce Type: new Abstract: Large language model (LLM) agents now autonomously complete long-horizon tasks, such as repository-level coding, through hundreds of think-act steps, yet the experience each run produces is rarely reused on the next. Prior memory sy

RITE: Evaluating End-to-End Resilience to Silent Tool Errors in E-Commerce Agentic Systems

arXiv:2610.27777v1 Announce Type: new Abstract: Tool-using LLM agents increasingly answer e-commerce queries through multi-step tool interactions. The expected output of a tool call can shift between agent development and execution, allowing a plausible but incorrect value to be

mini-Lean-Agent: A Coding-Agent Baseline for Lean

从邮箱客户端开发反思 AI Agent 与 Vibe Coding

📌 一句话摘要 作者通过尝试使用 Grok Bot 管理邮箱，反思认为 AI 时代应将任务交给 Agent 而非追求开发 Vibe coding 小工具。 📝 详细摘要 作者在开发个人邮箱客户端时尝试使用 Grok Bot 建立 Mail Bot，通过设定规则成功实现了邮箱管理。基于此体验，作者反思认为之前的软件开发思维有误，主张在 AI 时代应当直接利用 Agent 处理事情，而非通过 Vibe coding 开发传统意义上的软件小工具。 💡 主要观点 从软件思维转向 Agent 思维 传统的软件开发（包括 Vibe coding）仍是在构建一个工具，

VEOcc++: Learning Persistent Voxel Feature Maps for Embodied Online Semantic Occupancy Mapping and Offline Refinement

arXiv:2610.27704v1 Announce Type: new Abstract: Semantic occupancy mapping supports embodied perception during exploration and detailed scene understanding afterward, giving rise to two complementary requirements: online incremental mapping for maintaining an up-to-date global ma

WildAIGC-Bench: Can AI Image Detectors Handle the Wild? A Benchmark of Generated and Edited Images in Online Circulation

arXiv:2610.28087v1 Announce Type: new Abstract: AI-generated images linked to news events and online claims can mislead audiences, enable fraud, or cause public controversy. We introduce WildAIGC-Bench, built through recurring collection across news websites, fact-checking source

AutoRedClaw: Automated Red-Teaming for Claw-Style Agents

arXiv:2610.27244v1 Announce Type: cross Abstract: Agent systems are shifting from task-scoped tool-use agents to persistent, personalized assistants such as OpenClaw. These claw-style agents maintain configuration files (e.g., AGENTS.md), long-term memory, and reusable skills. On

Which Depth Estimator for Jetson AGX Orin? A 2026 Benchmark of Monocular and Stereo Methods

arXiv:2610.27199v1 Announce Type: cross Abstract: The Jetson AGX Orin is a common computer for robots and vehicles, yet measured accuracy, speed and energy of current depth estimators on it are hard to find. We benchmark nine model families as of October 2026: the monocular Depth

Embodied Playground: A Long-Horizon Agentic Robotics Benchmark for Physical, Strategic, and Spatial Reasoning

arXiv:2610.27039v1 Announce Type: new Abstract: Foundation models offer a promising basis for robotic agents, but long-horizon tasks require them to sustain coherent reasoning and reliable physical execution across interdependent stages. We introduce Embodied Playground, a benchm

HexaShell: A Thin Harness for Exploring the Limits of VLM Agents as Robot Coding Agents

arXiv:2610.23479v2 Announce Type: replace Abstract: Agents built on vision language models (VLMs) now compete with learned policies on robot manipulation leaderboards. We ask how far a coding agent can go when it controls a robot instead of a computer, and where it stops. Our sys

Physical AI for Intralogistics: Benchmarking the NVIDIA Sim-to-Real Ecosystem for Cosmos-Enhanced COMPASS Navigation on Industrial Mobile Robots

arXiv:2610.28432v1 Announce Type: new Abstract: Classical robot navigation stacks are reliable but often require re-engineering for each new robot type and site. Learning-based navigation shifts this effort from hand-tuned parameters to data generation and training, where a polic

ActionRoPE: Action as Coordinates for Spatially Consistent Control in Planar-Camera Game World Models

arXiv:2610.25861v2 Announce Type: replace Abstract: Game world models have made rapid progress toward real-time interactive generation. Yet action control is still predominantly implemented as feature conditioning, leaving the model to learn from training data how each action sho

AgentStripes: An Evidence-Graded Maturity Model for AI Agents

arXiv:2610.18603v2 Announce Type: replace Abstract: Benchmarks report what an Artificial Intelligence (AI) agent can do on a task suite, but not whether one released version has been engineered with the controls that make it reliable, safe and observable, or how strongly that is

Agentic Ensemble with Vision-Language Models for Visual Anomaly Detection

arXiv:2610.28560v1 Announce Type: new Abstract: Visual anomaly detection aims to detect abnormal objects, while anomaly samples are often scarce during model training. Assembling multiple pretrained detectors can help by leveraging their transferable image scores and anomaly maps

Agentic discovery of interpretable car-following laws from multi-source trajectory data

arXiv:2610.27836v1 Announce Type: new Abstract: Car-following (CF) models is a fundamental component of microscopic traffic simulation and underpin the analysis of collective traffic dynamics, while its modeling becomes more challenging due to diverse human driving behaviors and

Automated Multi-Agent Organization Design: From Architecture Search to Organizational Learning

arXiv:2610.27531v1 Announce Type: new Abstract: Agentic system design is increasingly shifting from hand-crafted prompts and workflows toward automated search over agentic programs and interaction structures, with multi-agent systems receiving growing attention. Yet a central que

AgenticRoboRL: Bringing Coding-Agent Post-Training to Agentic Robotics

arXiv:2610.28589v1 Announce Type: new Abstract: Agentic robotics systems combine planners with robot harnesses to perform complex manipulation tasks. Recent work improves these systems through better harnesses and external memory, but relies on powerful closed-source models as pl

Bulk migrate branch protections to rulesets for a repository

You can now convert all branch protection rules of a repository to rulesets in a single action. This is an extension of the ability to convert a single branch protection rule to a ruleset that we released previously, now allowing for at-scale conversions. How it works Navigate to

Updates to GitHub secret scanning

This update covers improvements and bug fixes for GitHub secret scanning. Update to audit log events On November 12, 2026 , the business field in affected secret_scanning_alert.* and secret_scanning_push_protection.bypass audit log events will change from the enterprise display n

OpenAI API 支持 12 小时 prompt 缓存，Codex 与 Work 用户缓存永久保留

OpenAI 开发者账号宣布 API 现已可选 12 小时 prompt 缓存，持续运行的智能体可在请求间隔数小时的情况下继续以更低的缓存输入费率复用指令、工具和上下文。Tibo 补充称，Codex 与 Work 中已登录用户的缓存会永久保留，返回旧对话或数小时后恢复会话保证命中缓存，且无额外惩罚。 🔗 阅读原文 via AIHOT · https://aihot.news/items/gf3h1rt1vsjw3dnw4hxlie3fv

GitHub package firewall in public preview

An increasing number of malware attacks exploit systems at package installation time, harvesting credentials on CI/CD hosts or developer workstations without even running the package’s code. To stem this tide of attacks, GitHub is introducing a new package firewall capability. Th

早报｜小天才回应封闭社交争议/车灯新规拟禁止显示文字/马斯克回应SpaceX与特斯拉合并

SpaceX 与特斯拉成立合并讨论委员会，AI 已成最大收入来源 混元扩至近 2000 人，员工称姚顺雨想把 DeepSeek「卷到加班」 报告：中国以外头部平台三季度新短剧，三分之二由 AI 生成 车灯国标征求意见：拟限 L3 及以上车辆安装 ADS 标志灯 小天才回应封闭社交争议：跨品牌可通话，微聊和好友圈暂不互通 工信部：防范未经充分测试的汽车进入市场，拟强化创新设计准入要求 Anthropic 上半年亏损超 800 亿美元，大部分来自融资工具会计计提 Waymo 将在马德里测试自动驾驶，目标 2027 年接待首批乘客 第一财经：9 月 BBA 主