#53
Trusting-Trust Attack against an Entire Linux Distribution
推荐理由「Trusting-Trust Attack against an…」对造物者有用的是方法能不能迁到自己的 Agent 或生成管线，不是把标题再念一遍。
Pai 的 AI 雷达
今天的简报围绕“代理正在如何工作、被评估和治理”展开：从 agent-sessions、前沿模型的 AEO 选择与应对，到数学代理作弊、Agent 线上 Badcase 和不确定性闸门，再延伸至 ChatGPT 与 Gemini 的流量变化、内容产品与独立新闻支持；同时夹带视频压缩、地图投影、洛杉矶城市演化和 MovieMaker 等可直接动手或轻松浏览的技术小品。阅读时可先把前半部分当作一组关于 Agent 可靠性、分发与社会影响的连贯线索，再用后半部分的工具、可视化和旧邮件案例换气，最后回到“如何在不确定中使用 Agent”这一实践问题。
Philippines-based dev here, on Claude Code since day one and on Max 20x since it existed. It costs me $224/month ($200 + 12% VAT). ChatGPT Pro 20x costs me ₱9,990, about $160 with VAT baked in. Same top tier, $64/month cheaper, because OpenAI actually prices for the market and bi
GPT-6 Astra上线后，大量用户用它自主操控Blender、Houdini、Unity、Aseprite等专业软件，做出游戏Demo、3D复刻旧金山艺术宫等作品，其中旧金山艺术宫案例里Astra自己搜索几百张参考图、翻到美国国会图书馆的老扫描文件找柱子尺寸，多数工作在夜间自主完成。 🔗 阅读原文 via AIHOT · https://aihot.virxact.com/items/cmtqhifvf04pwrohqvdfqmfxp
Evaluating the calibration of Large Language Models (LLMs) is critical for their safe deployment as zero-shot classifiers. Yet, commercial API providers increasingly hide the continuous output probabilities required by standard calibration metrics. To bypass this opacity, we demo
Computer-use agents have advanced on benchmarks like OSWorld and AndroidWorld, but still act mostly through the GUI, often producing inefficient trajectories. Real-world computer work is hybrid, combining visual-state inspection with precise, high-throughput command-line operatio
Use Claude, Anthropic's AI assistant, right from your terminal. Claude can understand your codebase, edit files, run terminal commands, and handle entire workflows for you.
Large language models (LLMs) show potential for medical tasks, but their single-turn question-answer format does not reflect how clinical diagnosis is performed in practice. As a result, they remain limited in complex diagnostic settings. We developed Debate-Mixture-of-Agents (DM
Building on the pioneering paper of Kearns, Roth, and Ryu (SODA'26), we study information aggregation in a networked learning model. The model captures a central pattern in agentic AI: each agent sees only part of the data and passes on only its own conclusion. Their model consid
A rapidly expanding ecosystem of actors is removing built-in safety guardrails from open-weight AI models. We profile this ecosystem by identifying key producers, downstream reproductions, and emerging applications. Between January 2024 and March 2026, we identified 3,471 origina
Autonomous AI agents increasingly select actions in environments whose memory, execution-time, runtime, compute, and operational constraints determine what counts as a suitable plan. We call the absence of this execution context from an agent's planning state substrate blindness.
Hey everyone! Has anyone else experienced an exponential increase in token usage on Claude Pro after the latest update? I always try to follow good practices to get the most out of Claude’s usage limits. For example, I avoid leaving unnecessary chats open, and when I’m working on
Explainability is increasingly seen as a crucial requirement in AI-based medical diagnosis, particularly in safety-critical clinical decision-making. Most existing explainability methods in healthcare operate in a post-hoc manner and are predominantly designed for unimodal data,
Automotive infotainment validation still relies on manual testing, slow, costly, and incompatible with agile releases and OTA updates. Scripted automation only partly helps: it couples test logic to implementation, yielding brittle, high-maintenance suites. Existing LLM-driven fr
OpenAI 发布内部数据称已达到自动化研究实习生里程碑，即可在人类监督下完成熟练研究员需数天的明确任务。截至 8 月中旬，其研究组织每投入 1 个人工工作日，就使用 3.1 个 agent 工作日的运行时长，该比值衡量的是运行时间而非等效生产力；原文作者援引 OpenAI 员工观点称递归自我改进或成为未来几年 AI 能力的关键，并呼吁其他 AI 公司同样公开数据。 🔗 阅读原文 via AIHOT · https://aihot.virxact.com/items/cmtq5pxc10275roiufscxyie4
Large language models become consequential agents when surrounding systems let outputs change external state. Models now call tools, operate interfaces, delegate work, retain state, inhabit generated worlds, and control robots or laboratory equipment. Such advances are often narr
Evaluation of medical artificial intelligence agents remains predominantly answer-centric, assessing only the correctness of final outputs while overlooking the quality of intermediate reasoning. In clinical settings, however, a correct answer reached through fabricated evidence
Autonomous coding agents are increasingly proposed as AI-scientist systems that conduct analyses and write research reports, but executing a prescribed analysis is not the same as making a discovery. Existing benchmarks are configured for reproduction: tasks, data, and rubrics ar
Long-running LLM agents are stateful: beyond the transcript they accrete compressed summaries, plaintext memory, pending tool plans, and, under every serving API, a KV cache. Yet today's "forget" operations delete a plaintext memory record and stop, leaving every artifact derived
Skill libraries improve the sample efficiency of agentic reinforcement learning (RL) by enabling large language model (LLM) agents to reuse procedural knowledge. Yet existing paradigms exhibit structural shortcomings: they either decouple skill evolution from policy optimization
Enterprise work spans messages, decisions, projects, and obligations that change over time. An AI agent that starts without this context must reconstruct it before contributing. To provide agents with this necessary context, our team used NVIDIA NemoClaw to build a memory-driven
This paper presents new tokenization resources for Irish and evaluation measures of alignment with the morphological boundaries of the language. We present MoirfEolas, a dataset of over 35,000 Irish words mapped to their respective eclipses, prefixes and suffixes as well as an ev
OpenAI 在模型文档中说明 GPT-6 Astra 相比 GPT-5.6 Sol 更常提出澄清问题、对上下文更敏感，并给出让模型更主动、审计 AGENTS.md 等技能文件、控制写作风格、约束子智能体委派和测试规模的提示词建议。 🔗 阅读原文 via AIHOT · https://aihot.virxact.com/items/cmtofayvo02uoroht238ucfjw