2026-10-03 科技日报

扫描 430 篇候选内容 · 覆盖 650 个信息源 · 纳入 414 篇

📌 今日要点

• Open-sourcing AstaBrief, the fast report-generation model in Asta
• AutoSynthData: Generating Training Data for Enterprise Agents
• Sweep thousands of leases for compliance using Amazon Quick and the Adjudicated Query pattern
• Add secure Web Search to Claude Desktop with Amazon Bedrock AgentCore
• Fine-tune a search agent with multi-turn RL on Amazon SageMaker AI
• A model guide for the GPT-6 family

📝 科技简讯

• Open-sourcing AstaBrief, the fast report-generation model in Asta — Hugging Face Blog
• AutoSynthData: Generating Training Data for Enterprise Agents — Hugging Face Blog
• Sweep thousands of leases for compliance using Amazon Quick and the Adjudicated Query pattern — AWS Machine Learning
• Add secure Web Search to Claude Desktop with Amazon Bedrock AgentCore — AWS Machine Learning
• Fine-tune a search agent with multi-turn RL on Amazon SageMaker AI — AWS Machine Learning
• A model guide for the GPT-6 family — OpenAI Blog
• Travel Time Prediction in Supply Chain Management Using Machine Learning — ArXiv ML (cs.LG)
• EHR2Trace: Auditable EHR Data Infrastructure for Patient World Models and Clinical Agents — ArXiv ML (cs.LG)
• A Moving-Horizon Approximate Branch-and-Reduce Method for Deep Classification Trees — ArXiv ML (cs.LG)
• A Data-Free Physics-Informed Neural Operator for Level-Set Interface Advection — ArXiv ML (cs.LG)
• Conformal Adversarial Generative Ensemble — ArXiv ML (cs.LG)
• DualCast: A Dual-Path Language Model for Bimodal Financial Time-Series Forecasting — ArXiv ML (cs.LG)
• FlashDiffusion: Fused Tiled Kernel Spectral Decomposition — ArXiv ML (cs.LG)
• The latest AI news we announced in September 2026 — Google AI Blog
• Meta wants your next gadget to be Muse-infused — TechCrunch AI
• Sean Parker is rebuilding Stability AI around music — TechCrunch AI
• Affected by layoffs? Don’t miss this $75 deal for your TechCrunch Disrupt 2026 Expo+ Pass — TechCrunch AI
• Apple says it’s tightening macOS ‘Full Disk Access’ controls due to new risks from AI agents — TechCrunch AI
• Call it AI, call it Super Intelligence, only 2% of consumers are buying it — TechCrunch AI
• It’s not AI anymore, it’s ‘super intelligence’ (according to the White House) — TechCrunch AI
• TechCrunch Disrupt 2026: Blackstone’s Jas Khaira on building the next generation of AI giants — TechCrunch AI
• Where the Agent Development Lifecycle Fits — Towards Data Science
• How to Build a Control Plane for AI Agents — Towards Data Science
• Anthropic 首曝秘密游说梵蒂冈：Claude 已经有意识了！ — 新智元
• 刚刚，Anthropic「第一夫人」出局了！ — 新智元
• Low-Background Minds — LessWrong
• What is recursive self-improvement and what would it mean to ban it? — LessWrong
• State of the Field: AI for Epistemics and Coordination — LessWrong
• How to apply to AI safety fellowships (and beyond) — LessWrong
• Petrov Day in Cusco, with a candle — LessWrong
• openJiuwen X-Router 自演进模型路由技术首发，昇腾亲和，Agent 越跑越省，实测减少 50+%Token 消耗 — 量子位
• 丘成桐新论文致谢了 GPT 和 Claude — 量子位
• arXiv 最严新规！每人每月最多提交 2 篇，拒稿不退额度 — 量子位
• Unpacking Dreamforce: Why Your AI Needs Trusted Context — Salesforce AI Blog
• Where’s the Party? Zero-Party Data Basics for Small and Growing Businesses — Salesforce AI Blog
• Heavy-Tailed Memory Traces in Long-Horizon Language Agents — arXiv AI (cs.AI)
• When Do Causal World Models Help Modular LLM Agents — arXiv AI (cs.AI)
• From Proposal to Verified Effect: Praxa, an Evidence-Bound Harness for Governed AI Agent Execution — arXiv AI (cs.AI)
• What Do Rationales Communicate? A Message-Intervention Study in Role-Specialized QA — arXiv AI (cs.AI)
• Measuring the Microtask Eligibility Gap: When Is an Off-the-Shelf SLM Enough for an Agent Harness? — arXiv AI (cs.AI)

📊 数据概览