全注意力、复杂推理不掉速：MiniMax M2把Agent做成了「可执行能力」
"Chain-of-Thought" 相关文章
Open-weight training practices and implications for CoT monitorability
Advancing Explainable AI in Radiology Research with NVIDIA Clara Reason
Google 白皮书核心解析：AI Agent 落地开发全指南
吴恩达关注的Ling-1T背后，蚂蚁Ling 2.0技术报告解密万亿模型开源配方
让机器人「不仅会想，还能准确去做」，VLA-R1把「推理+行动」带进真实世界
让机器人「不仅会想，还能准确去做」，VLA-R1把「推理+行动」带进真实世界
让机器人「不仅会想，还能准确去做」，VLA-R1把「推理+行动」带进真实世界
Can Reasoning Models Obfuscate Reasoning? Stress-Testing Chain-of-Thought Monitorability
Can Reasoning Models Obfuscate Reasoning? Stress-Testing Chain-of-Thought Monitorability
Highlights from Explaining AI Explainability
Highlights from Explaining AI Explainability
A Concrete Roadmap towards Safety Cases based on Chain-of-Thought Monitoring
SDC2025 精彩议题预告 | SCPGA：自认同CoT渐进式泛化攻击
First Try Matters，不是Aha Moment：邴立东团队揭示推理模型靠首答，不靠反思
AI-202X-slowdown: can CoT-based AIs become capable of aligning the ASI?
AI-202X-slowdown: can CoT-based AIs become capable of aligning the ASI?
NVIDIA Researchers Propose Reinforcement Learning Pretraining (RLP): Reinforcement as a Pretraining Objective for Building Reasoning During Pretraining
NVIDIA Researchers Propose Reinforcement Learning Pretraining (RLP): Reinforcement as a Pretraining Objective for Building Reasoning During Pretraining
CoT 之后，CoF 如何让帧间逻辑从「隐式对齐」变成「显式思考」？
