热点
"推理轨迹" 相关文章
Beyond the Final Answer: Evaluating the Reasoning Trajectories of Tool-Augmented Agents
cs.AI updates on arXiv.org 2025-10-06T04:20:07.000000Z
Beyond the Final Answer: Evaluating the Reasoning Trajectories of Tool-Augmented Agents
cs.AI updates on arXiv.org 2025-10-06T04:20:07.000000Z
The whole point of OpenAI's Responses API is to help them hide reasoning traces
https://www.seangoedecke.com/rss.xml 2025-10-02T12:52:13.000000Z
研究分享 | 当大模型开始 "思考"，谁来守护它的安全？ReasoningShield 重磅登场！
复旦白泽战队 2025-08-16T13:54:32.000000Z
微软：两个AI相互纠错，数学再涨5分
36kr 2024-12-02T08:29:23.000000Z