CP-Agent: A Calibrated Risk-Controlled Agent for Feedback-Driven Competitive Programming 事件

CP-Agent: A Calibrated Risk-Controlled Agent for Feedback-Driven Competitive Programming arXiv:2605.24693v1 Announce Type: new Abstract: Large language models still struggle with contest-level programming, while many agentic remedies rely on massive inference-time sampling or expensive multi-stage post-training. We study when execution feedback reliably helps an LLM CP solver and which mechanisms govern the gains. We model feedback-driven solving as a calibrated stopped process and identify thr