{"type": "article", "title": "Debugging Logprob Mismatches in LLM Reinforcement Learning", "publisher": "Web Pulse", "url": "https://wpnews.pro/news/debugging-logprob-mismatches-in-llm-reinforcement-learning", "original_source": "https://rocm.blogs.amd.com/software-tools-optimization/logprob-debug/README.html", "published": "2026-09-23T00:00:00+00:00", "accessed": "2026-09-23", "id": "debugging-logprob-mismatches-in-llm-reinforcement-learning"}