41554 blogs · [ { "id": "01a0875a-c101-7184-833a-494ae8506c7a", "title": "I Wrote \"NEVER STOP\" and Went to Bed", "url": "https://burtonye.github.io/posts/autoresearch-ralph/", "published_at": "2026-03-30T00:00:00+00:00" }, { "id": "01a0875a-c101-7184-833a-494ae92a0b30", "title": "Can a Model Teach Itself With Prompts Instead of Gradients?", "url": "https://burtonye.github.io/posts/training-free-grpo-experiment/", "published_at": "2026-03-26T00:00:00+00:00" }, { "id": "01a0875a-c101-7184-833a-494ae9c97e52", "title": "Which Models Actually Benefit From Prompt-Injected Experiences?", "url": "https://burtonye.github.io/posts/receiver-rejector-experience-injection/", "published_at": "2026-03-26T00:00:00+00:00" }, { "id": "01a0875a-c101-7184-833a-494aea51eb04", "title": "Hard Examples Are All You Need for GRPO", "url": "https://burtonye.github.io/posts/hard-examples-grpo/", "published_at": "2025-08-20T00:00:00+00:00" }, { "id": "01a0875a-c101-7184-833a-494aeac1c45f", "title": "Reward Hacking in RLHF: What Can Go Wrong", "url": "https://burtonye.github.io/posts/reward-hacking-in-rlhf/", "published_at": "2025-08-03T00:00:00+00:00" }, { "id": "01a0875a-c101-7184-833a-494aeacc4d16", "title": "What is OpenAI's Operator Good For?", "url": "https://burtonye.github.io/posts/what-is-openai-operator-good-for/", "published_at": "2025-02-01T00:00:00+00:00" }, { "id": "01a0875a-c101-7184-833a-494aeaf9a7b6", "title": "About", "url": "https://burtonye.github.io/about/", "published_at": null } ] posts Claim your blog

Blog · corpus.blog/blogs/burtonye.github.io/posts