This is a heavily interactive web application, and JavaScript is required. Simple HTML interfaces are possible, but that is not what this is.
Post
hakwan lau
hakwan.bsky.social
did:plc:nyx64cl7oqkc3nqwiu3cnps5
arxiv.org/abs/2506.10943
"To train the model to produce effective self-edits, we use a reinforcement learning loop with the downstream performance of the updated model as the reward signal. ... directly uses the model's own generation to control its adaptation process."
https://arxiv.org/abs/2506.10943
2025-10-15T22:21:05.337Z