News for hackers. Stuff that matters.
Re:Self-Correction Bench: Revealing and Addressing LLM Self-... by yubblegum on Wednesday October 08, 2025 @09:37PM (#45522532)
Note: Actual title is Self-Correction Bench: Revealing and Addressing the Self-Correction Blind Spot in LLMs, Ken Tsui, 2025.

Quite an interesting result. Per my chats with one of the online models, this magic token "wait" is generally applicable across all models.

There are no comments on this comment yet.