Chain-of-Thought Reasoning in the Wild Is Not Always Faithful a arxiv.org 2 flyingpumba 1 year ago 0 comments
Show HN: A Wordle for the command line (Spanish and English) g github.com 34 flyingpumba 4 years ago 16 comments
When doing the paper we noticed that models are very good at generating post-hoc plausible CoT, which to me knowledge can happen quite often with relatively easy tasks.
You might be interested in reading this other paper that came out after ours: https://arxiv.org/abs/2507.05246