Thought Anchors: Which LLM Reasoning Steps Matter? — mentioned by krackers

Sat, 05 Jul 2025 19:49:20 GMT

There are several such papers, off the top of my head one is https://arxiv.org/abs/2404.15758 It's a bit more subtle though, if I understand correctly this only works for parallelizable problems. Which makes intuitive sense since the model cannot pass information along with each dot. So in that sense COT can be seen as some form of sampling, which also tracks with findings that COT doesn't boost the "raw intelligence" but rather uncovers latent intelligence, converting pass@k to maj@k. Antirez touches upon this in [1]. On the other hand, I think problems with serial dependencies require…

Paper and discussion history. A mention is not an endorsement.

Thought Anchors: Which LLM Reasoning Steps Matter? — HN submission

Wed, 25 Jun 2025 21:06:00 GMT

Thought Anchors: Which LLM Reasoning Steps Matter?

2 points; 0 discussion replies on HN. Paper and source history. A submission is not an endorsement.

Thought Anchors: Which LLM Reasoning Steps Matter? · HackerPapers discussions

Thought Anchors: Which LLM Reasoning Steps Matter? — mentioned by krackers

Thought Anchors: Which LLM Reasoning Steps Matter? — HN submission