Chain-of-Thought Reasoning Without Prompting · HackerPapers discussions

Chain-of-Thought Reasoning Without Prompting — mentioned by activatedgeek

Mon, 03 Feb 2025 18:09:15 GMT

That has been my understanding too. More generally, a verifier at the end certainly helps. In our paper [1], we find that asking a follow up question like "Is the answer correct?" and taking the normalized probability of "Yes" or "No" token (or more generally any such token trained for) seems to be best bet so far to get well-calibrated probabilities out of the model. In general, the log-probability of tokens is not a good indicator of anything other than satisfying the pre-training loss function of predicting the "next token." (it likely is very well-calibrated on that task though) Semantics…

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

Chain-of-Thought Reasoning Without Prompting — HN submission

Tue, 24 Sep 2024 05:06:37 GMT

Cot Decoding – Eliciting Reasoning from LLMs

A recent talk (https://dennyzhou.github.io/LLM-Reasoning-Berkeley.pdf) from Denny Zhou covered a number of techniques that improve LLM reasoning. In the talk he a recent paper from Google Deepmind on "Chain-of-Thought Reasoning without Prompting" (https://arxiv.org/abs/2402.10200). The key idea in the paper is that existing models are capable of doing CoT style step-by-step reasoning via a new decoding strategy. I implemented their approach in optillm - https://github.com/codelion/optillm/blob/main/optillm/cot_decoding.py as I couldn't find any decent open-source implementation. I have also…

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

Chain-of-Thought Reasoning Without Prompting — mentioned by activatedgeek

Fri, 22 Mar 2024 16:58:32 GMT

I hesitate to the use description as "think," just biasing correlations for subsequent generations. In any case, there is at least one work that shows that CoT may not be necessary and biasing the decoding path via logit probabilities is also promising. [1] One could argue it still doesn't contradict the benefits of CoT, but I suspect there is nothing fundamental about CoT, except that we happened to have been pre-training on sequences that use certain prompts that were easy to conceive from a human's perspective. [1]: https://arxiv.org/abs/2402.10200

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

Chain-of-Thought Reasoning Without Prompting — mentioned by XenophileJKO

Sun, 10 Mar 2024 19:14:29 GMT

Actually you can get a very good proxy by looking at the probability distrobution of the "answer" tokens. The key here is you have to be able to identify the "answer" tokens. https://arxiv.org/abs/2402.10200

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

Chain-of-Thought Reasoning Without Prompting — HN submission

Sat, 17 Feb 2024 16:06:40 GMT

Chain-of-Thought Reasoning Without Prompting

94 points; 25 discussion replies on HN. Paper and source history. A submission is not an endorsement.