[HN Gopher] COSP and USP: New methods to advance reasoning in LLMs
___________________________________________________________________
COSP and USP: New methods to advance reasoning in LLMs
Author : TheIronYuppie
Score : 23 points
Date : 2024-01-25 19:10 UTC (3 hours ago)
(HTM) web link (pub.towardsai.net)
(TXT) w3m dump (pub.towardsai.net)
| TwentyPosts wrote:
| This appears to be for "paid subscribers".
| sgt101 wrote:
| Google can be your friend brother (still).
|
| https://pub.towardsai.net/inside-cosp-and-usp-google-researc...
| dang wrote:
| Ok, we'll use that instead of
| https://thesequence.substack.com/p/about-cosp-and-usp-two-
| ne..., which looks hardwalled. Thanks!
| sgt101 wrote:
| Self adaptive prompting (prompt generation based on some sort of
| feedback from the model I think).
|
| USP: https://arxiv.org/abs/2305.14926
|
| COSP: https://aclanthology.org/2023.findings-acl.216/
|
| A hall of mirrors or a wonderful telescope? Time will tell.
___________________________________________________________________
(page generated 2024-01-25 23:01 UTC)