GitHub/quartz
1
0
Fork 2
mirror of https://github.com/jackyzha0/quartz.git synced 2025-12-19 10:54:06 -06:00
Code Issues Actions Packages Projects Releases Wiki Activity

Fix typo

Browse Source
This commit is contained in:
Daniel Balcells 2025-08-26 16:45:40 -04:00
parent 0756773de7
commit 24a7b4c839
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
content/research/SPIRAL - Letting LLMs Teach Themselves Through Self-Play.md
View File

@ -10,7 +10,7 @@ tags:
---
![[selfplay.png]]
*Source: [Liu, Guertler et al., 2025](https://arxiv.org/abs/2506.24119)*.
*Source: [Liu, Guertler et al., 2025](https://arxiv.org/abs/2506.24119).*
## TL;DR
_We collaborated with the TextArena team to develop SPIRAL, a novel RL framework that allows LLMs to develop complex reasoning capabilities by playing text-based games against themselves. Using SPIRAL on a simplified variant of poker with no mathematical content, a 4B-parameter Qwen model improved its performance on math and reasoning benchmarks by 8.6% and 8.4% respectively. It does this by learning specific strategies, such as case-by-case analysis and expected value calculation, that generalize beyond poker better than simple game heuristics. We're excited to explore whether self-play on social deduction games like Mafia can lead to general improvements in LLMs' social cognition._