Game of Thought: Robust Information Seeking with Large Language Models Using Game Theory
Langyuan Cui, Chun Kai Ling, Hwee Tou Ng · 2026
Read Game of Thought: Robust Information Seeking with Large Language Models Using Game Theory by Langyuan Cui, Chun Kai Ling, Hwee Tou Ng (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2602-01708.txt.
Description
Langyuan Cui and colleagues improve information seeking where language models lack facts they need. Methods assume simplifications that degrade worst case performance, a flaw in high stakes settings. The authors use Twenty Questions to test information seeking. They formalize its adversarial counterpart, Strategic Language Search, as a two player zero sum game in extensive form. Their Game of Thought uses game theory to approximate a Nash equilibrium strategy for the restricted game. Tests show it improves worst case performance over direct prompting and heuristic guided search across all settings.
Catalog record
| CREATOR | Langyuan Cui, Chun Kai Ling, Hwee Tou Ng |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 85017 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2602.01708