Can LLMs Recover Strategies from Gameplay Behavior?

OfirPress · x · 2026-07-12

An AIWILD workshop poster discusses: Can SOTA LLMs recover hidden strategic heuristics purely from behavioral outputs, much like scientific research?

The post also asks whether models can identify and synthesize the core heuristics forming a policy in game strategies like poker, inviting attendees to discuss at the #icml2026 poster session.

Original post →

More from Research

Research channel →