Build, Don't Play: Self-Improving LLM Research Agents Develop a Hearthstone Battlegrounds AI That Rivals Top Players
AutoTavern chains specialized LLM research agents through a shared typed research memory and verification gates, with live games closing the loop; in 20 days it built a faithful simulator of Hearthstone Battlegrounds and 23 versions of a PPO-trained playing agent. TavernAgent's live rating rose from 6,000 to about 15,000, and its last 20 records average a 1.60 placement with 80% first places, beating the server's ten highest-rated humans on every statistic (2.34, 59%).
Why it matters
Closes the loop from agent-built research artifacts to a system that outperforms top humans in a game agents cannot play well themselves.
Importance: 3/5
Notable preprint with a live-ladder result; not yet peer-reviewed.
Sources
official
arXiv:2610.33845 Build, Don't Play