Multi-Agent RL Recipe Cracks Chess Puzzles Without Predefined Roles
The Inkling team demonstrated a multi-agent RL recipe where several Inkling 2 Small agents, without predefined roles, learn to divide labor, communicate and delegate on hard chess puzzles—turning pass@k performance into pass@1.
2026-11-18 ~ 2026-11-20 · 2 related posts
- Multi-agent RL recipe turns pass@k into pass@1 with role-free Inkling 2 Small teams — keirp1 · 2026-11-18
- Multi-agent RL recipe turns pass@k into pass@1 on hard chess puzzles — shizhediao · 2026-11-20