RLCR trains LLM agent organisations: Avalon-M1-Mini emerges hierarchies and verifiers
rosstaylor90 · x · 2026-10-23
GenReasoning released Avalon-M1-Mini (120B), its first long-horizon reasoning model trained with Reinforcement Learning from Collective Rewards (RLCR).
RLCR trains language models as agent organisations that share a message board and workspace, optimising collective rewards. Reported findings:
- Emergent organisational hierarchies and divisions of labour among agents
- Increased collective verification, including hiring verifier and tester agents
- The rise of collective thinking: agents reason more about other agents and how to help them
The team positions RLCR as a laboratory for studying collective intelligence and multi-agent safety. Blog post and technical report are available.
More from Research
- 20+ ICLR 2027 Submissions Found With Fabricated Results, Sparking Reproducibility Critique — srchvrs · 2026-10-23
- Qdrant's 64-byte-per-page recipe: 128x compression with 99.2% retrieval quality — qdrant_engine · 2026-10-23
- Qdrant benchmarks Cohere Embed 5 Pro: one dense vector per PDF page vs ColPali — qdrant_engine · 2026-10-23
- 20+ ICLR 2027 submissions self-confess fabricated placeholder results — thegautamkamath · 2026-10-23
- Pure FP vs C: Roc's DEFLATE grind closes in on libdeflate's performance — johnlindquist · 2026-10-23
- rosinality Fixes Numerical Precision Issues in Softmax, Praised by Lucas Beyer — giffmana · 2026-10-23