YAML Metadata Warning:empty or missing yaml metadata in repo card
Check out the documentation for more information.
🏛️ RMAD — Recursive Multi-Agent Debate GRPO
Trains the model's <thought> block to implement 3-persona internal debate:
- Logician — deductive reasoning, formal logic chains
- Fact-Checker — verifies claims, flags unsupported assertions
- Domain Specialist — applies domain expertise
Reward System (5 signals)
| Weight | Function | Signal |
|---|---|---|
| 0.40 | consensus_correctness |
+1.5 correct consensus, -1.0 wrong, None malformed |
| 0.25 | logic_leap_punishment |
-2.0 if FC critique ignored, +0.5 if addressed |
| 0.20 | rmad_format_reward |
Graduated 0→1.0 for structural compliance |
| 0.10 | debate_depth_reward |
Multi-round substantive debate quality |
| 0.05 | soft_overlong |
Length penalty (DAPO-style) |
Launch
accelerate launch rmad_grpo_train.py
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support