MiDashengLM-Gen RL-Round2-v2 (rl-r2v2)

Single clean Rank-64 Flow-DPO + Stop-Head RL LoRA anchored directly to ModelsLab/midashenglm-gen-wer-lora (scene-v8).

Paired 132-Clip A/B vs scene-v8

  • Blind Gemini 3 Flash overall win-rate: 75.0% (36–12), 0 real regressions
  • Ambience: 87.5% (7–1) blind win-rate
  • Music: 80.0% (4–1) blind win-rate, CLAP 0.481 -> 0.522 (+0.041)
  • Multi-Tag Mixed: 78.3% (18–5) blind win-rate, WER 26.7% -> 22.3%
  • SFX: 58.3% (7–5) blind win-rate, score 0.660 -> 0.682
  • English Speech WER: 8.99% -> 2.74%

Load stacked on top of scene-v8: ADAPTER_DIR="adapters/scene-v8,adapters/rl-r2v2".

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. πŸ™‹ Ask for provider support

Model tree for ModelsLab/midashenglm-gen-rl-r2v2

Adapter
(2)
this model