#
gapo
Here are 2 public repositories matching this topic...
Reproduction study of Group-Aware Policy Optimization (GAPO) for LLM output diversity - Qwen2.5-7B + LoRA + TRL/GRPO, achieving a 3.3x JSD reduction. ECEN 743 team project (showcase: report, results, analysis).
nlp diversity machine-learning reinforcement-learning lora peft mode-collapse trl gapo large-language-models llm rlhf qwen grpo
-
Updated
Jun 22, 2026
Add this topic to your repo
To associate your repository with the gapo topic, visit your repo's landing page and select "manage topics."