nancui0000/adaptive-mogrpo

Adaptive Weight Scheduling for Multi-Objective GRPO in Code Generation. Fixed multi-objective rewards cause reward hacking (short but broken code). Our curriculum approach—correctness first, then gradually adding efficiency/brevity—preserves 81.7% HumanEval while generating 11% shorter code.

语言：Python · GitHub 星标：49 · Fork：6

在 GitHub 上查看原仓库