ViperEkura 2c7a71a9c0 refactor: separate old policy and ref model in GRPO strategy ... - Split single ref_model into old_model (importance sampling ratio) and ref_model (frozen KL regularizer) - Move ref_model/old_model creation from strategy __init__ to TrainContextBuilder, pass as explicit parameters - Remove periodic sync_ref_model + sync_interval; add sync_old_model for external rollout loop to call - DPOStrategy also receives ref_model from builder - Fix std to use unbiased=False (population std per GRPO paper) - Remove redundant tests (test_grpo_kl_zero_at_init, test_grpo_no_sync_interval_param) - Remove --grpo_sync_interval CLI arg		2026-07-14 20:03:45 +08:00
..
demo	feat: add model_path temperature top_p top_k max_tokens system_prompt args to stream_chat	2026-07-04 07:33:32 +08:00
eval	fix: resolve IFD token-set asymmetry and support single-token answers	2026-07-05 17:48:26 +08:00
tools	refactor: separate old policy and ref model in GRPO strategy	2026-07-14 20:03:45 +08:00
docker.sh	fix: docker-compose UID/GID 添加默认值，修复 docker.sh logs 命令	2026-05-18 14:24:00 +08:00
pre_commit.sh	ci: 优化 GitHub Actions 工作流	2026-04-05 22:40:16 +08:00