Skip to content

Pull requests: opendilab/LightRFT

New pull request New
16 Open 28 Closed
16 Open 28 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

dev(hansbug): better docker builder and launcher enhancement New feature or request
#45 opened Feb 12, 2026 by HansBug • Draft
@HansBug
3
feature(wjy): add R1-AQA (Audio Question Answering) training examples enhancement New feature or request
#44 opened Feb 12, 2026 by JOY-SWang Loading…
24
feature(pu): add init version of on_policy_distillation enhancement New feature or request
#43 opened Feb 10, 2026 by puyuan1996 Loading…
feature(wzn): add language switcher and complete some Chinese documentation for LightRFT docs documentation Improvements or additions to documentation style Code or comments formatting
#42 opened Feb 10, 2026 by zunian-wan Loading…
3 of 40 tasks
52
feature(wzn): add LoRA training demo for Geo3K enhancement New feature or request
#41 opened Feb 10, 2026 by zunian-wan Loading…
1 of 47 tasks
17
feature(sunjx): implement dynamic sampling strategy in DAPO
#40 opened Feb 10, 2026 by Jiaxuan-Sun Loading…
WIP: feature(pu): adapt to npu device
#39 opened Feb 9, 2026 by puyuan1996 Loading…
feature(sunjx): add rejection sampling in grm_training
#38 opened Feb 6, 2026 by Jiaxuan-Sun Loading…
doc(pu): add init version of fast_exp_maker best practice documentation Improvements or additions to documentation
#37 opened Feb 3, 2026 by puyuan1996 Loading…
polish(nyz): fix redundant requires and vllm compatibility polish Polish algorithms, tests or configs
#36 opened Feb 3, 2026 by PaParaZz1 Loading…
4 of 9 tasks
feature(pu): add run_ppo_geo3k_qwen2.5_vl_7b.sh enhancement New feature or request polish Polish algorithms, tests or configs
#35 opened Feb 3, 2026 by puyuan1996 Loading…
7
feature(luyd): add partial rollout in training process enhancement New feature or request
#29 opened Jan 22, 2026 by AltmanD Loading…
8
feature(sunjx): add GSPO and GMPO algorithms support enhancement New feature or request
#22 opened Jan 9, 2026 by Jiaxuan-Sun Loading…
1
refactor(sunjx): refactor loss-filter implementation enhancement New feature or request refactor Cleanup, formatting, or restructuring of existing code.
#17 opened Jan 1, 2026 by Jiaxuan-Sun Loading…
15
refactor(sunjx): refactor dataset and reward module refactor Cleanup, formatting, or restructuring of existing code.
#13 opened Dec 31, 2025 by Jiaxuan-Sun Loading…
3
feature(sunjx): add rejective sampling pipeline in t2i demo enhancement New feature or request
#3 opened Dec 25, 2025 by Jiaxuan-Sun Loading…
3
ProTip! Type g p on any issue or pull request to go back to the pull request listing page.