-
Notifications
You must be signed in to change notification settings - Fork 70
Pull requests: RL-Align/RL-Kernel
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
fix(gemm): BF16 store + mid-split K-tree for simulated TP=2
#318
opened Aug 18, 2026 by
frank-2077
Collaborator
Loading…
1 task
[WS2][Logp] Expose the lse_vocab_tile_reduction ablation switch on linear-logp ops
#317
opened Aug 17, 2026 by
zhangj1an
Collaborator
Loading…
[WS2][Logp] Deterministic config option for operator
platform: cuda
Specific optimizations or bugs in NVIDIA graphics cards (such as FlashInfer, TMA optimizations)
#314
opened Aug 16, 2026 by
KJLdefeated
Collaborator
Loading…
[CI][refator]: migrate GPU workflow
needs-gpu-ci
#309
opened Aug 14, 2026 by
Flink-ddd
Collaborator
Loading…
[WS2][Mismatch] Add Qwen3 FFN implementation factor
#308
opened Aug 13, 2026 by
bitborne
Collaborator
Loading…
[WS2][GEMM] Add Qwen3 FFN orchestration and consistency validation
#304
opened Aug 13, 2026 by
bitborne
Collaborator
Loading…
5 of 10 tasks
[WS2][GEMM][Forward]: implement PR4 TP+CP+SP with FFN and sequence collectives
#298
opened Aug 12, 2026 by
Flink-ddd
Collaborator
Loading…
[draft][WS1][kernels] Batch-invariant logprob Ascend C Kernel
#297
opened Aug 12, 2026 by
erfgss
Loading…
[WS2][GEMM][Forward]: implement PR4 TP+CP with FFN and collective reductions
#295
opened Aug 12, 2026 by
Flink-ddd
Collaborator
Loading…
[WS2][GEMM][Forward]: implement PR3 TP with FFN and batch-invariance tests
#293
opened Aug 12, 2026 by
Flink-ddd
Collaborator
Loading…
[WS2][PR5][Logp] Add deterministic DP-aware GRPO loss on the TP-aware logprob path
#291
opened Aug 12, 2026 by
KJLdefeated
Collaborator
Loading…
[WS2][PR4][Logp] feat: add distributed TP/CP logprob drift report
#289
opened Aug 11, 2026 by
hihaluemen
Contributor
Loading…
[WS2] Ablation Matrix API v2.0 on top of PR230
#288
opened Aug 9, 2026 by
zhangj1an
Collaborator
Loading…
[WS2][PR5][Attention] Add distributed CP drift benchmark artifacts
#285
opened Aug 8, 2026 by
inaniloquentee
Collaborator
Loading…
feat(attention): add CP backward drift validation
#284
opened Aug 7, 2026 by
inaniloquentee
Collaborator
Loading…
[WS2][PR7][Attention] Add FlashInfer RoPE-fused paged attention scaffold
#279
opened Aug 5, 2026 by
inaniloquentee
Collaborator
Loading…
[WS2][PR3][Logp] Add deterministic vocab-parallel TP logprob reference
#265
opened Aug 5, 2026 by
KJLdefeated
Collaborator
Loading…
[WS2][Cross-config][Attention] Integrate the Qwen3-8B TP=2 CP=2 BF16 Attention setting to cross-config alignment framework
#263
opened Aug 4, 2026 by
zhangj1an
Collaborator
Loading…
[WS2][PR2][Logp] feat: add TP=1 logprob comparison harness
#262
opened Aug 4, 2026 by
hihaluemen
Contributor
Loading…
feat(attention): add decode-stage KV-cache CP replay
#260
opened Aug 3, 2026 by
JLiu4Coding
Loading…
[WS2][PR1][Logp] Add TP-aware logprob contract and dispatch metadata
#259
opened Aug 2, 2026 by
ryankert01
Loading…
[PERF][kernels] Write ratio_kl backward directly to policy dtype
needs-gpu-ci
#257
opened Aug 1, 2026 by
hongleng
Loading…
[FEAT][kernels] Add fused ratio-clip-aggregate loss primitive
needs-gpu-ci
#255
opened Jul 31, 2026 by
Chen-BUPT
Loading…
feat(attention): add single-gpu comparison harness
#253
opened Jul 30, 2026 by
inaniloquentee
Collaborator
Loading…
Previous Next
ProTip!
Type g i on any issue or pull request to go back to the issue listing page.