JY
Junmei Yang
cs.AIcs.LG
On Valency
published · living versionsW_k9rnuzz4·v1 · currentpublished
Stochastic Pruning of the DPO Computation Graph: 1.4x Faster Training and Lower Peak Memory
with Yubing Wu, Wei Chen, Delu Zeng, Tiewen Pan
1 version
Preprints & journals
4 papers in the corpus · 2026RubricOpt: Self-Revision of Semantic Rubrics for Preference Optimization2610.29491v1 · Wei Chen, Yubing Wu, Shigui Li et al.2026 · 0 citationsarXivon Valency
Stochastic Pruning of the DPO Computation Graph: 1.4x Faster Training and Lower Peak Memory2610.29538v1 · Yubing Wu, Wei Chen, Junmei Yang et al.2026 · 0 citationsarXivon Valency
Hindsight-Privileged RLVR: Single-Rollout Token Credit Without Group Sampling2610.20497v2 · Wei Chen, Yubing Wu, Zhou Xu et al.2026 · 0 citationsarXiv
Towards Disentangled Preference Optimization Dynamics: Suppress the Loser, Preserve the Winner2604.18239v4 · Wei Chen, Yubing Wu, Junmei Yang et al.2026 · 0 citationsInternational Conference on Machine Learning(ICML) 2026
Career total: 5 works. 4 are in this corpus.
Profile built from the corpus for this byline.
Author records are still filling in while the Hub is in alpha. If this is your page, you'll be able to claim it soon. Spot a mistake? Tell us.