Balancing the Experts: Unlocking LoRA-MoE for GRPO via Mechanism-Aware Rewards

Apr 24, 2026·

Changlian Ma

,

Zizheng Huang

,

Xiangyu Zeng

,

Yi Wang

,

Cheng Liang

,

Kun Tian

,

Xinhai Zhao

Limin Wang

Limin Wang

· 0 min read

Type

Conference paper

Publication

The Fourteenth International Conference on Learning Representations

Last updated on Apr 24, 2026

Limin Wang

Authors

Nanjing University

← Arbitrary Generative Video Interpolation Apr 24, 2026

CaReBench: A Fine-grained Benchmark for Video Captioning and Retrieval Apr 24, 2026 →