SyncAI.news, a Varaisys broadcasting
Relax Within, Balance Across: Load Balancing under Composition Shifts in Vision-Language Mixture-of-Experts
ZW

Ziang Wu, Peng Jin, Qishen Yin, Munan Ning, Hao Li, Peizhen Zhang, Huihui Zhou, Li Yuan

· 1 min read

ResearcharXiv cs.AI

Relax Within, Balance Across: Load Balancing under Composition Shifts in Vision-Language Mixture-of-Experts

arXiv:2608.00574v2 Announce Type: replace-cross Abstract: Vision-language mixture-of-experts (MoE) models such as Qwen3-VL need balanced loads for fast parallel serving. Standard training pools image and text tokens into one auxiliary loss, assuming training balance holds at inference. Yet we find a surprising mismatch: on a fixed router, balance shifts sharply whenever resolution, dynamic tiling, or image count changes the image-to-text token ratio. Splitting load by modality reveals cross-modal load cancellation: opposing image and text errors cancel near the training ratio, hiding strong skew until the mix changes. To quantify this shift, we derive an exact quadratic law: under fixed modality profiles, one reference run predicts the full response curve, and the image-text gap governs sensitivity. Meanwhile, patches within an image route similarly while patch counts vary, letting large images dominate flat pooling. We propose ReBA (Relax Within, Balance Across) to separate modality losses by token fraction and weight images equally, letting each image specialize while keeping overall visual load even. Across four split backbones, ReBA turns fragile single-ratio balance into stable balance across resolutions, tilings, and image counts without hurting accuracy. Large modality gaps in released native MoEs also show its promise for native post-training. Code is available at https://github.com/ZiangWu-77/ReBA.

Original source

This story was published by arXiv cs.AI and written by Ziang Wu, Peng Jin, Qishen Yin, Munan Ning, Hao Li, Peizhen Zhang, Huihui Zhou, Li Yuan. SyncAI.news shows a preview; the complete article is on the publisher's site.

Read the full story on arxiv.org

Similar News