一作ACL 2026Poster· Natural Language Processing
MACS : Modality-Aware Capacity Scaling for Efficient Multimodal M o E Inference
Bo Li · Chuan Wu · Shaolin Zhu
摘要
Mixture-of-Experts Multimodal Large Language Models (MoE MLLMs) suffer from a significant efficiency bottleneck during Expert Parallelism (EP) inference due to the straggler effect. This issue is worsened in the multimodal context, as existing token-count-based load balancing methods fail to addr...