3 papers
cs.CV2026
Relax Within, Balance Across: Geometry-Guided Load Balancing for Vision-Language Mixture-of-Experts
Ziang Wu, Peng Jin, Qishen Yin +4
Vision-language MoE batches contain different numbers of image and text tokens. Image resolution, image count, tiling, and prompt length all change this token mix. We call the stan…
cs.AI2026
Divergence Decoding: Training-Free Capability Fusion
Yimi Wang, Hao Li, Shuo Yang +6
While large language models excel in reasoning, these generalists often lack knowledge for specialized scientific domains. Conversely, domain models~(specialists), while knowledgea…
cs.CL2024
LOOK-M: Look-Once Optimization in KV Cache for Efficient Multimodal Long-Context Inference
Zhongwei Wan, Ziang Wu, Che Liu +5
Long-context Multimodal Large Language Models (MLLMs) demand substantial computational resources for inference as the growth of their multimodal Key-Value (KV) cache, in response t…