3 papers
cs.CV2026
PhoStream: Benchmarking Real-World Streaming for Omnimodal Assistants in Mobile Scenarios
Xudong Lu, Huankang Guan, Yang Bo +10
Multimodal Large Language Models excel at offline audio-visual understanding, but their ability to serve as mobile assistants in continuous real-world streams remains underexplored…
cs.CV2025
CoIDO: Efficient Data Selection for Visual Instruction Tuning via Coupled Importance-Diversity Optimization
Yichen Yan, Ming Zhong, Qi Zhu +3
Multimodal large language models (MLLMs) rely heavily on instruction tuning to align vision and language capabilities, yet the computational cost of training on large-scale dataset…
cs.CV2025
SpecVLM: Enhancing Speculative Decoding of Video LLMs via Verifier-Guided Token Pruning
Yicheng Ji, Jun Zhang, Heming Xia +4
Video large language models (Vid-LLMs) have shown strong capabilities in understanding video content. However, their reliance on dense video token representations introduces substa…