Video large multimodal models have improved at identifying cross-frame visual evidence and conducting open-web retrieval, as reflected in recent video deep research (VDR) benchmarks. However, existing VDR benchmarks primarily evaluate tool-use trajectories and final responses without verifying whether a model actually...
Xi-Yuan Liang, Yu Wang, Zu-Hao Yang et al.· 0 citations
Test-time agentic computation can improve multimodal large language models (MLLMs) through planning, tool use, iterative perception, and reflection, but indiscriminate computation may duplicate resolved work, introduce noisy evidence, and perturb correct predictions. We propose that such computation should be residual:...
Yi-Ke Wu, Keyan Chen, Qi-Xiong Wang et al.· 0 citations
Despite their impressive performance on a wide range of video understanding tasks, current Vision Language Models (VLMs) are predominantly designed for offline scenarios and struggle to handle online streaming videos that demand low latency response. Several studies have explored memory and token compression strategies...
Yi-Heng Li, Xu He, Shao-Bo Wang et al.· 0 citations
CodeActionBench is introduced, a benchmark of 25 manipulation tasks that evaluates this capability through agentic Code-as-Policy and provides a controlled testbed for measuring how general-purpose models translate their capabilities into manipulation behavior and for examining typical failure scenarios in that process...
Yiheng Lyu, Xueying Jiang, Wen-Hao Li et al.· 2 citations
Despite rapid progress, most existing vision-language models (VLMs) built from 2D visual inputs often struggle when handling various 3D tasks that require fine-grained spatial understanding and reasoning. To bridge this gap, we present VLM-IE3D, a unified framework that enhances the 3D spatial awareness of VLMs by equi...
