arXiv Computer Vision
BVB: Benchmarking Agentic Video Understanding via Programmatic Reconstruction in Blender
Read the original on arXiv Computer Vision →The Flow has not summarised this story yet — read it at arXiv Computer Vision.
Related stories
arXiv AI
Sep 21
AgentVidBench: A Multi-Hop Video Question Answering Benchmark for Evaluating MLLM Agents
AgentVidBench is a new multi‑hop video question‑answering benchmark designed to evaluate spatial, temporal, and causal reasoning in multimodal large language models (MLLMs). Unlike existing tests that focus on simple scene queries or global summaries, AgentVidBench includes step‑by‑step solution traces to assess whether agents gather the necessary evidence to justify their answers. Experiments with 12 MLLMs show limited single‑turn performance, but integrating these models into agentic workflows improves both accuracy and trajectory scores, establishing AgentVidBench as a comprehensive testbed for future research on agentic video understanding.
arXiv Computer Vision
Oct 8
Video2World: Benchmarking Coding Agents for Interactive World Modeling from Embodied Videos
arXiv:2610.04432v2 Announce Type: replace Abstract: Building interactive simulators from real-world observations is a promising way to scale embodied data, but current pipelines still rely heavily on...
arXiv AI
Jun 26
Confidence-Aware Tool Orchestration for Robust Video Understanding
arXiv:2606. 26904v1 Announce Type: cross Abstract: Video reasoning language models implicitly assume that every input frame is equally reliable.
arXiv Computer Vision
Oct 28
VibeVideoBench: Evaluating Multimodal Agents on Code-Driven Video Generation
VibeVideoBench is a benchmark that evaluates multimodal agents capable of generating videos by writing code to render each frame and sound. The benchmark contains 239 self‑contained tasks in English and Chinese across 12 categories, requiring agents to produce a finished MP4 with sound within one hour without using any pre‑trained image or video generation model. Evaluation is hierarchical, using program checks, 13k task‑specific rubric checks, and five quality dimensions scored by three VLM judge panels, and it shows that GPT‑6 Astra and Claude Opus 5.5 perform best among nine frontier agents.
arXiv AI
Sep 30
Video-RSI: Recursive Self-Improvement of Video Understanding Agents via Harness Evolution
arXiv:2609.37950v1 Announce Type: new Abstract: Video understanding agents acquire evidence through an executable harness that controls what they observe and how they use those observations. However,...
arXiv Computer Vision
Oct 1
VidHarness: Evolving Agent Harnesses for Cost-Efficient Long Video Understanding
arXiv:2609.38413v1 Announce Type: new Abstract: Vision-language models (VLMs) can answer questions about hour-long videos, but processing every frame is prohibitively expensive, even though the evide...