BVB: Benchmarking Agentic Video Understanding via Programmatic Reconstruction in Blender

arXiv:2609.15478 · cs.CV · Submitted 2026-09-14 · Read on arXiv

cs.CV

Submitted: 2026-09-14

Updated: 2026-09-26

Comments: The 3rd version. V1 was released on Sept. 14, 2026. Project Page: https://yoloytang.me/BVB/

Code: https://github.com/ahujasid/blender-mcp

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Benchmarking — 33× in this paper · explained in 3 episodes
• Benchmark — 28× in this paper · explained in 2 episodes
• Dual VQA — 26× in this paper
• Latent Similarity — 23× in this paper
• Conference on Computer Vision and Pattern Recognition — 8× in this paper
• Accuracy — 6× in this paper · explained in 1 episode
• Diffusion Model — 2× in this paper · explained in 8 episodes
• Diffusion Models — 2× in this paper · explained in 6 episodes
• Grounding — 2× in this paper · explained in 3 episodes
• Multimodal Reasoning — 2× in this paper · explained in 5 episodes
• Aggregation — 1× in this paper · explained in 1 episode
• Alignment — 1× in this paper · explained in 1 episode
• CVF International Conference on Computer Vision — 1× in this paper
• Calibration — 1× in this paper · explained in 7 episodes
• Failure Modes — 1× in this paper · explained in 3 episodes
• Fine-Tuning — 1× in this paper · explained in 5 episodes
• Foundation Model — 1× in this paper · explained in 3 episodes
• Foundation Models — 1× in this paper · explained in 7 episodes
• In International Conference on Learning Representations — 1× in this paper
• Reinforcement Learning — 1× in this paper · explained in 4 episodes
• Robustness — 1× in this paper · explained in 5 episodes
• Self-Supervised Learning — 1× in this paper · explained in 7 episodes
• Trace — 1× in this paper · explained in 2 episodes

Sources

• V-JEPA 2: Self-Supervised Video Models Enable Understanding, Prediction and Planning
• Seed1.5-VL Technical Report
• Thinking with Spatial Code for Physical-World Video Reasoning
• Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities
• GLM-4.5V and GLM-4.1V-Thinking: Towards Versatile Multimodal Reasoning with Scalable Reinforcement Learning
• Kimi K2: Open Agentic Intelligence
• VisPhyWorld: Probing Physical Reasoning via Code-Driven Video Reconstruction
• MiniMax-01: Scaling Foundation Models with Lightning Attention
• V-JEPA 2.1: Unlocking Dense Features in Video Self-Supervised Learning
• Qwen2.5-VL Technical Report
• Code-as-Room: Generating 3D Rooms from Top-Down View Images via Agentic Code Synthesis
• Vision-as-Inverse-Graphics Agent via Interleaved Multimodal Reasoning
• SceneActBench: Can Agents Act on the 3D Scenes They See?

Related papers

• Loss Knows Best: Detecting Annotation Errors in Videos via Loss Trajectories
• AnchorWeave: World-Consistent Video Generation with Retrieved Local Spatial Memories
• Benchmarking the Robustness of Foundation Models for Mammography under Domain Shift
• MambaX-Net: Dual-Input Mamba-Enhanced Cross-Attention Network for Longitudinal MRI Segmentation
• TeleOCR: Navigating Document Parsing Across Digital and Camera-Captured Documents
• A Survey on Efficient Vision-Language-Action Models