EgoTools: Towards Tool-Centric Reasoning in Real-World Egocentric Videos
EgoTools adds 100 hours of egocentric tool-use video and a 1,000-question benchmark where current VLMs still struggle.
EgoTools is a suite for egocentric tool-use understanding. EgoTools-Data provides 100 hours of tool-centric recordings with synchronized audio, dense captions, reasoning-heavy narrations, and supplementary 3D information, while EgoTools-Bench has 1,000 QA pairs across perception, geometry, procedure, and causal reasoning. Gemini-3.1-Pro reaches 66.9% overall accuracy but only 51.7% on Perception and Grounding. Full supervised fine-tuning on EgoTools-Data raises Qwen3-VL-8B-Instruct from 50.0% to 60.9% on the full benchmark under strict source-video separation.
48