Researchers have introduced CoMind, a novel dataset designed to study human collaboration by capturing real-world cooking scenarios. This dataset integrates multi-perspective video, audio, gaze tracking, and 3D scans, annotated with details on shared attention, social cues, and agent-object interactions. CoMind aims to facilitate research into the cognitive processes supporting collaboration, particularly Theory of Mind, and establishes benchmarks for tasks like Joint Attention Estimation and Collaborative Handover Prediction. AI
IMPACT Enables development and evaluation of AI systems capable of modeling complex social interactions and reasoning about human behaviors in collaborative environments.
RANK_REASON The cluster contains a research paper introducing a new dataset and benchmarks. [lever_c_demoted from research: ic=1 ai=1.0]
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
