← all datasets

A-2A-Bench

Emerging
1papers using it
2026first seen

A-2A-Bench is a manipulation-oriented benchmark that evaluates scene-level, task-conditioned part affordance grounding in everyday scenes, addressing both single-region and multi-region instruction correspondences.

Papers using A-2A-Bench (1)

A-2A-Bench β€” datasets β€” robotics