A4Bench

From the catalogue. Listed by one researcher from one source page.

• id: a4bench
• kind: language and reasoning annotation
• robot or device: images (class: sensor only)
• how collected: curated QA
• size as stated: 1,282 QA pairs in the constitutive part; rest not captured
• hours: not stated | hours only claimed: none | episodes: not stated
• year: 2025
• organisation: not stated
• licence: not stated (class: not stated)
• access: not stated (class: not stated)
• link (paper): https://arxiv.org/abs/2506.00893
• note: Affordance perception benchmark for MLLMs.

Confirm the licence at the link before relying on it.