← back
arXivSichao Li, Sai Ma, Daniel Kilov, Secil Yanik Guyot, Zhuang Li, Seth LazarWed, Jun 3, 2026, 5:30 AM PDT
score 17.1

Benchmark tests if AI agents justify decisions with visible facts

Original: NoRA: Evaluating Grounded Reasonableness in Visual First-person Normative Action Reasoning

Source: arxiv.org

Writing ELI5 summary…