arXivJeffrey M. Girard, Jason Z. Zheng, Jacqueline R. Vertino, Antony D'Avirro, Benjamin PeloquinFri, Jul 31, 2026, 9:33 AM PDT
score 15.3
AI models match humans at reading familiarity from brief chats
Original: FriendBench: Benchmarking Dyadic Familiarity Inference in Humans and Multimodal Large Language Models
Source: arxiv.org ↗
Writing ELI5 summary…