作者位次 #4/6ACL 2026Findings· Findings of ACL
Spatiotemporal Sycophancy: Negation-Based Gaslighting in Video Large Language Models
Ziyao Tang · Pengkun Jiao · Bin Zhu · Huiyan Qi · Jingjing Chen · Yu-Gang Jiang
摘要
Video Large Language Models (Vid-LLMs) have demonstrated remarkable performance in video understanding tasks, yet their robustness under conversational interaction remains largely underexplored. In this paper, we identify spatiotemporal sycophancy, a failure mode in which Vid-LLMs retract initial...