作者位次 #5/8ACL 2026Findings· Findings of ACL
CCTVB ench: Contrastive Consistency Traffic V ideo QA Benchmark for Multimodal LLM s
Xingcheng Zhou · Hao Guo · Rui Song · Walter Zimmer · Mingyu Liu · André Schamschurko · Hu Cao · Alois Christian Knoll
摘要
Safety-critical traffic reasoning requires contrastive consistency: models must detect true hazards when an accident occurs, and reliably reject plausible-but-false hypotheses under near-identical counterfactual scenes. We present CCTVBench, a Contrastive Consistency Traffic VideoQA Benchmark bui...