作者位次 #5/5ACL 2026Poster· Natural Language Processing
Benchmarking Deflection and Hallucination in Large Vision-Language Models
Nicholas Moratelli · Christopher Davis · Leonardo F. R. Ribeiro · Bill Byrne · Gonzalo Iglesias
摘要
Large Vision–Language Models (LVLMs) increasingly rely on retrieval to answer knowledge-intensive multimodal questions. Existing benchmarks overlook conflicts between visual and textual evidence and the importance of generating deflections (e.g., “Sorry, I cannot answer...”) when retrieved knowle...