RelCheck: Dual-Evidence Spatial Grounding for VLM Hallucination Correction
Multimodal large language models (MLLMs) fre- quently generate text that is inconsistent with the input image. While object- and attribute-level hallucinations have received considerable attention, relational hallucinations (incorrect de- scriptions of spatial or interactive relationships between objects) remain largel...
Siddhi Patil, N. Saxena, William B. Andreopoulos
· 0 citations