Data as of Jul 25, 2026 · Based on 2,718,867 AI responses across 9,511 prompts · See how Parse measures this
VBackCheckergithub.com · Low evidenceVBackChecker is a visual back-grounding framework for hallucination detection in Multimodal Large Language Models (MLLMs), operating without extra references or experts. It introduces the R2 HalBench benchmark and R Instruct dataset, achieving state-of-the-art performance on hallucination detection and pixel-level grounding tasks.
Parse Score