Skip to main content
Search

# RealmMatch

1 related articles

Subscribe to updates
  1. arXiv

    DreamRealm turns a model's own hallucinations into edited images: MMVP pair accuracy rises 11.34 points on Qwen3.5-9B

    Multimodal large language models still answer against the image when correct responses hinge on subtle visual details, and existing contrastive supervision is usually built from retrieved or generated images rather than the model's own confusions. DreamRealm mines the incorrect responses a target model assigns high likelihood, minimally edits the original image so each response becomes valid, and trains the model to assign every response to the image that supports it. On Qwen3.5-9B the method improves all seven metrics across six visual perception and hallucination benchmarks, including 7.26 percentage points on HallusionBench accuracy and 11.34 points on MMVP pair accuracy.

You’re all caught up