Reproduce “ReVEAL: Eliciting Role-Aware Latent Visual Evidence from MLLMs for Reasoning Video Object Segmentation” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Zhongyuan Bao, Cunjie He, Lejun Zhang, Zenglin Xu, Yanwei Fu
ReVEAL segments video objects identified by indirect queries by turning role-marked attention from a fine-tuned multimodal language model into prompts for frozen SAM2. It trains visual evidence tokens and a layer router, avoiding segmenter-alignment training. With Qwen2.5-VL-7B, it scores 59.0 J&F on ReVOS versus DecAF’s 54.2, but on MeViS scores 47.7 versus 48.1.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.