Vision-Language Models
4 stories tagged “Vision-Language Models”.
-
RenderRank: Learning Text Reranking from Compressed Visual Tokens
RenderRank reranks documents using compressed visual tokens, reducing input length 16.5-35.5% while achieving …
-
Self-Evolving Vision-Language Models with Program Verification
VQS reduces VLM label noise via structured image records and program-computed answers, boosting Qwen3-VL up to…
-
SpatialSpeak: QA-Native Reconstruction Leveraging Local and Global Contexts for Spatial Chain-of-Thought Reasoning
SpatialSpeak is a 4B-parameter vision-language framework designed for spatial chain-of-thought reasoning acros…
-
World Action Agent: Leveraging VLMs for Robotic Manipulation through World Action Rehearsal
World Action Agent is a VLM harness for robot manipulation, using contact views, rehearsal and skill transfer …