VLM-R3 uses reinforcement learning to teach a multimodal model when and where to crop an image during its chain-of-thought, improving scores on several visual reasoning benchmarks.
https://huggingface.co/datasets/TheEighthDay/SeekWorld
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
VLM-R$^3$: Region Recognition, Reasoning, and Refinement for Enhanced Multimodal Chain-of-Thought
VLM-R3 uses reinforcement learning to teach a multimodal model when and where to crop an image during its chain-of-thought, improving scores on several visual reasoning benchmarks.