Reasoning-Guided Part-Level Visual Grounding via Reinforcement Learning

A new method called Object-Part Hierarchical Reflective Grounding (OP-HRG) improves part-level visual grounding in multimodal large language models by introducing a coarse-to-fine reasoning pipeline that first localizes the parent object and then the part, outperforming 7B grounding LLMs and SAM3 on PascalPart, PartImageNet, and InstructPart with a 4B model.

arXiv:2607.15374v1 Announce Type: new Abstract: Multimodal large language models MLLMs ground whole objects well from free-form language queries, but they struggle when the query names a part rather than the object. We trace this to a missing object-part hierarchy, since parts are localized in the same single step used for objects. We propose Object-Part Hierarchical Reflective Grounding OP-HRG , a coarse-to-fine reasoning-guided grounding strategy that first localizes the parent object and then the part within it. A self-check then reflects on the result, with an extension to re-encode the predicted crop to inspect the region it is correcting. We introduce a part-aware GRPO framework to train our pipeline with stage-wise rewards. A 4B model trained this way outperforms 7B grounding LLMs and SAM3 across PascalPart, PartImageNet, and InstructPart, and transfers to reasoning segmentation.