Metraipolis › archive › dc_e06edbfb37b3

Region-Level Policy Optimization for Fine-grained MLLM Perception — arXiv:2609.19745

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.19745
License (as archived)
CC0-1.0
Archived
2026-10-10T05:05:31.332Z
SHA-256 of the archived text
ecf99aa7b275a077f267e6a0a999ab48b69769bc6bbdfe903b4406111c325862
Stable URL of this page
https://metraipolis.com/documento/dc_e06edbfb37b3

How to cite

Region-Level Policy Optimization for Fine-grained MLLM Perception — arXiv:2609.19745 — arXiv (en), https://arxiv.org/abs/2609.19745 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_e06edbfb37b3 (Metraipolis archive dc_e06edbfb37b3, sha256 ecf99aa7b275a077f267e6a0a999ab48b69769bc6bbdfe903b4406111c325862)
BibTeX
@misc{dce06edbfb37b3,
  title = {Region-Level Policy Optimization for Fine-grained MLLM Perception — arXiv:2609.19745},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.19745},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_e06edbfb37b3; sha256 ecf99aa7b275a077f267e6a0a999ab48b69769bc6bbdfe903b4406111c325862},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.19745. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Fine-grained visual perception in MLLMs is commonly improved by raising the resolution, but the added visual tokens inflate vision-encoding and language-model prefilling costs. We show that the two operations underlying fine-grained perception, localizing the region of interest (RoI) and recognizing its content, have different resolution requirements. In a controlled diagnostic, localization tolerates roughly 3 to 4 times stronger token compression than recognition, which motivates localizing from a coarse view and concentrating resolution on the selected evidence. Decoding coordinates with the MLLM can be trained end-to-end from answers, but costs a full model pass per query and depends on grounding ability. A lightweight proposal network distilled from the model's attention is fast, but inherits the noise of its attention targets. The RoI from the proposal network reaches the answer through a discrete region choice, so its faithfulness to the answer cannot supervise the network. We therefore optimize the proposal network with region-level reinforcement learning, which we call Vision-RL2. It treats coherent regions as actions, and a frozen MLLM reader scores each one by how its removal changes the answer likelihood. Complementary subtractive and additive objectives suppress distracting proposals and recover missing evidence, updating only the predictor without region annotations, response sampling, or reasoning trajectories. The refined proposal further enables a sparse encoding that magnifies evidence and excludes background tokens. Across six fine-grained benchmarks and four MLLM backbones, Vision-RL2 improves accuracy over the base model at every token budget and surpasses its largest-budget accuracy with about 4 times fewer visual tokens. Code is available at this https URL .

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-18 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2609.19745 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-10T04:26:09.533Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-18?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).