Region-Level Policy Optimization for Fine-grained MLLM Perception
cs.CV
Submitted: 2026-09-17
Updated: 2026-09-17
Code: https://github.com/YuHengsss/VisionRL2
Terminology
Sources
- Qwen2.5-VL Technical Report
- Mini-o3: Scaling Up Reasoning Patterns and Interaction Turns for Visual Search
- Latent Visual Reasoning
- LLaVA-OneVision: Easy Visual Task Transfer
- Perceive-to-Reason: Decoupling Perception and Reasoning for Fine-Grained Visual Reasoning
- HiDe: Rethinking The Zoom-IN method in High Resolution MLLMs via Hierarchical Decoupling
- Ovis2.5 Technical Report
- What Does Vision Tool-Use Reinforcement Learning Really Learn? Disentangling Tool-Induced and Intrinsic Effects for Crop-and-Zoom
- DocVQA: A Dataset for VQA on Document Images
- Q-Zoom: Query-Aware Adaptive Perception for Efficient Multimodal Large Language Models
- Gemma 4 Technical Report
- Kimi K2: Open Agentic Intelligence
- Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution
- Monet: Reasoning in Latent Visual Space Beyond Images and Language
- Zooming without Zooming: Region-to-Image Distillation for Fine-Grained Multimodal Perception
- V*: Guided Visual Search as a Core Mechanism in Multimodal LLMs
- Vision-OPD: Learning to See Fine Details for Multimodal LLMs via On-Policy Self-Distillation
- Thyme: Think Beyond Images
- DeepEyes: Incentivizing "Thinking with Images" via Reinforcement Learning
Related papers
- Loss Knows Best: Detecting Annotation Errors in Videos via Loss Trajectories
- AnchorWeave: World-Consistent Video Generation with Retrieved Local Spatial Memories
- Benchmarking the Robustness of Foundation Models for Mammography under Domain Shift
- MambaX-Net: Dual-Input Mamba-Enhanced Cross-Attention Network for Longitudinal MRI Segmentation
- TeleOCR: Navigating Document Parsing Across Digital and Camera-Captured Documents
- A Survey on Efficient Vision-Language-Action Models