NAS-PO: Native Attention-Strategy Policy Optimization for Vision-Language Models
cs.CV
Submitted: 2026-11-01
Updated: 2026-11-01
Terminology
Sources
- Qwen3-VL Technical Report
- DRA-GRPO: Your GRPO Needs to Know Diverse Reasoning Paths for Mathematical Reasoning
- Agentic Entropy-Balanced Policy Optimization
- OneThinker: All-in-one Reasoning Model for Image and Video
- Spotlight on Token Perception for Multimodal Reinforcement Learning
- Vision-R1: Incentivizing Reasoning Capability in Multimodal Large Language Models
- Reinforced Attention Learning
- Understanding R1-Zero-Like Training: A Critical Perspective
- MM-Eureka: Exploring the Frontiers of Multimodal Reasoning with Rule-based Reinforcement Learning
- Seeing with You: Perception-Reasoning Coevolution for Multimodal Reasoning
- CPPO: Contrastive Perception Policy Optimization for VLM Agents
- Proximal Policy Optimization Algorithms
- DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
- Perception-Aware Policy Optimization for Multimodal Reasoning
- Teaching the Way, Not the Answer: Privileged Tutoring Distillation for Multimodal Policy Optimization
- LogicVista: Multimodal LLM Logical Reasoning Benchmark in Visual Contexts
- Vision-OPD: Learning to See Fine Details for Multimodal LLMs via On-Policy Self-Distillation
- VAPO: Efficient and Reliable Reinforcement Learning for Advanced Reasoning Tasks
- Geometric-Mean Policy Optimization
- Group Sequence Policy Optimization
Related papers
- TEA: Text Encoder Alignment for Robust Concept Erasure in Text-to-Image Models
- Do Egocentric Video-Language Models Capture Both Hand- and Object-Centric Cues?
- Aura: Consistent Multi-Subject Video Generation via VLM-Grounded Semantic Alignment
- PVRF: All-in-one Adverse Weather Removal via Prior-modulated and Velocity-constrained Rectified Flow
- DLWM: Diverse Latent World Models for Efficient Multimodal Reasoning
- Out of Sight, Still in Mind: Token Compression for Omni-LLMs