Adaptive Look, Flexible Thought: Learning How and How Long to Think in Multimodal Reasoning
cs.CV, cs.AI, cs.CL
Submitted: 2026-10-11
Updated: 2026-10-11
Code: https://github.com/hiyouga/EasyR1
Terminology
Sources
- Ground-R1: Incentivizing Grounded Visual Reasoning via Reinforcement Learning
- Latent Visual States for Efficient Multimodal Reasoning
- ATLAS: Agentic or Latent Visual Reasoning? One Word is Enough for Both
- DeepEyesV2: Toward Agentic Multimodal Model
- Mull-Tokens: Modality-Agnostic Latent Thinking
- Kimi k1.5: Scaling Reinforcement Learning with LLMs
- Qwen3 Technical Report
- More Thought, Less Accuracy? On the Dual Nature of Reasoning in Vision-Language Models
- SeePhys Pro: Diagnosing Modality Transfer and Blind-Training Effects in Multimodal RLVR for Physics Reasoning
- Think When You Need: Self-Adaptive Chain-of-Thought Learning
- Group Sequence Policy Optimization
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