Cover Image

Semi-off-Policy Reinforcement Learning for Vision-Language Slow-Thinking Reasoning

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Shen, Junhao, Zhao, Haiteng, Gu, Yuzhe, Gao, Songyang, Liu, Kuikun, Huang, Haian, Gao, Jianfei, Lin, Dahua, Zhang, Wenwei, Chen, Kai
Format:	Preprint
Published:	2025
Subjects:	Machine Learning Computer Vision and Pattern Recognition
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Achieving Olympia-Level Geometry Large Language Model Agent via Complexity Boosting Reinforcement Learning
by: Zhao, Haiteng, et al.
Published: (2025)

RIG: Synergizing Reasoning and Imagination in End-to-End Generalist Policy
by: Zhao, Zhonghan, et al.
Published: (2025)

The Imitation Game: Turing Machine Imitator is Length Generalizable Reasoner
by: Hua, Zhouqi, et al.
Published: (2025)

Exploring the Limit of Outcome Reward for Learning Mathematical Reasoning
by: Lyu, Chengqi, et al.
Published: (2025)

Long-horizon Reasoning Agent for Olympiad-Level Mathematical Problem Solving
by: Gao, Songyang, et al.
Published: (2025)

Are Your LLMs Capable of Stable Reasoning?
by: Liu, Junnan, et al.
Published: (2024)

ANAH: Analytical Annotation of Hallucinations in Large Language Models
by: Ji, Ziwei, et al.
Published: (2024)

Mask-DPO: Generalizable Fine-grained Factuality Alignment of LLMs
by: Gu, Yuzhe, et al.
Published: (2025)

ANAH-v2: Scaling Analytical Hallucination Annotation of Large Language Models
by: Gu, Yuzhe, et al.
Published: (2024)

T-Eval: Evaluating the Tool Utilization Capability of Large Language Models Step by Step
by: Chen, Zehui, et al.
Published: (2023)

Agent-FLAN: Designing Data and Methods of Effective Agent Tuning for Large Language Models
by: Chen, Zehui, et al.
Published: (2024)

CIBench: Evaluating Your LLMs with a Code Interpreter Plugin
by: Zhang, Chuyu, et al.
Published: (2024)

From Pixels to Graphs: Open-Vocabulary Scene Graph Generation with Vision-Language Models
by: Li, Rongjie, et al.
Published: (2024)

Think Visually, Reason Textually: Vision-Language Synergy in ARC
by: Zhang, Beichen, et al.
Published: (2025)

CompassVerifier: A Unified and Robust Verifier for LLMs Evaluation and Outcome Reward
by: Liu, Shudong, et al.
Published: (2025)

Fast-Slow Thinking GRPO for Large Vision-Language Model Reasoning
by: Xiao, Wenyi, et al.
Published: (2025)

DAST: Difficulty-Adaptive Slow-Thinking for Large Reasoning Models
by: Shen, Yi, et al.
Published: (2025)

Attention-Guided Reward for Reinforcement Learning-based Jailbreak against Large Reasoning Models
by: Lin, Zheng, et al.
Published: (2026)

OPV: Outcome-based Process Verifier for Efficient Long Chain-of-Thought Verification
by: Wu, Zijian, et al.
Published: (2025)

Think or Not? Selective Reasoning via Reinforcement Learning for Vision-Language Models
by: Wang, Jiaqi, et al.
Published: (2025)

AlchemistCoder: Harmonizing and Eliciting Code Capability by Hindsight Tuning on Multi-source Data
by: Song, Zifan, et al.
Published: (2024)

ExpLang: Improved Exploration and Exploitation in LLM Reasoning with On-Policy Thinking Language Selection
by: Gao, Changjiang, et al.
Published: (2026)

AtomThink: Multimodal Slow Thinking with Atomic Step Reasoning
by: Xiang, Kun, et al.
Published: (2024)

Thinking with Deltas: Incentivizing Reinforcement Learning via Differential Visual Reasoning Policy
by: Gao, Shujian, et al.
Published: (2026)

InternLM-Math: Open Math Large Language Models Toward Verifiable Reasoning
by: Ying, Huaiyuan, et al.
Published: (2024)

InternLM-Law: An Open Source Chinese Legal Large Language Model
by: Fei, Zhiwei, et al.
Published: (2024)

Ada-LEval: Evaluating long-context LLMs with length-adaptable benchmarks
by: Wang, Chonghua, et al.
Published: (2024)

MindSearch: Mimicking Human Minds Elicits Deep AI Searcher
by: Chen, Zehui, et al.
Published: (2024)

CriticEval: Evaluating Large Language Model as Critic
by: Lan, Tian, et al.
Published: (2024)

Navigating the OverKill in Large Language Models
by: Shi, Chenyu, et al.
Published: (2024)

Time Series Forecasting as Reasoning: A Slow-Thinking Approach with Reinforced LLMs
by: Zhou, Yitong, et al.
Published: (2025)

Thinking Slow, Fast: Scaling Inference Compute with Distilled Reasoners
by: Paliotta, Daniele, et al.
Published: (2025)

VisualThink-VLA: Visual Intermediate Reasoning for Effective and Low-Latency Vision-Language-Action Policies
by: Gao, Mingjian, et al.
Published: (2026)

ThinkAct: Vision-Language-Action Reasoning via Reinforced Visual Latent Planning
by: Huang, Chi-Pin, et al.
Published: (2025)

MM-HELIX: Boosting Multimodal Long-Chain Reflective Reasoning with Holistic Platform and Adaptive Hybrid Policy Optimization
by: Zhao, Xiangyu, et al.
Published: (2025)

Fast-Slow Test-Time Adaptation for Online Vision-and-Language Navigation
by: Gao, Junyu, et al.
Published: (2023)

Non-myopic Generation of Language Models for Reasoning and Planning
by: Ma, Chang, et al.
Published: (2024)

Learning to Think Fast and Slow for Visual Language Models
by: Lin, Chenyu, et al.
Published: (2025)

MathBench: Evaluating the Theory and Application Proficiency of LLMs with a Hierarchical Mathematics Benchmark
by: Liu, Hongwei, et al.
Published: (2024)

Interactive Continual Learning: Fast and Slow Thinking
by: Qi, Biqing, et al.
Published: (2024)