Cover Image

Why Distillation can Outperform Zero-RL: The Role of Flexible Reasoning

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Hu, Xiao, Lu, Xingyu, Mao, Liyuan, Zhang, YiFan, Zhang, Tianke, Wen, Bin, Yang, Fan, Gao, Tingting, Zhou, Guorui
Format:	Preprint
Published:	2025
Subjects:	Artificial Intelligence
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

ContextRL: Enhancing MLLM's Knowledge Discovery Efficiency with Context-Augmented RL
by: Lu, Xingyu, et al.
Published: (2026)

InstructEngine: Instruction-driven Text-to-Image Alignment
by: Lu, Xingyu, et al.
Published: (2025)

VLM as Policy: Common-Law Content Moderation Framework for Short Video Platform
by: Lu, Xingyu, et al.
Published: (2025)

Thyme: Think Beyond Images
by: Zhang, Yi-Fan, et al.
Published: (2025)

Kwai-STaR: Transform LLMs into State-Transition Reasoners
by: Lu, Xingyu, et al.
Published: (2024)

SpatialReward: Bridging the Perception Gap in Online RL for Image Editing via Explicit Spatial Reasoning
by: Long, Yancheng, et al.
Published: (2026)

R1-Reward: Training Multimodal Reward Model Through Stable Reinforcement Learning
by: Zhang, Yi-Fan, et al.
Published: (2025)

VCap: Hypergeometric Rewards for Weak-to-Strong Visual Captioning
by: Lu, Xingyu, et al.
Published: (2026)

TaskGalaxy: Scaling Multi-modal Instruction Fine-tuning with Tens of Thousands Vision Task Types
by: Chen, Jiankang, et al.
Published: (2025)

SFT-then-RL Outperforms Mixed-Policy Methods for LLM Reasoning
by: Limozin, Alexis, et al.
Published: (2026)

Rethinking On-Device LLM Reasoning: Why Analogical Mapping Outperforms Abstract Thinking for IoT DDoS Detection
by: Pan, William, et al.
Published: (2026)

Why Large Language Models can Secretly Outperform Embedding Similarity in Information Retrieval
by: Benescu, Matei, et al.
Published: (2026)

LoongRL: Reinforcement Learning for Advanced Reasoning over Long Contexts
by: Wang, Siyuan, et al.
Published: (2025)

MCIE: Multimodal LLM-Driven Complex Instruction Image Editing with Spatial Guidance
by: Bai, Xuehai, et al.
Published: (2026)

TIME: Temporal-Sensitive Multi-Dimensional Instruction Tuning and Robust Benchmarking for Video-LLMs
by: Wang, Yunxiao, et al.
Published: (2025)

StreamingCoT: A Dataset for Temporal Dynamics and Multimodal Chain-of-Thought Reasoning in Streaming VideoQA
by: Hu, Yuhang, et al.
Published: (2025)

PEER: Unified Process-Outcome Reinforcement Learning for Structured Empathetic Reasoning
by: Wang, Yunxiao, et al.
Published: (2025)

Debiasing Multimodal Large Language Models via Penalization of Language Priors
by: Zhang, YiFan, et al.
Published: (2024)

Spatial Chain-of-Thought: Bridging Understanding and Generation Models for Spatial Reasoning Generation
by: Chen, Wei, et al.
Published: (2026)

Unlocking Exploration in RLVR: Uncertainty-aware Advantage Shaping for Deeper Reasoning
by: Xie, Can, et al.
Published: (2025)

MoL-RL: Distilling Multi-Step Environmental Feedback into LLMs for Feedback-Independent Reasoning
by: Yang, Kang, et al.
Published: (2025)

HeteroMorpheus: Universal Control Based on Morphological Heterogeneity Modeling
by: Hao, YiFan, et al.
Published: (2024)

Why Softmax Attention Outperforms Linear Attention
by: Deng, Yichuan, et al.
Published: (2023)

Federated Reasoning Distillation Framework with Model Learnability-Aware Data Allocation
by: Guo, Wei, et al.
Published: (2026)

Joint Reward Modeling: Internalizing Chain-of-Thought for Efficient Visual Reward Models
by: Yang, Yankai, et al.
Published: (2026)

The Challenge of Teaching Reasoning to LLMs Without RL or Distillation
by: Du, Wei, et al.
Published: (2025)

Addressing Concept Shift in Online Time Series Forecasting: Detect-then-Adapt
by: Zhang, YiFan, et al.
Published: (2024)

Attention as a Compass: Efficient Exploration for Process-Supervised RL in Reasoning Models
by: Liu, Runze, et al.
Published: (2025)

Why Does Dropping Edges Usually Outperform Adding Edges in Graph Contrastive Learning?
by: Xu, Yanchen, et al.
Published: (2024)

When and Why Does Unsupervised RL Succeed in Mathematical Reasoning? A Manifold Envelopment Perspective
by: Zhang, Zelin, et al.
Published: (2026)

RemoteZero: Geospatial Reasoning with Zero Human Annotations
by: Yao, Liang, et al.
Published: (2026)

Multi-frequency and multi-GNSS PPP phase bias estimation and ambiguity resolution
by: Xiao, Guorui
Published: (2022)

UniRef-Image-Edit: Towards Scalable and Consistent Multi-Reference Image Editing
by: Wei, Hongyang, et al.
Published: (2026)

Skill-Aware Data Selection and Fine-Tuning for Data-Efficient Reasoning Distillation
by: Zhang, Lechen, et al.
Published: (2026)

DAMA: Data- and Model-aware Alignment of Multi-modal LLMs
by: Lu, Jinda, et al.
Published: (2025)

Accelerating RL for LLM Reasoning with Optimal Advantage Regression
by: Brantley, Kianté, et al.
Published: (2025)

AffectGPT-RL: Revealing Roles of Reinforcement Learning in Open-Vocabulary Emotion Recognition
by: Lian, Zheng, et al.
Published: (2026)

Decoupling Contrastive Decoding: Robust Hallucination Mitigation in Multimodal Large Language Models
by: Chen, Wei, et al.
Published: (2025)

Generalized Correspondence Matching via Flexible Hierarchical Refinement and Patch Descriptor Distillation
by: Han, Yu, et al.
Published: (2024)

BrainRing: An Interactive Web-Based Tool for Brain Connectivity Chord Diagram Visualization
by: Fan, Xiao, et al.
Published: (2026)