Cover Image

The Synergy Dilemma of Long-CoT SFT and RL: Investigating Post-Training Techniques for Reasoning VLMs

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Chen, Jierun, Yu, Tiezheng, Bai, Haoli, Yao, Lewei, Wu, Jiannan, Li, Kaican, Mi, Fei, Tao, Chaofan, Zhu, Lei, Zhang, Manyi, Li, Xiaohui, Hou, Lu, Shang, Lifeng, Liu, Qun
Format:	Preprint
Published:	2025
Subjects:	Computation and Language
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

InSight-o3: Empowering Multimodal Foundation Models with Generalized Visual Search
by: Li, Kaican, et al.
Published: (2025)

From Verifiable Dot to Reward Chain: Harnessing Verifiable Reference-based Rewards for Reinforcement Learning of Open-ended Generation
by: Jiang, Yuxin, et al.
Published: (2026)

GRPO-VPS: Enhancing Group Relative Policy Optimization with Verifiable Process Supervision for Effective Reasoning
by: Wang, Jingyi, et al.
Published: (2026)

Quantization Hurts Reasoning? An Empirical Study on Quantized Reasoning Models
by: Liu, Ruikang, et al.
Published: (2025)

SWE-Lego: Pushing the Limits of Supervised Fine-tuning for Software Issue Resolving
by: Tao, Chaofan, et al.
Published: (2026)

When Does RL Help Medical VLMs? Disentangling Vision, SFT, and RL Gains
by: Jeddi, Ahmadreza, et al.
Published: (2026)

ReasonGen-R1: CoT for Autoregressive Image generation models through SFT and RL
by: Zhang, Yu, et al.
Published: (2025)

Benchmarking Post-Training Quantization of Large Language Models under Microscaling Floating Point Formats
by: Zhang, Manyi, et al.
Published: (2026)

REAgent: Requirement-Driven LLM Agents for Software Issue Resolution
by: Kuang, Shiqi, et al.
Published: (2026)

GAC: Noise-Aware Adaptive Mixing for Hybrid SFT-RL Post-Training
by: Hu, Yuelin, et al.
Published: (2026)

From SFT to RL: Demystifying the Post-Training Pipeline for LLM-based Vulnerability Detection
by: Li, Youpeng, et al.
Published: (2026)

FastCode: Fast and Cost-Efficient Code Understanding and Reasoning
by: Li, Zhonghang, et al.
Published: (2026)

Quagmires in SFT-RL Post-Training: When High SFT Scores Mislead and What to Use Instead
by: Kang, Feiyang, et al.
Published: (2025)

Learning to Align Multi-Faceted Evaluation: A Unified and Robust Framework
by: Xu, Kaishuai, et al.
Published: (2025)

Beyond Two-Stage Training: Cooperative SFT and RL for LLM Reasoning
by: Chen, Liang, et al.
Published: (2025)

Visual CoT Makes VLMs Smarter but More Fragile
by: Xu, Chunxue, et al.
Published: (2025)

Chain-of-Probe: Examining the Necessity and Accuracy of CoT Step-by-Step
by: Wang, Zezhong, et al.
Published: (2024)

Empowering Lightweight MLLMs with Reasoning via Long CoT SFT
by: Ou, Linyu, et al.
Published: (2025)

AceReason-Nemotron 1.1: Advancing Math and Code Reasoning through SFT and RL Synergy
by: Liu, Zihan, et al.
Published: (2025)

OVD: On-policy Verbal Distillation
by: Xiong, Jing, et al.
Published: (2026)

Subtle Errors in Reasoning: Preference Learning via Error-injected Self-editing
by: Xu, Kaishuai, et al.
Published: (2024)

BATQuant: Outlier-resilient MXFP4 Quantization via Learnable Block-wise Optimization
by: Li, Ji-Fu, et al.
Published: (2026)

Why Does RL Generalize Better Than SFT? A Data-Centric Perspective on VLM Post-Training
by: Lu, Aojun, et al.
Published: (2026)

SFT Memorizes, RL Generalizes: A Comparative Study of Foundation Model Post-training
by: Chu, Tianzhe, et al.
Published: (2025)

TurnGuide: Enhancing Meaningful Full Duplex Spoken Interactions via Dynamic Turn-Level Text-Speech Interleaving
by: Cui, Wenqian, et al.
Published: (2025)

PepThink-R1: LLM for Interpretable Cyclic Peptide Optimization with CoT SFT and Reinforcement Learning
by: Wang, Ruheng, et al.
Published: (2025)

What Makes Low-Bit Quantization-Aware Training Work for Reasoning LLMs? A Systematic Study
by: Lv, Keyu, et al.
Published: (2026)

CoT‐Based Clinical‐BERT for Risk Stratification After Acute Coronary Syndrome
by: Linhui Mi, et al.
Published: (2025)

QuantClaw: Precision Where It Matters for OpenClaw
by: Zhang, Manyi, et al.
Published: (2026)

Post-Training is About States, Not Tokens: A State Distribution View of SFT, RL, and On-Policy Distillation
by: Nie, Dong
Published: (2026)

First SFT, Second RL, Third UPT: Continual Improving Multi-Modal LLM Reasoning via Unsupervised Post-Training
by: Wei, Lai, et al.
Published: (2025)

KDRL: Post-Training Reasoning LLMs via Unified Knowledge Distillation and Reinforcement Learning
by: Xu, Hongling, et al.
Published: (2025)

RL makes MLLMs see better than SFT
by: Song, Junha, et al.
Published: (2025)

RL Fine-Tuning Heals OOD Forgetting in SFT
by: Jin, Hangzhan, et al.
Published: (2025)

Uniqueness results for skew‐self‐adjoint Dirac system with rectangular potential
by: Tiezheng Li
Published: (2025)

Does Your 3D Encoder Really Work? When Pretrain-SFT from 2D VLMs Meets 3D VLMs
by: Li, Haoyuan, et al.
Published: (2025)

BREAD: Branched Rollouts from Expert Anchors Bridge SFT & RL for Reasoning
by: Zhang, Xuechen, et al.
Published: (2025)

SFT or RL? An Early Investigation into Training R1-Like Reasoning Large Vision-Language Models
by: Chen, Hardy, et al.
Published: (2025)

Bridging SFT and RL: Dynamic Policy Optimization for Robust Reasoning
by: Zhu, Taojie, et al.
Published: (2026)

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