ViSurf: Visual Supervised-and-Reinforcement Fine-Tuning for Large Vision-and-Language Models

• URL: http://arxiv.org/abs/2510.10606v2
• Date: Sun, 09 Nov 2025 08:50:17 GMT
• Title: ViSurf: Visual Supervised-and-Reinforcement Fine-Tuning for Large Vision-and-Language Models
• Abstract summary: Supervised Fine-Tuning (SFT) and Reinforcement Learning with Verifiable Rewards (RLVR) are typical post-training paradigms for Large Vision-and-Language Models (LVLMs)<n>We propose ViSurf, a unified post-training paradigm that integrates the strengths of both SFT and RLVR within a single stage.<n>The core of ViSurf involves injecting ground-truth labels into the RLVR rollouts, thereby providing simultaneous external supervision and internal reinforcement.
• Score: 52.819394001139955
• License: http://creativecommons.org/publicdomain/zero/1.0/
• Abstract: Typical post-training paradigms for Large Vision-and-Language Models (LVLMs) include Supervised Fine-Tuning (SFT) and Reinforcement Learning with Verifiable Rewards (RLVR). SFT leverages external guidance to inject new knowledge, whereas RLVR utilizes internal reinforcement to enhance reasoning capabilities and overall performance. However, our analysis reveals that SFT often leads to sub-optimal performance, while RLVR struggles with tasks that exceed the model's internal knowledge base. To address these limitations, we propose ViSurf (\textbf{Vi}sual \textbf{Su}pervised-and-\textbf{R}einforcement \textbf{F}ine-Tuning), a unified post-training paradigm that integrates the strengths of both SFT and RLVR within a single stage. We analyze the derivation of the SFT and RLVR objectives to establish the ViSurf objective, providing a unified perspective on these two paradigms. The core of ViSurf involves injecting ground-truth labels into the RLVR rollouts, thereby providing simultaneous external supervision and internal reinforcement. Furthermore, we introduce three novel reward control strategies to stabilize and optimize the training process. Extensive experiments across several diverse benchmarks demonstrate the effectiveness of ViSurf, outperforming both individual SFT, RLVR, and two-stage SFT \textrightarrow RLVR. In-depth analysis corroborates these findings, validating the derivation and design principles of ViSurf.

Related papers

• Trust Your Instincts: Confidence-Driven Test-Time RL for Vision-Language-Action Models [16.942019996884273]
  Reinforcement learning (RL) has become indispensable for pushing Vision-Language-Action Models (VLAs) beyond static imitation learning.<n>In this work, we show that VLA models possess useful internal evaluative capabilities.<n>We introduce T2VLA (Test-time VLA), an architecture-agnostic test-time RL framework that enables VLA models to achieve self-bootstrapping policy improvement.
  arXiv  Detail & Related papers  (2026-06-29T07:31:41Z)
• On the Implicit Reward Overfitting and the Low-rank Dynamics in RLVR [51.935533482549545]
  We show that RLVR may exhibit implicit reward overfitting to the training dataset.<n>The model can achieve satisfactory performance on the test set even when its rewards remain relatively low during the training process.
  arXiv  Detail & Related papers  (2026-05-07T16:30:28Z)
• Decouple before Integration: Test-time Synthesis of SFT and RLVR Task Vectors [26.233592394784868]
  We analyze SFT and RLVR through the lens of task vectors.<n>We propose Decoupled Test-time Synthesis (DoTS) to allow SFT and RLVR checkpoints to be trained independently.
  arXiv  Detail & Related papers  (2026-05-01T12:20:44Z)
• Enhancing Radiology Report Generation and Visual Grounding using Reinforcement Learning [15.894854593567963]
  Reinforcement learning can incorporate task-specific feedback, and its combination with explicit intermediate reasoning ("thinking") has demonstrated substantial gains on verifiable math and coding tasks.<n>We build an updated vision-language model based on Qwen3-VL, followed by a cold-start SFT stage that equips the model with basic thinking ability.<n>We find that while strong SFT remains crucial for high base performance, RL provides additional gains on both tasks, whereas explicit thinking does not appear to further improve results.
  arXiv  Detail & Related papers  (2025-12-11T14:36:14Z)
• The Path Not Taken: RLVR Provably Learns Off the Principals [85.41043469428365]
  We show that sparsity is a surface artifact of a model-conditioned optimization bias.<n>We mechanistically explain these dynamics with a Three-Gate Theory.<n>We provide a parameter-level characterization of RLVR's learning dynamics.
  arXiv  Detail & Related papers  (2025-11-11T18:49:45Z)
• SAIL-VL2 Technical Report [65.45818722427506]
  We introduce SAIL-VL2, an open-suite vision foundation model (LVM) for comprehensive multimodal understanding and reasoning.<n>SAIL-VL2 achieves state-of-the-art performance at the 2B and 8B parameter scales across diverse image and video benchmarks.
  arXiv  Detail & Related papers  (2025-09-17T14:34:02Z)
• Empowering Small VLMs to Think with Dynamic Memorization and Exploration [5.2613925143497635]
  Small-scale Vision-Language Models (SVLMs) with reliable thinking capabilities remains fundamentally challenging.<n>Existing training paradigms, including Supervised Fine-Tuning (SFT) and Reinforcement Learning with Verifiable Reward (RLVR), impose substantial demands on the base VLM.<n>We propose DyME, a novel training paradigm that Dynamically selects between Memorization (via SFT) and Exploration (via RLVR) modes at each optimization step.
  arXiv  Detail & Related papers  (2025-06-29T02:19:51Z)
• VLN-R1: Vision-Language Navigation via Reinforcement Fine-Tuning [77.34267241692706]
  Vision-Language Navigation (VLN) is a core challenge in embodied AI, requiring agents to navigate real-world environments using natural language instructions.<n>We propose VLN-R1, an end-to-end framework that leverages Large Vision-Language Models (LVLM) to directly translate egocentric video streams into continuous navigation actions.
  arXiv  Detail & Related papers  (2025-06-20T17:59:59Z)
• Reshaping Reasoning in LLMs: A Theoretical Analysis of RL Training Dynamics through Pattern Selection [35.268183415853976]
  We provide an explanation of the RL training process through empirical analysis and rigorous theoretical modeling.<n>We develop a theoretical framework to understand the training dynamics of RL with two typical rewards: reward (RLVR) and model's internal feedback (RLIF)
  arXiv  Detail & Related papers  (2025-06-05T07:17:04Z)
• SVQA-R1: Reinforcing Spatial Reasoning in MLLMs via View-Consistent Reward Optimization [57.484274282231226]
  We propose SVQA-R1, the first framework to extend R1-style training to spatial VQA.<n>In particular, we introduce Spatial-GRPO, a novel group-wise RL strategy that constructs view-consistent rewards by perturbing spatial relations between objects.<n>Our model, SVQA-R1, not only dramatically improved accuracy on spatial VQA benchmarks but also exhibits interpretable reasoning paths even without using supervised fine-tuning data.
  arXiv  Detail & Related papers  (2025-06-02T06:58:43Z)
• SFT or RL? An Early Investigation into Training R1-Like Reasoning Large Vision-Language Models [39.551767637896404]
  This work revisits the dominant supervised fine-tuning (SFT) then reinforcement learning (RL) paradigm for training Large Vision-Language Models (LVLMs)<n>We show that SFT can significantly undermine subsequent RL by inducing pseudo reasoning paths'' imitated from expert models.<n>We introduce VLAA-Thinking, a new multimodal dataset designed to support reasoning in LVLMs.
  arXiv  Detail & Related papers  (2025-04-10T16:54:05Z)
• SpaceR: Reinforcing MLLMs in Video Spatial Reasoning [70.7401015322983]
  Video spatial reasoning poses a significant challenge for existing Multimodal Large Language Models (MLLMs)<n>This limitation stems primarily from 1) the absence of high-quality datasets for this task, and 2) the lack of effective training strategies to develop spatial reasoning capabilities.<n>Motivated by the success of Reinforcement Learning with Verifiable Reward (RLVR) in unlocking spatial reasoning abilities, this aims to improve MLLMs in video spatial reasoning through the RLVR paradigm.
  arXiv  Detail & Related papers  (2025-04-02T15:12:17Z)
• Reason-RFT: Reinforcement Fine-Tuning for Visual Reasoning of Vision Language Models [42.75418134743927]
  Reason-RFT is a two-stage reinforcement fine-tuning framework for visual reasoning.<n>First,Supervised Fine-Tuning (SFT) with curated CoT data activates the reasoning potential of Vision-Language Models (VLMs)<n>Second, reinforcement learning based on Group Relative Policy Optimization (GRPO) generates multiple reasoning-response pairs to enhance adaptability to domain shifts.
  arXiv  Detail & Related papers  (2025-03-26T17:38:06Z)
• OpenVLThinker: Complex Vision-Language Reasoning via Iterative SFT-RL Cycles [91.88062410741833]
  We introduce OpenVLThinker, one of the first open-source large vision-language models (LVLMs) to exhibit sophisticated chain-of-thought reasoning.<n>We show that OpenVLThinker-7B consistently advances performance across six benchmarks demanding mathematical and general reasoning.
  arXiv  Detail & Related papers  (2025-03-21T17:52:43Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.