NoisyRollout: Reinforcing Visual Reasoning with Data Augmentation

• URL: http://arxiv.org/abs/2504.13055v3
• Date: Tue, 27 May 2025 02:15:18 GMT
• Title: NoisyRollout: Reinforcing Visual Reasoning with Data Augmentation
• Authors: Xiangyan Liu, Jinjie Ni, Zijian Wu, Chao Du, Longxu Dou, Haonan Wang, Tianyu Pang, Michael Qizhe Shieh,
• Abstract summary: NoisyRollout is a simple yet effective data augmentation method that mixes trajectories from both clean and distorted images during RL training.<n>By injecting targeted diversity in visual perception and the resulting reasoning patterns, NoisyRollout promotes better policy exploration through vision-oriented inductive biases.<n>NoisyRollout achieves state-of-the-art performance among open-source RL-tuned models across $5$ out-of-domain reasoning and perception benchmarks.
• Score: 34.806610134389366
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Recent advances in reinforcement learning (RL) have strengthened the reasoning capabilities of vision-language models (VLMs). However, enhancing policy exploration to better scale test-time compute remains largely underexplored. In addition, VLMs continue to struggle with imperfect visual perception, which in turn affects the subsequent reasoning process. To this end, we propose NoisyRollout, a simple yet effective data augmentation method that mixes trajectories from both clean and moderately distorted images during RL training. By injecting targeted diversity in visual perception and the resulting reasoning patterns, NoisyRollout promotes better policy exploration through vision-oriented inductive biases, ultimately leading to more robust reasoning behaviors. We further adopt a noise annealing schedule that gradually reduces distortion strength over training, leveraging noisy signals early on while ensuring training stability in later stages. Crucially, our method is easy-to-adopt--requiring no additional training cost and no modifications to the RL objective. Extensive experiments on $2$ distinct training datasets demonstrate that NoisyRollout achieves state-of-the-art performance among open-source RL-tuned models across $5$ out-of-domain reasoning and perception benchmarks. Furthermore, we validate the effectiveness of NoisyRollout across model sizes ($7$B and $32$B) and data scales (from $1$K to $6$K), highlighting its generalizability and scalability.

Related papers

• Reinforcement Learning via Implicit Imitation Guidance [49.88208134736617]
  A natural approach is to incorporate an imitation learning objective, either as regularization during training or to acquire a reference policy.<n>We propose to use prior data solely for guiding exploration via noise added to the policy, sidestepping the need for explicit behavior cloning constraints.<n>Our approach achieves up to 2-3x improvement over prior reinforcement learning from offline methods across seven simulated continuous control tasks.
  arXiv  Detail & Related papers  (2025-06-09T07:32:52Z)
• Behavior Injection: Preparing Language Models for Reinforcement Learning [24.46625106928253]
  Reinforcement fine-tuning (RFT) has emerged as a powerful post-training technique to incentivize the reasoning ability of large language models (LLMs)<n>LLMs can respond very inconsistently to RFT: some show substantial performance gains, while others plateau or even degrade.<n>We propose behavior injection, a task-agnostic data-augmentation scheme applied prior to RL.
  arXiv  Detail & Related papers  (2025-05-25T00:54:50Z)
• J1: Exploring Simple Test-Time Scaling for LLM-as-a-Judge [24.607213170485743]
  This paper introduces $textbfJ1-7B$, which is first supervised fine-tuned on reflection-enhanced datasets collected via rejection-sampling.<n>At inference time, we apply Simple Test-Time Scaling (STTS) strategies for additional performance improvement.<n> Experimental results demonstrate that $textbfJ1-7B$ surpasses the previous state-of-the-art LLM-as-a-Judge by $ textbf4.8$% and exhibits a $ textbf5.1$% stronger scaling trend under STTS.
  arXiv  Detail & Related papers  (2025-05-17T06:58:42Z)
• Echo Chamber: RL Post-training Amplifies Behaviors Learned in Pretraining [74.83412846804977]
  Reinforcement learning (RL)-based fine-tuning has become a crucial step in post-training language models.<n>We present a systematic end-to-end study of RL fine-tuning for mathematical reasoning by training models entirely from scratch.
  arXiv  Detail & Related papers  (2025-04-10T17:15:53Z)
• Fast Adaptation with Behavioral Foundation Models [82.34700481726951]
  Unsupervised zero-shot reinforcement learning has emerged as a powerful paradigm for pretraining behavioral foundation models.<n>Despite promising results, zero-shot policies are often suboptimal due to errors induced by the unsupervised training process.<n>We propose fast adaptation strategies that search in the low-dimensional task-embedding space of the pre-trained BFM to rapidly improve the performance of its zero-shot policies.
  arXiv  Detail & Related papers  (2025-04-10T16:14:17Z)
• OpenVLThinker: An Early Exploration to Complex Vision-Language Reasoning via Iterative Self-Improvement [91.88062410741833]
  This study investigates whether similar reasoning capabilities can be successfully integrated into large vision-language models (LVLMs)<n>We consider an approach that iteratively leverages supervised fine-tuning (SFT) on lightweight training data and Reinforcement Learning (RL) to further improve model generalization.<n>OpenVLThinker, a LVLM exhibiting consistently improved reasoning performance on challenging benchmarks such as MathVista, MathVerse, and MathVision, demonstrates the potential of our strategy for robust vision-language reasoning.
  arXiv  Detail & Related papers  (2025-03-21T17:52:43Z)
• Towards Better Alignment: Training Diffusion Models with Reinforcement Learning Against Sparse Rewards [52.90573877727541]
  reinforcement learning (RL) has been considered for diffusion model fine-tuning.<n>RL's effectiveness is limited by the challenge of sparse reward.<n>$textB2text-DiffuRL$ is compatible with existing optimization algorithms.
  arXiv  Detail & Related papers  (2025-03-14T09:45:19Z)
• SFO: Piloting VLM Feedback for Offline RL [1.3597551064547502]
  Vision-Language Models (VLMs) are limited in their ability to solve control tasks due to their lack of action-conditioned training data. A key challenge in Reinforcement Learning from AI Feedback is determining how best to integrate VLM-derived signals into the learning process. We propose a simple yet effective approach--filtered and weighted behavior cloning--consistently outperforms more complex reinforcement learning from human feedback-based methods.
  arXiv  Detail & Related papers  (2025-03-02T23:52:46Z)
• Dynamic Loss-Based Sample Reweighting for Improved Large Language Model Pretraining [55.262510814326035]
  Existing reweighting strategies primarily focus on group-level data importance.<n>We introduce novel algorithms for dynamic, instance-level data reweighting.<n>Our framework allows us to devise reweighting strategies deprioritizing redundant or uninformative data.
  arXiv  Detail & Related papers  (2025-02-10T17:57:15Z)
• E2ED^2:Direct Mapping from Noise to Data for Enhanced Diffusion Models [15.270657838960114]
  Diffusion models have established themselves as the de facto primary paradigm in visual generative modeling.<n>We present a novel end-to-end learning paradigm that establishes direct optimization from the final generated samples to initial noises.<n>Our method achieves substantial performance gains in terms of Fr'eche't Inception Distance (FID) and CLIP score, even with fewer sampling steps.
  arXiv  Detail & Related papers  (2024-12-30T16:06:31Z)
• ResFlow: Fine-tuning Residual Optical Flow for Event-based High Temporal Resolution Motion Estimation [50.80115710105251]
  Event cameras hold significant promise for high-temporal-resolution (HTR) motion estimation.<n>We propose a residual-based paradigm for estimating HTR optical flow with event data.
  arXiv  Detail & Related papers  (2024-12-12T09:35:47Z)
• Avoiding mode collapse in diffusion models fine-tuned with reinforcement learning [0.0]
  Fine-tuning foundation models via reinforcement learning (RL) has proven promising for aligning to downstream objectives. We exploit the hierarchical nature of diffusion models (DMs) and train them dynamically at each epoch with a tailored RL method. We show that models trained with HRF achieve better preservation of diversity in downstream tasks, thus enhancing the fine-tuning robustness and at uncompromising mean rewards.
  arXiv  Detail & Related papers  (2024-10-10T19:06:23Z)
• Representation Alignment for Generation: Training Diffusion Transformers Is Easier Than You Think [72.48325960659822]
  One main bottleneck in training large-scale diffusion models for generation lies in effectively learning these representations.<n>We study this by introducing a straightforward regularization called REPresentation Alignment (REPA), which aligns the projections of noisy input hidden states in denoising networks with clean image representations obtained from external, pretrained visual encoders.<n>The results are striking: our simple strategy yields significant improvements in both training efficiency and generation quality when applied to popular diffusion and flow-based transformers, such as DiTs and SiTs.
  arXiv  Detail & Related papers  (2024-10-09T14:34:53Z)
• VIRL: Volume-Informed Representation Learning towards Few-shot Manufacturability Estimation [0.0]
  This work introduces VIRL, a Volume-Informed Representation Learning approach to pre-train a 3D geometric encoder. The model pre-trained by VIRL shows substantial enhancements on demonstrating improved generalizability with limited data.
  arXiv  Detail & Related papers  (2024-06-18T05:30:26Z)
• Enhancing Noise Robustness of Retrieval-Augmented Language Models with Adaptive Adversarial Training [39.21885486667879]
  Large Language Models (LLMs) exhibit substantial capabilities yet encounter challenges, including hallucination, outdated knowledge, and untraceable reasoning processes. Retrieval-augmented generation (RAG) has emerged as a promising solution, integrating knowledge from external databases to mitigate these challenges. We propose a novel RAG approach known as Retrieval-augmented Adaptive Adrial Training (RAAT)
  arXiv  Detail & Related papers  (2024-05-31T16:24:53Z)
• Reinforcement Learning from Delayed Observations via World Models [10.298219828693489]
  In reinforcement learning settings, agents assume immediate feedback about the effects of their actions after taking them. In practice, this assumption may not hold true due to physical constraints and can significantly impact the performance of learning algorithms. We propose leveraging world models, which have shown success in integrating past observations and learning dynamics, to handle observation delays.
  arXiv  Detail & Related papers  (2024-03-18T23:18:27Z)
• Take the Bull by the Horns: Hard Sample-Reweighted Continual Training Improves LLM Generalization [165.98557106089777]
  A key challenge is to enhance the capabilities of large language models (LLMs) amid a looming shortage of high-quality training data. Our study starts from an empirical strategy for the light continual training of LLMs using their original pre-training data sets. We then formalize this strategy into a principled framework of Instance-Reweighted Distributionally Robust Optimization.
  arXiv  Detail & Related papers  (2024-02-22T04:10:57Z)
• Understanding and Mitigating the Label Noise in Pre-training on Downstream Tasks [91.15120211190519]
  This paper aims to understand the nature of noise in pre-training datasets and to mitigate its impact on downstream tasks. We propose a light-weight black-box tuning method (NMTune) to affine the feature space to mitigate the malignant effect of noise.
  arXiv  Detail & Related papers  (2023-09-29T06:18:15Z)
• VIBR: Learning View-Invariant Value Functions for Robust Visual Control [3.2307366446033945]
  VIBR (View-Invariant Bellman Residuals) is a method that combines multi-view training and invariant prediction to reduce out-of-distribution gap for RL based visuomotor control. We show that VIBR outperforms existing methods on complex visuo-motor control environment with high visual perturbation.
  arXiv  Detail & Related papers  (2023-06-14T14:37:34Z)
• Challenges and Opportunities in Offline Reinforcement Learning from Visual Observations [58.758928936316785]
  offline reinforcement learning from visual observations with continuous action spaces remains under-explored. We show that modifications to two popular vision-based online reinforcement learning algorithms suffice to outperform existing offline RL methods.
  arXiv  Detail & Related papers  (2022-06-09T22:08:47Z)
• AAMDRL: Augmented Asset Management with Deep Reinforcement Learning [5.801876281373619]
  We show how Deep Reinforcement Learning can tackle this challenge. Our contributions are threefold: (i) the use of contextual information also referred to as augmented state in DRL, (ii) the impact of a one period lag between observations and actions, and (iii) the implementation of a new repetitive train test method called walk forward analysis. Although our experiment is on trading bots, it can easily be translated to other bot environments that operate in sequential environment with regime changes and noisy data.
  arXiv  Detail & Related papers  (2020-09-30T03:55:47Z)

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.