A Semantic Decoupling-Based Two-Stage Rainy-Day Attack for Revealing Weather Robustness Deficiencies in Vision-Language Models

• URL: http://arxiv.org/abs/2601.13238v1
• Date: Mon, 19 Jan 2026 17:16:30 GMT
• Title: A Semantic Decoupling-Based Two-Stage Rainy-Day Attack for Revealing Weather Robustness Deficiencies in Vision-Language Models
• Abstract summary: We introduce the first adversarial framework that exploits realistic weather to attack Vision-Language Models (VLMs)<n>We use a two-stage, parameterized perturbation model based on semantic decoupling to analyze rain-induced shifts in decision-making.<n>Our framework generates perturbations that are both physically grounded and interpretable.
• Score: 9.92574905741706
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Vision-Language Models (VLMs) are trained on image-text pairs collected under canonical visual conditions and achieve strong performance on multimodal tasks. However, their robustness to real-world weather conditions, and the stability of cross-modal semantic alignment under such structured perturbations, remain insufficiently studied. In this paper, we focus on rainy scenarios and introduce the first adversarial framework that exploits realistic weather to attack VLMs, using a two-stage, parameterized perturbation model based on semantic decoupling to analyze rain-induced shifts in decision-making. In Stage 1, we model the global effects of rainfall by applying a low-dimensional global modulation to condition the embedding space and gradually weaken the original semantic decision boundaries. In Stage 2, we introduce structured rain variations by explicitly modeling multi-scale raindrop appearance and rainfall-induced illumination changes, and optimize the resulting non-differentiable weather space to induce stable semantic shifts. Operating in a non-pixel parameter space, our framework generates perturbations that are both physically grounded and interpretable. Experiments across multiple tasks show that even physically plausible, highly constrained weather perturbations can induce substantial semantic misalignment in mainstream VLMs, posing potential safety and reliability risks in real-world deployment. Ablations further confirm that illumination modeling and multi-scale raindrop structures are key drivers of these semantic shifts.

Related papers

• WeatherOcc3D: VLM-Assisted Adverse Weather Aware 3D Semantic Occupancy Prediction [0.0]
  We propose a VLM-assisted framework to guide multi-sensor integration via linguistic environmental cues.<n>We utilize a parameter-efficient adapter to align weather-specific text embeddings with sensor features, coupled with a gating strategy.<n>This enables the model to dynamically modulate the fusion ratio - prioritizing semantic camera features in clear daylight and shifting to geometric LiDAR priors during rainy nights.
  arXiv  Detail & Related papers  (2026-05-15T16:12:50Z)
• Cross-Scenario Deraining Adaptation with Unpaired Data: Superpixel Structural Priors and Multi-Stage Pseudo-Rain Synthesis [48.809049043865905]
  Image deraining plays a pivotal role in low-level computer vision.<n>Deep learning paradigms often suffer from severe performance degradation when generalized to unseen Out-of-Distribution scenarios.<n>This paper proposes a pioneering cross-scenario deraining adaptation framework.
  arXiv  Detail & Related papers  (2026-03-23T07:38:55Z)
• Derain-Agent: A Plug-and-Play Agent Framework for Rainy Image Restoration [33.7820079889815]
  We present Derain-Agent, a plug-and-play refinement framework that transitions deraining from static processing to dynamic, agent-based restoration.<n>Derain-Agent equips a base deraining model with two core capabilities: 1) a Planning Network that intelligently schedules an optimal sequence of restoration tools for each instance, and 2) a Strength Modulation mechanism that applies these tools with spatially adaptive intensity.<n>Our method demonstrates strong generalization, consistently boosting the performance of state-of-the-art deraining models on both synthetic and real-world benchmarks.
  arXiv  Detail & Related papers  (2026-03-12T12:38:23Z)
• RoSe: Robust Self-supervised Stereo Matching under Adverse Weather Conditions [58.37558408672509]
  We propose a robust self-supervised training paradigm, consisting of two key steps: robust self-supervised scene correspondence learning and adverse weather distillation.<n>Experiments demonstrate the effectiveness and versatility of our proposed solution, which outperforms existing state-of-the-art self-supervised methods.
  arXiv  Detail & Related papers  (2025-09-23T15:41:40Z)
• Semi-Supervised State-Space Model with Dynamic Stacking Filter for Real-World Video Deraining [73.5575992346396]
  We propose a dual-branch-temporal state-space model to enhance rain streak removal in video sequences.<n>To improve multi-frame feature fusion, we derive a dynamic filter stacking, which adaptively approximates statistical filters for pixel-wise feature refinement.<n>To further explore the capacity of deraining models in supporting other vision-based tasks in rainy environments, we introduce a novel real-world benchmark.
  arXiv  Detail & Related papers  (2025-05-22T15:50:00Z)
• Multi-Modality Driven LoRA for Adverse Condition Depth Estimation [61.525312117638116]
  We propose Multi-Modality Driven LoRA (MMD-LoRA) for Adverse Condition Depth Estimation.<n>It consists of two core components: Prompt Driven Domain Alignment (PDDA) and Visual-Text Consistent Contrastive Learning (VTCCL)<n>It achieves state-of-the-art performance on the nuScenes and Oxford RobotCar datasets.
  arXiv  Detail & Related papers  (2024-12-28T14:23:58Z)
• Learning Robust Precipitation Forecaster by Temporal Frame Interpolation [65.5045412005064]
  We develop a robust precipitation forecasting model that demonstrates resilience against spatial-temporal discrepancies. Our approach has led to significant improvements in forecasting precision, culminating in our model securing textit1st place in the transfer learning leaderboard of the textitWeather4cast'23 competition.
  arXiv  Detail & Related papers  (2023-11-30T08:22:08Z)
• Residual Corrective Diffusion Modeling for Km-scale Atmospheric Downscaling [58.456404022536425]
  State of the art for physical hazard prediction from weather and climate requires expensive km-scale numerical simulations driven by coarser resolution global inputs. Here, a generative diffusion architecture is explored for downscaling such global inputs to km-scale, as a cost-effective machine learning alternative. The model is trained to predict 2km data from a regional weather model over Taiwan, conditioned on a 25km global reanalysis.
  arXiv  Detail & Related papers  (2023-09-24T19:57:22Z)
• Rethinking Real-world Image Deraining via An Unpaired Degradation-Conditioned Diffusion Model [51.49854435403139]
  We propose RainDiff, the first real-world image deraining paradigm based on diffusion models. We introduce a stable and non-adversarial unpaired cycle-consistent architecture that can be trained, end-to-end, with only unpaired data for supervision. We also propose a degradation-conditioned diffusion model that refines the desired output via a diffusive generative process conditioned by learned priors of multiple rain degradations.
  arXiv  Detail & Related papers  (2023-01-23T13:34:01Z)
• Semi-Supervised Video Deraining with Dynamic Rain Generator [59.71640025072209]
  This paper proposes a new semi-supervised video deraining method, in which a dynamic rain generator is employed to fit the rain layer. Specifically, such dynamic generator consists of one emission model and one transition model to simultaneously encode the spatially physical structure and temporally continuous changes of rain streaks. Various prior formats are designed for the labeled synthetic and unlabeled real data, so as to fully exploit the common knowledge underlying them.
  arXiv  Detail & Related papers  (2021-03-14T14:28:57Z)

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.