OPD-IAD: From Language Judgment to Industrial Anomaly Detection via On-Policy Self-Distillation

• URL: http://arxiv.org/abs/2607.18850v1
• Date: Tue, 21 Jul 2026 08:37:01 GMT
• Title: OPD-IAD: From Language Judgment to Industrial Anomaly Detection via On-Policy Self-Distillation
• Abstract summary: Large vision-language models (LVLMs) have recently shown strong potential for industrial anomaly detection (IAD)<n>We propose textbfOPD-IAD, an evidence-privileged dense on-policy self-distillation framework for LVLM-based IAD.<n> OPD-IAD distills privileged defect evidence onto the model's own on-policy judgment trajectory, enabling the final generated judgment to be learned under dense supervision.
• Score: 42.193887957394345
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Large vision-language models (LVLMs) have recently shown strong potential for industrial anomaly detection (IAD) by providing image-level anomaly judgments and interpretable defect reasoning. However, current LVLM-based IAD methods still struggle to produce precise pixel-level anomaly maps from generated language judgments. We aim to achieve precise pixel-level localization while using language as guidance rather than letting it dominate the visual response. Specifically, we propose \textbf{OPD-IAD}, an evidence-privileged dense on-policy self-distillation framework for LVLM-based IAD. OPD-IAD distills privileged defect evidence onto the model's own on-policy judgment trajectory, enabling the final generated judgment to be learned under dense supervision rather than treated only as a textual answer. The resulting judgment serves as a semantic condition for dense anomaly perception. To turn this condition into dense visual evidence, we introduce \textbf{Language-guided Visual Anchoring}, which uses a judgment reforward to re-encode the image and question under the final-judgment condition into semantic anchors and contrasts them with dense visual features through a contrastive heatmap head to generate anomaly maps. The language judgment therefore provides compact semantic guidance, while dense visual features remain the basis for pixel-level scoring, allowing language to guide anomaly localization without letting language quality directly dictate the pixel-level response. Extensive experiments show that OPD-IAD achieves the best overall performance among LVLM-based IAD methods, leading on most image-level, pixel-level, and QA metrics.

Related papers

• Can Vision-Language Models Reason about AI Edits in Images? [78.04125956307838]
  Vision-Language Models (VLMs) offer a promising alternative due to their strong visual understanding and reasoning capabilities.<n>We propose a GRPO-based training framework that utilizes simple accuracy and format rewards.<n>Our approach achieves competitive detection and localization performance compared to state-of-the-art image forgery detectors.
  arXiv  Detail & Related papers  (2026-07-30T16:23:40Z)
• GenAU: Language-Grounded Industrial Anomaly Understanding with Vision-Language Models [12.502139331711673]
  GenAU is a vision-language framework for industrial Anomaly Understanding.<n>It unifies image-level detection, pixel-level segmentation, multi-type anomaly detection, and defect analysis in a single instruction-following model.<n>It attains the strongest image-level detection among CLIP-based zero-shot methods on VisA and Real-IAD.
  arXiv  Detail & Related papers  (2026-07-01T15:11:29Z)
• ExACT: Exemplar-Driven Calibrated Refinement for Training-Free Visual Grounding in Remote Sensing Images [51.53960096934913]
  Remote sensing visual grounding (RSVG) aims to locate specific objects in high-resolution RS imagery using free-form natural language descriptions.<n>Recent advances in multimodal large language models (MLLMs) show great potential for such open-vocabulary RSVG.<n>We propose Exemplar-driven Calibrated Refinement (ExACT), a training-free framework driven by a one-shot visual prompting mechanism.
  arXiv  Detail & Related papers  (2026-06-27T13:50:39Z)
• Reasoning-Driven Anomaly Detection and Localization with Image-Level Supervision [41.94832792442927]
  In this work, we activate the intrinsic reasoning potential of MLLMs to perform anomaly detection, pixel-level localization, and interpretable reasoning solely from image-level supervision.<n>Specifically, we propose Reasoning-Driven Anomaly localization (ReAL), which extracts anomaly-related tokens from the autoregressive reasoning process and aggregates their attention responses to produce pixel-level anomaly maps.
  arXiv  Detail & Related papers  (2026-03-28T07:34:33Z)
• IAD-GPT: Advancing Visual Knowledge in Multimodal Large Language Model for Industrial Anomaly Detection [70.02774285130238]
  This paper explores the combination of rich text semantics with both image-level and pixel-level information from images.<n>We propose IAD-GPT, a novel paradigm based on MLLMs for Industrial Anomaly Detection.<n>Experiments on MVTec-AD and VisA datasets demonstrate our state-of-the-art performance.
  arXiv  Detail & Related papers  (2025-10-16T02:48:05Z)
• Weakly-Supervised Image Forgery Localization via Vision-Language Collaborative Reasoning Framework [16.961220047066792]
  ViLaCo is a vision-language collaborative reasoning framework that introduces auxiliary semantic supervision distilled from pre-trained vision-language models.<n>ViLaCo substantially outperforms existing WSIFL methods, achieving state-of-the-art performance in both detection and localization accuracy.
  arXiv  Detail & Related papers  (2025-08-02T12:14:29Z)
• Multimodal LLM-Guided Semantic Correction in Text-to-Image Diffusion [52.315729095824906]
  MLLM Semantic-Corrected Ping-Pong-Ahead Diffusion (PPAD) is a novel framework that introduces a Multimodal Large Language Model (MLLM) as a semantic observer during inference.<n>It performs real-time analysis on intermediate generations, identifies latent semantic inconsistencies, and translates feedback into controllable signals that actively guide the remaining denoising steps.<n>Extensive experiments demonstrate PPAD's significant improvements.
  arXiv  Detail & Related papers  (2025-05-26T14:42:35Z)
• Do LLMs Understand Visual Anomalies? Uncovering LLM's Capabilities in Zero-shot Anomaly Detection [18.414762007525137]
  Large vision-language models (LVLMs) are proficient in deriving visual representations guided by natural language.<n>Recent explorations have utilized LVLMs to tackle zero-shot visual anomaly detection (VAD) challenges.<n>We present ALFA, a training-free approach designed to address these challenges via a unified model.
  arXiv  Detail & Related papers  (2024-04-15T10:42:22Z)
• Bootstrap Fine-Grained Vision-Language Alignment for Unified Zero-Shot Anomaly Localization [63.61093388441298]
  Contrastive Language-Image Pre-training models have shown promising performance on zero-shot visual recognition tasks. In this work, we propose AnoCLIP for zero-shot anomaly localization.
  arXiv  Detail & Related papers  (2023-08-30T10:35:36Z)
• VoLTA: Vision-Language Transformer with Weakly-Supervised Local-Feature Alignment [52.489874804051304]
  VoLTA is a new vision-language pre-training paradigm that only utilizes image-caption data but fine-grained region-level image understanding. VoLTA pushes multi-modal fusion deep into the uni-modal backbones during pre-training. Experiments on a wide range of vision- and vision-language downstream tasks demonstrate the effectiveness of VoLTA.
  arXiv  Detail & Related papers  (2022-10-09T01:49:58Z)

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.