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

arXiv: https://arxiv.org/abs/2607.18850 · HTML: https://mlsanity.com/paper/2607.18850/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Chart understanding and anomaly detection, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 20 of its 29 references that are in the map.
- First posted on arXiv 21 Jul 2026.

## Placed in

- Multimodal LLMs / vision-language models / [Chart understanding and anomaly detection](https://mlsanity.com/mllm/chart-understanding-and-anomaly-detection/)
- Knowledge distillation / [Detection and dense prediction distillation](https://mlsanity.com/distill/detection-and-dense-prediction-distillation/)

## Builds on (most cited first, 20 of 20)

- [V-Net: Fully Convolutional Neural Networks for Volumetric Medical Image Segmentation](https://mlsanity.com/paper/1606.04797/) (2016)
- [On-Policy Distillation of Language Models: Learning from Self-Generated Mistakes](https://mlsanity.com/paper/2306.13649/) (2023)
- [WinCLIP: Zero-/Few-Shot Anomaly Classification and Segmentation](https://mlsanity.com/paper/2303.14814/) (2023)
- [AnomalyCLIP: Object-agnostic Prompt Learning for Zero-shot Anomaly Detection](https://mlsanity.com/paper/2310.18961/) (2023)
- [AnomalyGPT: Detecting Industrial Anomalies Using Large Vision-Language Models](https://mlsanity.com/paper/2308.15366/) (2023)
- [Segmentation-Based Deep-Learning Approach for Surface-Defect Detection](https://mlsanity.com/paper/1903.08536/) (2019)
- [AdaCLIP: Adapting CLIP with Hybrid Learnable Prompts for Zero-Shot Anomaly Detection](https://mlsanity.com/paper/2407.15795/) (2024)
- [Mixed supervision for surface-defect detection: from weakly to fully supervised learning](https://mlsanity.com/paper/2104.06064/) (2021)
- [A Unified Anomaly Synthesis Strategy with Gradient Ascent for Industrial Anomaly Detection and Localization](https://mlsanity.com/paper/2407.09359/) (2024)
- [Towards Zero-Shot Anomaly Detection and Reasoning with Multimodal Large Language Models](https://mlsanity.com/paper/2502.07601/) (2025)
- [MuSc: Zero-Shot Industrial Anomaly Classification and Segmentation with Mutual Scoring of the Unlabeled Images](https://mlsanity.com/paper/2401.16753/) (2024)
- [MMAD: A Comprehensive Benchmark for Multimodal Large Language Models in Industrial Anomaly Detection](https://mlsanity.com/paper/2410.09453/) (2024)
- [Myriad: Large Multimodal Model by Applying Vision Experts for Industrial Anomaly Detection](https://mlsanity.com/paper/2310.19070/) (2023)
- [AnomalyR1: A GRPO-based End-to-end MLLM for Industrial Anomaly Detection](https://mlsanity.com/paper/2504.11914/) (2025)
- [PKU-GoodsAD: A Supermarket Goods Dataset for Unsupervised Anomaly Detection and Segmentation](https://mlsanity.com/paper/2307.04956/) (2023)
- [IAD-R1: Reinforcing Consistent Reasoning in Industrial Anomaly Detection](https://mlsanity.com/paper/2508.09178/) (2025)
- [VMAD: Visual-enhanced Multimodal Large Language Model for Zero-Shot Anomaly Detection](https://mlsanity.com/paper/2409.20146/) (2024)
- [VisualAD: Language-Free Zero-Shot Anomaly Detection via Vision Transformer](https://mlsanity.com/paper/2603.07952/) (2026)
- [Reasoning-Driven Anomaly Detection and Localization with Image-Level Supervision](https://mlsanity.com/paper/2603.27179/) (2026)
- [AG-VAS: Anchor-Guided Zero-Shot Visual Anomaly Segmentation with Large Multimodal Models](https://mlsanity.com/paper/2603.01305/) (2026)

## Built on by (most cited first, 1 of 1)

- [ADOPD: Reference-Privileged On-Policy Distillation for MLLM-Based Industrial Anomaly Detection](https://mlsanity.com/paper/2608.09789/) (2026)

## Abstract (arXiv)

Large vision-language models (LVLMs) have shown strong potential for industrial anomaly detection (IAD) by providing image-level anomaly judgments and interpretable reasoning. However, reliably translating generated judgments into precise pixel-level localization remains challenging. We propose Mixed-Trajectory On-Policy Distillation for Language-Guided Industrial Anomaly Detection (MOPDA), the first framework to introduce on-policy self-distillation into LVLM-based IAD. For judgment learning, method introduces Mixed-Trajectory Supervision, combining student-generated on-policy trajectories with evidence-conditioned teacher trajectories under a shared token-level distillation objective. Student trajectories preserve supervision on deployment-relevant response paths, while teacher trajectories provide complementary evidence-conditioned supervision. For dense localization, Language-guided Visual Anchoring uses the final judgment as a compact semantic condition to construct image-specific normal and abnormal anchors, which are contrasted with dense visual features to produce anomaly maps. This keeps language as semantic guidance while grounding pixel-level responses in visual evidence. Under a strict cross-dataset zero-shot protocol on five IAD benchmarks, method outperforms the evaluated LVLM-based baselines on most detection, localization, and judgment metrics while remaining competitive with CLIP-based methods. Ablations further validate both mixed-trajectory supervision and final-judgment conditioning.