MMR-AD：面向多模态大语言模型通用异常检测基准的大规模多模态数据集
MMR-AD: A Large-Scale Multimodal Dataset for Benchmarking General Anomaly Detection with Multimodal Large Language Models
提供了一个面向多模态大语言模型通用异常检测的大规模训练与评测基准，并给出基于思维链和强化学习的基线模型 Anomaly-R1，可作为视觉研究者探索 MLLM 在工业异常检测中后训练与推理增强的起点。
原文依据 we present MMR-AD, which is a comprehensive benchmark for both training and evaluating MLLM-based AD models.
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论在工业异常检测的发展进程中，通用异常检测（GAD）是一个新兴趋势，也是最终目标。与传统的单类和多类异常检测不同，通用异常检测旨在训练一个通用的异常检测模型，无需对目标数据进行任何重训练或微调即可直接检测多种新类别中的异常。近年来，多模态大语言模型（MLLM）凭借其革命性的视觉理解与语言推理能力，在实现通用异常检测方面展现出巨大潜力。然而，MLLM 的通用异常检测能力仍未得到充分探索，原因在于：（1）MLLM 在大量来自网络的数据上进行预训练，这些数据与异常检测场景中的数据仍存在显著差距。此外，预训练期间的图像-文本对也并非专门针对异常检测任务。（2）当前主流的异常检测数据集基于图像，尚不适合用于 MLLM 的后训练。为促进基于 MLLM 的通用异常检测研究，我们提出了 MMR-AD，这是一个用于训练和评估基于 MLLM 的异常检测模型的综合基准。借助 MMR-AD，我们揭示出当前最先进的通用 MLLM 的异常检测性能仍远未达到工业要求。基于 MMR-AD，我们还提出了一个基线模型 Anomaly-R1，这是一个基于推理的异常检测模型，它从 MMR-AD 中的思维链数据中学习，并通过强化学习进一步增强。大量实验表明，我们的 Anomaly-R1 在异常检测和定位方面均比通用 MLLM 取得了显著提升。
英文原文摘要 1558 字符
In the progress of industrial anomaly detection, general anomaly detection (GAD) is an emerging trend and also the ultimate goal. Unlike the conventional single- and multi-class AD, general AD aims to train a general AD model that can directly detect anomalies in diverse novel classes without any retraining or fine-tuning on the target data. Recently, Multimodal Large Language Models (MLLMs) have shown great promise in achieving general anomaly detection due to their revolutionary visual understanding and language reasoning capabilities. However, MLLM's general AD ability remains underexplored due to: (1) MLLMs are pretrained on amounts of data sourced from the Web, these data still have significant gaps with the data in AD scenarios. Moreover, the image-text pairs during pretraining are also not specifically for AD tasks. (2) The current mainstream AD datasets are image-based and not yet suitable for post-training MLLMs. To facilitate MLLM-based general AD research, we present MMR-AD, which is a comprehensive benchmark for both training and evaluating MLLM-based AD models. With MMR-AD, we reveal that the AD performance of current SOTA generalist MLLMs still falls far behind the industrial requirements. Based on MMR-AD, we also propose a baseline model, Anomaly-R1, which is a reasoning-based AD model that learns from the CoT data in MMR-AD and is further enhanced by reinforcement learning. Extensive experiments show that our Anomaly-R1 achieves remarkable improvements over generalist MLLMs in both anomaly detection and localization.