# Industrial Anomaly Understanding on MMAD 1.0 (test) Source: [Industrial Anomaly Understanding on MMAD 1.0 (test)]() Collection index: [llms.txt]() Data updated: 2026-04-15T08:52:33.660Z Evaluation of Industrial Anomaly Understanding performance on the test split of the MMAD 1.0 dataset, reporting metrics for anomaly detection, defect classification, defect localization, defect description, defect analysis, object classification, and object analysis. A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. | Scope | Recorded value | | --- | --- | | Task | Industrial Anomaly Understanding | | Dataset (including any stated split) | MMAD 1.0 (test) | | Metric | Anomaly Detection Score | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 6 of 6 | | Coverage | All results currently returned for this benchmark | Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol. ## Metrics - [Anomaly Detection Score](): Higher is better; unit: PERCENT. - [Defect Classification Score](): Higher is better; unit: PERCENT. - [Defect Localization Score](): Higher is better; unit: PERCENT. - [Defect Description Score](): Higher is better; unit: PERCENT. - [Defect Analysis Score](): Higher is better; unit: PERCENT. - [Object Classification Score](): Higher is better; unit: PERCENT. - [Object Analysis Score](): Higher is better; unit: PERCENT. - [Overall Score](): Higher is better; unit: PERCENT. ## Recorded results | Method | Anomaly Detection Score | Result date | Report paper | | --- | --- | --- | --- | | IAD-Unify | 75.2 | Not recorded | 69df1b7a7d0172140ec7b83b | | Gemini 3 Flash | 73.62 | Not recorded | 69df1b7a7d0172140ec7b83b | | Qwen3.5-4B | 68.3 | Not recorded | 69df1b7a7d0172140ec7b83b | | Show-o2-7B | 65.9 | Not recorded | 69df1b7a7d0172140ec7b83b | | GPT-5-mini | 54.6 | Not recorded | 69df1b7a7d0172140ec7b83b | | Janus-Pro-7B | 51.9 | Not recorded | 69df1b7a7d0172140ec7b83b | ### IAD-Unify Anomaly Detection Score: 75.2. Result ID: 69df50d247f34ce924c82d0c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Unified Models","Evaluation Protocol":"1-shot"}. This record does not assert independent reproduction. ### Gemini 3 Flash Anomaly Detection Score: 73.62. Result ID: 69df50d247f34ce924c82d08. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"API Models"}. This record does not assert independent reproduction. ### Qwen3.5-4B Anomaly Detection Score: 68.3. Result ID: 69df50d247f34ce924c82d0d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Unified Models"}. This record does not assert independent reproduction. ### Show-o2-7B Anomaly Detection Score: 65.9. Result ID: 69df50d247f34ce924c82d0a. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model Category":"Unified Models"}. This record does not assert independent reproduction. ### GPT-5-mini Anomaly Detection Score: 54.6. Result ID: 69df50d247f34ce924c82d09. [Reporting paper]() Recorded evaluation configuration: {"Model Category":"API Models"}. This record does not assert independent reproduction. ### Janus-Pro-7B Anomaly Detection Score: 51.9. Result ID: 69df50d247f34ce924c82d0b. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model Category":"Unified Models"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.