# Discriminative QA on MMAD Source: [Discriminative QA on MMAD]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:47:06.317Z Evaluation of model performance on the MMAD dataset for Discriminative Question Answering, reporting category-specific scores across various component types. 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 | Discriminative QA | | Dataset (including any stated split) | MMAD | | Metric | Industrial Components Score | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 12 of 12 | | 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 - [Industrial Components Score](): Higher is better; unit: PERCENT. - [Electronic Components Score](): Higher is better; unit: PERCENT. - [Construction Materials Score](): Higher is better; unit: PERCENT. - [Mechanical Parts Score](): Higher is better; unit: PERCENT. - [Consumer Products Score](): Higher is better; unit: PERCENT. - [Optical Inspection Score](): Higher is better; unit: PERCENT. - [Average Score](): Higher is better; unit: PERCENT. ## Recorded results | Method | Industrial Components Score | Result date | Report paper | | --- | --- | --- | --- | | MAU-GPT | 89.37 | Not recorded | 698abba8c97a0b3e3b5ddfe6 | | UIO2-Large | 66.61 | Not recorded | 698abba8c97a0b3e3b5ddfe6 | | Gemma-3 | 63.06 | Not recorded | 698abba8c97a0b3e3b5ddfe6 | | Yi-VL | 62.39 | Not recorded | 698abba8c97a0b3e3b5ddfe6 | | AnomalyGPT | 58.14 | Not recorded | 698abba8c97a0b3e3b5ddfe6 | | AnomalyGPT | 57.04 | Not recorded | 698abba8c97a0b3e3b5ddfe6 | | AnomalyGPT | 56.2 | Not recorded | 698abba8c97a0b3e3b5ddfe6 | | UIO2-Xl | 55.55 | Not recorded | 698abba8c97a0b3e3b5ddfe6 | | AnomalyGPT | 50.93 | Not recorded | 698abba8c97a0b3e3b5ddfe6 | | InternVL-2.5 | 46.58 | Not recorded | 698abba8c97a0b3e3b5ddfe6 | | LLaVA-1.5 | 40.1 | Not recorded | 698abba8c97a0b3e3b5ddfe6 | | LLaVA-1.5 | 11.57 | Not recorded | 698abba8c97a0b3e3b5ddfe6 | ### MAU-GPT Industrial Components Score: 89.37. Result ID: 69976ab05698665ec4be493d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Type":"Specialized Model","Number of Parameters":"4B"}. This record does not assert independent reproduction. ### UIO2-Large Industrial Components Score: 66.61. Result ID: 69976ab05698665ec4be4934. [Reporting paper]() Recorded evaluation configuration: {"Type":"Generalist Model","Number of Parameters":"1B"}. This record does not assert independent reproduction. ### Gemma-3 Industrial Components Score: 63.06. Result ID: 69976ab05698665ec4be4936. [Reporting paper]() Recorded evaluation configuration: {"Type":"Generalist Model","Number of Parameters":"4B"}. This record does not assert independent reproduction. ### Yi-VL Industrial Components Score: 62.39. Result ID: 69976ab05698665ec4be4937. [Reporting paper]() Recorded evaluation configuration: {"Type":"Generalist Model","Number of Parameters":"6B"}. This record does not assert independent reproduction. ### AnomalyGPT Industrial Components Score: 58.14. Result ID: 69976ab05698665ec4be493c. [Reporting paper]() Recorded evaluation configuration: {"Type":"Specialized Model","Few-shot settings":"4-shot"}. This record does not assert independent reproduction. ### AnomalyGPT Industrial Components Score: 57.04. Result ID: 69976ab05698665ec4be493b. [Reporting paper]() Recorded evaluation configuration: {"Type":"Specialized Model","Few-shot settings":"3-shot"}. This record does not assert independent reproduction. ### AnomalyGPT Industrial Components Score: 56.2. Result ID: 69976ab05698665ec4be493a. [Reporting paper]() Recorded evaluation configuration: {"Type":"Specialized Model","Few-shot settings":"2-shot"}. This record does not assert independent reproduction. ### UIO2-Xl Industrial Components Score: 55.55. Result ID: 69976ab05698665ec4be4935. [Reporting paper]() Recorded evaluation configuration: {"Type":"Generalist Model","Number of Parameters":"3B"}. This record does not assert independent reproduction. ### AnomalyGPT Industrial Components Score: 50.93. Result ID: 69976ab05698665ec4be4939. [Reporting paper]() Recorded evaluation configuration: {"Type":"Specialized Model","Few-shot settings":"1-shot"}. This record does not assert independent reproduction. ### InternVL-2.5 Industrial Components Score: 46.58. Result ID: 69976ab05698665ec4be4938. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Type":"Generalist Model","Number of Parameters":"8B"}. This record does not assert independent reproduction. ### LLaVA-1.5 Industrial Components Score: 40.1. Result ID: 69976ab05698665ec4be4932. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Type":"Generalist Model","Number of Parameters":"7B"}. This record does not assert independent reproduction. ### LLaVA-1.5 Industrial Components Score: 11.57. Result ID: 69976ab05698665ec4be4933. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Type":"Generalist Model","Number of Parameters":"13B"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.