# Audio Reasoning on MMAR Agent Track Source: [Audio Reasoning on MMAR Agent Track]() Collection index: [llms.txt]() Data updated: 2026-03-19T10:02:09.904Z Evaluation of Audio Reasoning performance specifically on the MMAR Agent Track dataset, reporting results based on MMAR-Rubrics and general Accuracy. 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 | Audio Reasoning | | Dataset (including any stated split) | MMAR Agent Track | | Metric | Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 8 of 8 | | 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 - [Accuracy](): Higher is better; unit: PERCENT. - [MMAR Rubrics](): Higher is better; unit: PERCENT. ## Recorded results | Method | Accuracy | Result date | Report paper | | --- | --- | --- | --- | | Team B | 77.4 | Not recorded | 69bb83060636e5bf4afc19f2 | | TalTech | 76.9 | Not recorded | 69bb83060636e5bf4afc19f2 | | Team C | 75.1 | Not recorded | 69bb83060636e5bf4afc19f2 | | Gemini 2.5 Pro | 74.7 | Not recorded | 69bb83060636e5bf4afc19f2 | | Team D | 72.2 | Not recorded | 69bb83060636e5bf4afc19f2 | | Team E | 71 | Not recorded | 69bb83060636e5bf4afc19f2 | | Qwen3-Omni-Thinking | 66.4 | Not recorded | 69bb83060636e5bf4afc19f2 | | GPT-4o Audio | 63.5 | Not recorded | 69bb83060636e5bf4afc19f2 | ### Team B Accuracy: 77.4. Result ID: 69bbc99847f34ce924a63de2. [Reporting paper]() Recorded evaluation configuration: {"Model Category":"Agent-based"}. This record does not assert independent reproduction. ### TalTech Accuracy: 76.9. Result ID: 69bbc99847f34ce924a63de1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Agent-based"}. This record does not assert independent reproduction. ### Team C Accuracy: 75.1. Result ID: 69bbc99847f34ce924a63de3. [Reporting paper]() Recorded evaluation configuration: {"Model Category":"Agent-based"}. This record does not assert independent reproduction. ### Gemini 2.5 Pro Accuracy: 74.7. Result ID: 69bbc99847f34ce924a63de6. [Reporting paper]() Recorded evaluation configuration: {"Model Category":"Single-model"}. This record does not assert independent reproduction. ### Team D Accuracy: 72.2. Result ID: 69bbc99847f34ce924a63de4. [Reporting paper]() Recorded evaluation configuration: {"Model Category":"Agent-based"}. This record does not assert independent reproduction. ### Team E Accuracy: 71. Result ID: 69bbc99847f34ce924a63de5. [Reporting paper]() Recorded evaluation configuration: {"Model Category":"Agent-based"}. This record does not assert independent reproduction. ### Qwen3-Omni-Thinking Accuracy: 66.4. Result ID: 69bbc99847f34ce924a63de8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Single-model"}. This record does not assert independent reproduction. ### GPT-4o Audio Accuracy: 63.5. Result ID: 69bbc99847f34ce924a63de7. [Reporting paper]() Recorded evaluation configuration: {"Model Category":"Single-model"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.