# Audio-Visual Question Answering on MUSIC-AVQA (Modality Combination Breakdown) Source: [Audio-Visual Question Answering on MUSIC-AVQA (Modality Combination Breakdown)]() Collection index: [llms.txt]() Data updated: 2026-03-24T15:47:31.130Z Evaluation of model performance on the MUSIC-AVQA dataset for Audio-Visual Question Answering, reporting accuracy metrics based on specific combinations of input modalities (e.g., Visual+Language, Visual+Audio+Language). 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-Visual Question Answering | | Dataset (including any stated split) | MUSIC-AVQA | | Metric | Accuracy (V+A+L) | | 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 - [Accuracy (V+A+L)](): Higher is better; unit: PERCENT. - [Accuracy (V+L)](): Higher is better; unit: PERCENT. - [Accuracy (V+I+L)](): Higher is better; unit: PERCENT. - [Accuracy (V+I+A+L)](): Higher is better; unit: PERCENT. - [Average Accuracy](): Higher is better; unit: PERCENT. ## Recorded results | Method | Accuracy (V+A+L) | Result date | Report paper | | --- | --- | --- | --- | | SSAM | 54.68 | Not recorded | 69c21aa10636e5bf4afc49fd | | WUDI | 54.1 | Not recorded | 69c21aa10636e5bf4afc49fd | | TSV | 53.63 | Not recorded | 69c21aa10636e5bf4afc49fd | | Task Arithmetic | 53.62 | Not recorded | 69c21aa10636e5bf4afc49fd | | OptMerge | 51.05 | Not recorded | 69c21aa10636e5bf4afc49fd | | DAMC | 50.91 | Not recorded | 69c21aa10636e5bf4afc49fd | | NaiveMC | 50.66 | Not recorded | 69c21aa10636e5bf4afc49fd | | X-InstructBLIP | 48.34 | Not recorded | 69c21aa10636e5bf4afc49fd | | OneLLM-7B | 47.6 | Not recorded | 69c21aa10636e5bf4afc49fd | | Proj-Only | 46.17 | Not recorded | 69c21aa10636e5bf4afc49fd | | ChatBridge-13B | 43 | Not recorded | 69c21aa10636e5bf4afc49fd | | ImageBind-LLM | 39.72 | Not recorded | 69c21aa10636e5bf4afc49fd | ### SSAM Accuracy (V+A+L): 54.68. Result ID: 69c2b20047f34ce924abe254. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training protocol":"Training-free model merging methods"}. This record does not assert independent reproduction. ### WUDI Accuracy (V+A+L): 54.1. Result ID: 69c2b20047f34ce924abe250. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training protocol":"Training-free model merging methods"}. This record does not assert independent reproduction. ### TSV Accuracy (V+A+L): 53.63. Result ID: 69c2b20047f34ce924abe24f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training protocol":"Training-free model merging methods"}. This record does not assert independent reproduction. ### Task Arithmetic Accuracy (V+A+L): 53.62. Result ID: 69c2b20047f34ce924abe24e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training protocol":"Training-free model merging methods"}. This record does not assert independent reproduction. ### OptMerge Accuracy (V+A+L): 51.05. Result ID: 69c2b20047f34ce924abe251. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training protocol":"Training-free model merging methods"}. This record does not assert independent reproduction. ### DAMC Accuracy (V+A+L): 50.91. Result ID: 69c2b20047f34ce924abe253. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training protocol":"Training-free model merging methods"}. This record does not assert independent reproduction. ### NaiveMC Accuracy (V+A+L): 50.66. Result ID: 69c2b20047f34ce924abe252. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training protocol":"Training-free model merging methods"}. This record does not assert independent reproduction. ### X-InstructBLIP Accuracy (V+A+L): 48.34. Result ID: 69c2b20047f34ce924abe24c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training protocol":"Models finetuned on paired multimodal data"}. This record does not assert independent reproduction. ### OneLLM-7B Accuracy (V+A+L): 47.6. Result ID: 69c2b20047f34ce924abe24a. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Training protocol":"Models finetuned on paired multimodal data"}. This record does not assert independent reproduction. ### Proj-Only Accuracy (V+A+L): 46.17. Result ID: 69c2b20047f34ce924abe24d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training protocol":"Models finetuned on paired multimodal data"}. This record does not assert independent reproduction. ### ChatBridge-13B Accuracy (V+A+L): 43. Result ID: 69c2b20047f34ce924abe249. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training protocol":"Models finetuned on paired multimodal data"}. This record does not assert independent reproduction. ### ImageBind-LLM Accuracy (V+A+L): 39.72. Result ID: 69c2b20047f34ce924abe24b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training protocol":"Models finetuned on paired multimodal data"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.