# Visual Question Answering (Multiple-Choice) on VQA 1.0 (test-standard) Source: [Visual Question Answering (Multiple-Choice) on VQA 1.0 (test-standard)]() Collection index: [llms.txt]() Data updated: 2026-02-27T06:07:04.552Z Evaluation of model performance on the test-standard split of the VQA 1.0 dataset specifically for multiple-choice visual question answering. 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 | Visual Question Answering (Multiple-Choice) | | Dataset (including any stated split) | VQA 1.0 (test-standard) | | Metric | Accuracy (All) | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 27 of 27 | | 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 (All)](): Higher is better; unit: PERCENT. - [Accuracy (Yes/No)](): Higher is better; unit: PERCENT. - [Accuracy (Number)](): Higher is better; unit: PERCENT. - [Accuracy (Other Types)](): Higher is better; unit: PERCENT. ## Recorded results | Method | Accuracy (All) | Result date | Report paper | | --- | --- | --- | --- | | Dual-MFA | 69.97 | Not recorded | 68fa640f1e209131cea026a9 | | DAN | 69 | Not recorded | 68fa63321e209131ce9fd480 | | MLB | 68.89 | Not recorded | 68fa640f1e209131cea026a9 | | VQA-Machine | 67.8 | Not recorded | 68fa640f1e209131cea026a9 | | Ours-ResNet | 67.3 | Not recorded | 68fa62d01e209131ce9fb4a0 | | RAU | 67.3 | Not recorded | 68fa63321e209131ce9fd480 | | MRN | 66.33 | Not recorded | 68fa640f1e209131cea026a9 | | (Kim et al. 2016) ResNet\* | 66.3 | Not recorded | 68fa62d01e209131ce9fb4a0 | | MRN | 66.3 | Not recorded | 68fa63321e209131ce9fd480 | | (Lu et al. 2016) ResNet\* | 66.1 | Not recorded | 68fa62d01e209131ce9fb4a0 | | HieCoAtt | 66.1 | Not recorded | 68fa63321e209131ce9fd480 | | HieCoAtt | 66.07 | Not recorded | 68fa640f1e209131cea026a9 | | QRU | 65.43 | Not recorded | 68fa640f1e209131cea026a9 | | FDA | 64.18 | Not recorded | 68fa640f1e209131cea026a9 | | (Antol et al. 2015) | 63.1 | Not recorded | 68fa62d01e209131ce9fb4a0 | | VQA team | 63.1 | Not recorded | 68fa63321e209131ce9fd480 | | (Noh, Seo, and Han 2016) | 62.7 | Not recorded | 68fa62d01e209131ce9fb4a0 | | DPPnet | 62.7 | Not recorded | 68fa63321e209131ce9fd480 | | DPPnet | 62.69 | Not recorded | 68fa640f1e209131cea026a9 | | DPPnet | 62.69 | Not recorded | 68fa622f1e209131ce9f7ffe | | Region Sel. | 62.43 | Not recorded | 68fa640f1e209131cea026a9 | | (Shih, Singh, and Hoiem 2016) | 62.4 | Not recorded | 68fa62d01e209131ce9fb4a0 | | (Zhou et al. 2015) | 62 | Not recorded | 68fa62d01e209131ce9fb4a0 | | iBOWIMG | 62 | Not recorded | 68fa63321e209131ce9fd480 | | iBOWING | 61.97 | Not recorded | 68fa640f1e209131cea026a9 | | iBOWIMG | 61.97 | Not recorded | 68fa622f1e209131ce9f7ffe | | LSTM Q+I | 57.57 | Not recorded | 68fa640f1e209131cea026a9 | ### Dual-MFA Accuracy (All): 69.97. Result ID: 698c7c1337894ef7d9cebaa6. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### DAN Accuracy (All): 69. Result ID: 698c6ff737894ef7d9ce624b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet"}. This record does not assert independent reproduction. ### MLB Accuracy (All): 68.89. Result ID: 698c7c1337894ef7d9cebaa5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### VQA-Machine Accuracy (All): 67.8. Result ID: 698c7c1337894ef7d9cebaa4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Ours-ResNet Accuracy (All): 67.3. Result ID: 698be21fecace47ef5599085. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"ResNet","type":"Single Model"}. This record does not assert independent reproduction. ### RAU Accuracy (All): 67.3. Result ID: 698c6ff737894ef7d9ce624a. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"ResNet"}. This record does not assert independent reproduction. ### MRN Accuracy (All): 66.33. Result ID: 698c7c1337894ef7d9cebaa2. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### (Kim et al. 2016) ResNet\* Accuracy (All): 66.3. Result ID: 698be21fecace47ef5599083. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet","type":"Single Model"}. This record does not assert independent reproduction. ### MRN Accuracy (All): 66.3. Result ID: 698c6ff737894ef7d9ce6248. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet"}. This record does not assert independent reproduction. ### (Lu et al. 2016) ResNet\* Accuracy (All): 66.1. Result ID: 698be21fecace47ef5599084. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"ResNet","type":"Single Model"}. This record does not assert independent reproduction. ### HieCoAtt Accuracy (All): 66.1. Result ID: 698c6ff737894ef7d9ce6249. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"ResNet"}. This record does not assert independent reproduction. ### HieCoAtt Accuracy (All): 66.07. Result ID: 698c7c1337894ef7d9cebaa3. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### QRU Accuracy (All): 65.43. Result ID: 698c7c1337894ef7d9cebaa1. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### FDA Accuracy (All): 64.18. Result ID: 698c7c1337894ef7d9ceba9f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### (Antol et al. 2015) Accuracy (All): 63.1. Result ID: 698be21fecace47ef559907f. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"type":"Single Model"}. This record does not assert independent reproduction. ### VQA team Accuracy (All): 63.1. Result ID: 698c6ff737894ef7d9ce6247. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### (Noh, Seo, and Han 2016) Accuracy (All): 62.7. Result ID: 698be21fecace47ef5599082. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"type":"Single Model"}. This record does not assert independent reproduction. ### DPPnet Accuracy (All): 62.7. Result ID: 698c6ff737894ef7d9ce6246. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### DPPnet Accuracy (All): 62.69. Result ID: 698c7c1337894ef7d9ceba9e. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### DPPnet Accuracy (All): 62.69. Result ID: 69a1347235dfcabb00e57c6b. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Region Sel. Accuracy (All): 62.43. Result ID: 698c7c1337894ef7d9cebaa0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### (Shih, Singh, and Hoiem 2016) Accuracy (All): 62.4. Result ID: 698be21fecace47ef5599081. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"type":"Single Model"}. This record does not assert independent reproduction. ### (Zhou et al. 2015) Accuracy (All): 62. Result ID: 698be21fecace47ef5599080. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"type":"Single Model"}. This record does not assert independent reproduction. ### iBOWIMG Accuracy (All): 62. Result ID: 698c6ff737894ef7d9ce6245. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### iBOWING Accuracy (All): 61.97. Result ID: 698c7c1337894ef7d9ceba9d. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### iBOWIMG Accuracy (All): 61.97. Result ID: 69a1347235dfcabb00e57c6c. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### LSTM Q+I Accuracy (All): 57.57. Result ID: 698c7c1337894ef7d9ceba9c. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.