# Fact-based Visual Question Answering on FVQA 1.0 (test) Source: [Fact-based Visual Question Answering on FVQA 1.0 (test)]() Collection index: [llms.txt]() Data updated: 2026-02-26T06:08:41.850Z Evaluation of model performance on the test split of the FVQA 1.0 dataset, focusing on fact-based visual question answering using WUPS metrics. 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 | Fact-based Visual Question Answering | | Dataset (including any stated split) | FVQA 1.0 (test) | | Metric | WUPS@0.0 (Top-1) | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 13 of 13 | | 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 - [WUPS@0.0 (Top-1)](): Higher is better; unit: PERCENT. - [WUPS@0.0 (Top-3)](): Higher is better; unit: PERCENT. - [WUPS@0.0 (Top-10)](): Higher is better; unit: PERCENT. ## Recorded results | Method | WUPS@0.0 (Top-1) | Result date | Report paper | | --- | --- | --- | --- | | Human | 87.3 | Not recorded | 68fa638d1e209131ce9ff826 | | KB-query model (gt QQmapping) | 73.98 | Not recorded | 68fa638d1e209131ce9ff826 | | KB-query model (top-3-QQmaping) | 72.34 | Not recorded | 68fa638d1e209131ce9ff826 | | Hie-Question+Image+Pre-VQA | 71.51 | Not recorded | 68fa638d1e209131ce9ff826 | | Hie-Question+Image | 66.11 | Not recorded | 68fa638d1e209131ce9ff826 | | KB-query model (top-1-QQmaping) | 64.96 | Not recorded | 68fa638d1e209131ce9ff826 | | LSTM-Question+Image+Pre-VQA | 63.42 | Not recorded | 68fa638d1e209131ce9ff826 | | LSTM-Question+Image | 61.86 | Not recorded | 68fa638d1e209131ce9ff826 | | SVM-Question+Image | 59.97 | Not recorded | 68fa638d1e209131ce9ff826 | | SVM-Image | 59.64 | Not recorded | 68fa638d1e209131ce9ff826 | | LSTM-Image | 59.53 | Not recorded | 68fa638d1e209131ce9ff826 | | SVM-Question | 56.88 | Not recorded | 68fa638d1e209131ce9ff826 | | LSTM-Question | 51.45 | Not recorded | 68fa638d1e209131ce9ff826 | ### Human WUPS@0.0 (Top-1): 87.3. Result ID: 698bfc55f729bc7f68e73b88. [Reporting paper]() Recorded evaluation configuration: {"Type":"Human evaluation"}. This record does not assert independent reproduction. ### KB-query model (gt QQmapping) WUPS@0.0 (Top-1): 73.98. Result ID: 698bfc55f729bc7f68e73b85. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Question-Query mapping":"ground truth"}. This record does not assert independent reproduction. ### KB-query model (top-3-QQmaping) WUPS@0.0 (Top-1): 72.34. Result ID: 698bfc55f729bc7f68e73b87. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Question-Query mapping":"top-3 predicted"}. This record does not assert independent reproduction. ### Hie-Question+Image+Pre-VQA WUPS@0.0 (Top-1): 71.51. Result ID: 698bfc55f729bc7f68e73b84. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Input Modalities":"Question + Image","Architecture":"Hierarchical","Use Pre-VQA":"true"}. This record does not assert independent reproduction. ### Hie-Question+Image WUPS@0.0 (Top-1): 66.11. Result ID: 698bfc55f729bc7f68e73b83. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Input Modalities":"Question + Image","Architecture":"Hierarchical"}. This record does not assert independent reproduction. ### KB-query model (top-1-QQmaping) WUPS@0.0 (Top-1): 64.96. Result ID: 698bfc55f729bc7f68e73b86. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Question-Query mapping":"top-1 predicted"}. This record does not assert independent reproduction. ### LSTM-Question+Image+Pre-VQA WUPS@0.0 (Top-1): 63.42. Result ID: 698bfc55f729bc7f68e73b82. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Input Modalities":"Question + Image","Architecture":"LSTM","Use Pre-VQA":"true"}. This record does not assert independent reproduction. ### LSTM-Question+Image WUPS@0.0 (Top-1): 61.86. Result ID: 698bfc55f729bc7f68e73b81. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Input Modalities":"Question + Image","Architecture":"LSTM"}. This record does not assert independent reproduction. ### SVM-Question+Image WUPS@0.0 (Top-1): 59.97. Result ID: 698bfc55f729bc7f68e73b7e. [Reporting paper]() Recorded evaluation configuration: {"Input Modalities":"Question + Image","Classifier":"SVM"}. This record does not assert independent reproduction. ### SVM-Image WUPS@0.0 (Top-1): 59.64. Result ID: 698bfc55f729bc7f68e73b7d. [Reporting paper]() Recorded evaluation configuration: {"Input Modality":"Image","Classifier":"SVM"}. This record does not assert independent reproduction. ### LSTM-Image WUPS@0.0 (Top-1): 59.53. Result ID: 698bfc55f729bc7f68e73b80. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Input Modality":"Image","Architecture":"LSTM"}. This record does not assert independent reproduction. ### SVM-Question WUPS@0.0 (Top-1): 56.88. Result ID: 698bfc55f729bc7f68e73b7c. [Reporting paper]() Recorded evaluation configuration: {"Input Modality":"Question","Classifier":"SVM"}. This record does not assert independent reproduction. ### LSTM-Question WUPS@0.0 (Top-1): 51.45. Result ID: 698bfc55f729bc7f68e73b7f. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Input Modality":"Question","Architecture":"LSTM"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.