# Video Question Answering on TGIF-QA-R (test) Source: [Video Question Answering on TGIF-QA-R (test)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:34:39.222Z Evaluation of model performance on the test split of the TGIF-QA-R dataset for Video Question Answering, focusing on Action, Frame, and Transition 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 | Video Question Answering | | Dataset (including any stated split) | TGIF-QA-R (test) | | Metric | Action Accuracy | | 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 - [Action Accuracy](): Higher is better; unit: PERCENT. - [Frame Accuracy](): Higher is better; unit: PERCENT. - [Transition Accuracy](): Higher is better; unit: PERCENT. ## Recorded results | Method | Action Accuracy | Result date | Report paper | | --- | --- | --- | --- | | MAMA-VIOLET | 0.665 | Not recorded | 69818174a8b01e76cc8e2525 | | LaViLa-VIOLET | 0.643 | Not recorded | 69818174a8b01e76cc8e2525 | | Vid2Seq-VIOLET | 0.642 | Not recorded | 69818174a8b01e76cc8e2525 | | VIOLET | 0.626 | Not recorded | 69818174a8b01e76cc8e2525 | | OmniVL | 0.62 | Not recorded | 69818174a8b01e76cc8e2525 | | MAMA-VGT | 0.617 | Not recorded | 69818174a8b01e76cc8e2525 | | MERLOT | 0.614 | Not recorded | 69818174a8b01e76cc8e2525 | | VGT | 0.61 | Not recorded | 69818174a8b01e76cc8e2525 | | VindLU | 0.595 | Not recorded | 69818174a8b01e76cc8e2525 | | LAVENDER | 0.571 | Not recorded | 69818174a8b01e76cc8e2525 | | Singularity | 0.531 | Not recorded | 69818174a8b01e76cc8e2525 | | All-in-one | 0.349 | Not recorded | 69818174a8b01e76cc8e2525 | ### MAMA-VIOLET Action Accuracy: 0.665. Result ID: 699767f85698665ec4bb5aa0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"adaptation":"VIOLET"}. This record does not assert independent reproduction. ### LaViLa-VIOLET Action Accuracy: 0.643. Result ID: 699767f85698665ec4bb5a9d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"adaptation":"VIOLET"}. This record does not assert independent reproduction. ### Vid2Seq-VIOLET Action Accuracy: 0.642. Result ID: 699767f85698665ec4bb5a9e. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"adaptation":"VIOLET"}. This record does not assert independent reproduction. ### VIOLET Action Accuracy: 0.626. Result ID: 699767f85698665ec4bb5a9b. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### OmniVL Action Accuracy: 0.62. Result ID: 699767f85698665ec4bb5a98. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### MAMA-VGT Action Accuracy: 0.617. Result ID: 699767f85698665ec4bb5a9f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"adaptation":"VGT"}. This record does not assert independent reproduction. ### MERLOT Action Accuracy: 0.614. Result ID: 699767f85698665ec4bb5a95. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### VGT Action Accuracy: 0.61. Result ID: 699767f85698665ec4bb5a9a. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### VindLU Action Accuracy: 0.595. Result ID: 699767f85698665ec4bb5a99. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### LAVENDER Action Accuracy: 0.571. Result ID: 699767f85698665ec4bb5a96. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Singularity Action Accuracy: 0.531. Result ID: 699767f85698665ec4bb5a97. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### All-in-one Action Accuracy: 0.349. Result ID: 699767f85698665ec4bb5a9c. [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.