# Fill-in-the-blank Video Question Answering on LSMDC (test)

Source: [Fill-in-the-blank Video Question Answering on LSMDC (test)](<https://www.sota2.com/research/sota/fill-in-the-blank-video-question-answering-on-lsmdc-test>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-02-26T05:33:26.633Z

Evaluation of model performance on the test split of the LSMDC dataset for Fill-in-the-blank Video Question Answering, measured by 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 | Fill-in-the-blank Video Question Answering |
| Dataset (including any stated split) | LSMDC (test) |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 0 of 0 |
| 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](<https://www.sota2.com/research/sota/fill-in-the-blank-video-question-answering-on-lsmdc-test.md?metric=Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |

Page 1 of 1. This is an index page, not the full collection.