Visual Question Answering (VQA) 벤치마크
Visual Question Answering (VQA) on MSVD-QA
Accuracy
- 2017-04-14 — ST-VQA: Accuracy 0.313
- 2018-03-29 — Co-Mem: Accuracy 0.317
- 2019-04-08 — HMEMA: Accuracy 0.337
- 2020-02-25 — HCRN: Accuracy 0.361
- 2021-07-10 — DualVGR: Accuracy 0.39
- 2021-12-17 — ALPRO: Accuracy 0.459
- 2022-03-14 — All-in-one-B: Accuracy 0.483
- 2022-05-27 — GIT: Accuracy 0.568
- 2022-08-01 — Co-Tokenization: Accuracy 486.0
- 2023-03-29 — MaMMUT (ours): Accuracy 602.0