AI Paper Atlas

Research map / Video / Video-language understanding

Open-vocabulary Video Question Answering: A New Benchmark for Evaluating the Generalizability of Video Question Answering Models

Dohwan Ko, Ji Soo Lee, Miso Choi, Jaewon Chu, Jihwan Park, Hyunwoo J. Kim

ICCV 2023 · Video › Video-language understanding · approach: Video question answering · 8 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
replace classification head with answer encoder similarity (OVQA baseline)FrozenBiLM · MSVD-QAmAcc: 17.2 -> 21.7Improved
replace classification head with answer encoder similarity (OVQA baseline)VIOLET · MSVD-QAmAcc: 2.7 -> 10.7Improved
replace classification head with answer encoder similarity (OVQA baseline)baselines · all four datasetsbase-answer accuracy: FrozenBiLM 72.7 -> 72.2 (MSVD B)Mixed
add GNN-based soft verbalizer (ablation)FrozenBiLM+ · MSVD-QAunseen accuracy: 13.7 -> 16.1Improved
add GNN-based soft verbalizer (ablation)FrozenBiLM+ · ActivityNet-QAmAcc: 10.4 -> 12.4Improved
add GNN-based soft verbalizer (ablation)All-in-one+ · ActivityNet-QAmAcc: 6.8 -> 8.7Improved
add GNN-based soft verbalizer (ablation)JustAsk+ · ActivityNet-QAcommon accuracy: 37.1 -> 35.6Mixed
use a hard verbalizer (mean-pooling similar words) instead of no verbalizer (ablation)FrozenBiLM+ · ActivityNet-QAmAcc: 10.4 -> 7.9Hurt
use GNN-based soft verbalizer over learnable MLP verbalizer (ablation)FrozenBiLM+ · ActivityNet-QAmAcc: 10.8 -> 12.4Improved
use DeBERTa instead of CLIP as answer encoder (ablation)All-in-one+ · MSVD-QAmAcc: 5.3 -> 9.4Improved
vary convex combination coefficient epsilon from 1.0 to 0.7 (ablation)FrozenBiLM+ GNN verbalizer · ActivityNet-QAmAcc: 10.4 -> 11.9Improved
set epsilon to 0.0 (use only GNN outputs) (ablation)FrozenBiLM+ GNN verbalizer · ActivityNet-QAtotal accuracy: 43.2 -> 41.6Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy