AI Paper Atlas

Research map / Video / Video-language understanding

VideoQA in the Era of LLMs: An Empirical Study

Junbin Xiao, Nanxin Huang, Hangyu Qin, Dongyang Li, Yicong Li, Fengbin Zhu, Zhulin Tao, Jianxing Yu, Liang Lin, Tat‐Seng Chua, Angela Yao

IJCV 2025 · Video › Video-language understanding · approach: Video question answering · 28 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
add TD-style data (about 4.5K) to NExT-QA trainingFrozenGQA · NExT-QA temporal description (TD)accuracy: 32.3 -> 92.5Improved
add TD-style data to NExT-QA trainingFrozenGQA · NExT-QA temporal exchange (TE)accuracy: 46.1 -> 52.0Improved
add TD-style data to NExT-QA trainingFrozenGQA · NExT-QA temporal original (TO)accuracy: 67.7 -> 68.3Improved
prompt GPT-4o to also output indexes of frames it usesGPT-4o · NExT-GQA NormalVQAaccuracy: 49.0 -> 64.7Improved
remove video input (BlindQA) for finetuned vs zero-shot models (ablation)CoVGT vs GPT-4 · NExT-GQA BlindQAaccuracy: 49.3 vs 45.3Mixed
rewrite answer options to remove question-answer and video-answer shortcutsVideo-LLMs · NExT-QA validation subsetaccuracy: 10% ~ 20% declineHurt
edit options to remove video-answer shortcut vs question-answer shortcutGPT-4o · NExT-QAaccuracy drop: 2.6% (QA) vs 9.9% (VA)Mixed
add spoken phrases before the questionGPT-4o · NExT-QAaccuracy: +3%Improved
shuffle video framesVideo-LLMs · NExT-QA and ActivityNet-QAaccuracyNo clear effect
shuffle question wordsVideo-LLMs · MCQA and OEQAaccuracy retained: 86%~97% (MCQA), 70%~80% (OEQA)Hurt
scale LLaMA-VID LLM from 7B to 13B (ablation)LLaMA-VID · OEQA with spoken prefixesaccuracyImproved
train on temporal or causal questions instead of descriptive questions for testing on descriptiveSeViLA and LLaMA-VQA · NExT-QA descriptive questionsaccuracyNo clear effect

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