AI Paper Atlas

Research map / Vision-language and multimodal / Visual question answering

Re-evaluating Continual VQA: Toward Fair and Robust Evaluation for Multimodal Continual Learning

Zijian Gao, Zicheng Sun, Xingxing Zhang, Kele Xu, Huaimin Wang

CVPR 2026 · Vision-language and multimodal › Visual question answering · approach: Attention and prior-robust VQA

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 answer prediction distillation (LAPD) with question-only replay (ablation)MaDQ · GQA v2 (PS)FAA: 4.98 -> 36.88Improved
add image-question matching loss (LIQM) on top of LAPD (ablation)MaDQ · GQA v2 (PS)CAA: 41.75 -> 43.36Improved
add matching consistency distillation (LMCD) on top of LAPD and LIQM (ablation)MaDQ · GQA v2 (PS)FFM: 11.07 -> 9.37Improved
add LMCD on top of LAPD and LIQM (ablation)MaDQ · VQA v3 (PS)FFM: 6.83 -> 4.54Improved
use LIQM without any distillation loss (ablation)MaDQ · GQA v2 (PS)FAA: 4.98 -> 5.06No clear effect
plug LIQM into existing continual learning methods (ablation)LwF, ER, CLS-ER · GQA v2 (PS) and VQA v3 (PS)FFM: CLS-ER 11.87 -> 10.71 (GQA v2); 7.56 -> 5.46 (VQA v3)Improved
replace full-triplet replay (ER, CLS-ER) with question-only replay plus dual-level distillation (MaDQ)ER / CLS-ER · VQA v3 (PS)CAA: 33.64 -> 37.82Improved
MaDQ vs ER/CLS-ER under standard splits on GQA v2ER / CLS-ER · GQA v2 (SS)FAA: 41.61 -> 43.51Improved
use token-level disjoint answer spaces (GQA v2 vs original GQA)continual VQA benchmarks · GQA vs GQA v2 (SS)FAA / CAA / FFMHurt
apply simplified MaDQ* to BLIP2 instead of BLIPMaDQ · VQA v3 (SS)FAA: 21.48 (MaDQ*) vs 23.46 (ER) and 24.45 (CLS-ER)Mixed
use BLIP2 LLM-based backbone instead of BLIPcontinual VQA methods · VQA v3 (SS)FAA / CAAHurt
introduce intra-task train-test answer shift (PS) versus standard splitscontinual VQA methods · VQA v3FAA: MaDQ 53.21 -> 36.15; Joint 69.41 -> 42.90Hurt

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