AI Paper Atlas

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

GRACE: Graph-Based Contextual Debiasing for Fair Visual Question Answering

Yifeng Zhang, Ming Jiang, Qi Zhao

ECCV 2024 · Vision-language and multimodal › Visual question answering · approach: Attention and prior-robust VQA · 1 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 semantic fairness loss LS to graph learning (ablation)GRACE context graph [sF] · VQA-CPaccuracy: 55.38 -> 56.87Improved
add structural fairness loss LT to graph learning (ablation)GRACE context graph [sF] · VQA-CPaccuracy: 55.38 -> 56.48Improved
add fairness losses LS and LT (ablation)GRACE context graph [sF] · OK-VQAaccuracy: 60.36 -> 60.42No clear effect
add reasoning similarity sR to semantic similarity with both fairness losses (ablation)GRACE · GQA-OOD tailaccuracy: 48.93 -> 50.14Improved
add reasoning similarity sR to semantic similarity with both fairness losses (ablation)GRACE · GQA-OODhead-tail gap: 9.31 -> 7.49Improved
add reasoning similarity sR without fairness losses (ablation)GRACE LR baseline · VQA-CPaccuracy: 55.38 -> 54.87Hurt
combine semantic and reasoning similarity for retrieval versus semantic only (ablation)GRACE retrieval · VQA-CPanswer hit rate: 74.82 -> 79.95Improved
swap Prophet for GRACE on different reasonersProphet · VQA-CPaccuracy: 54.02 -> 57.32 (Llama2-7B)Improved
swap Prophet for GRACE on OK-VQA across reasonersProphet · OK-VQAaccuracy: 61.11 -> 60.32 (GPT-3)Mixed
apply LMH and CSS debiasing to ProphetProphet · OK-VQAaccuracy: 61.11 -> 57.28Hurt
apply LMH and CSS debiasing to GRACEGRACE · VQA-CPaccuracy: 57.35 -> 61.37Improved
use GRACE instead of ProphetVQA-CP gender subsets · VQA-CP gendermale-female gap: 4.39 -> 0.90Improved

Later papers that compared with this method

4 later accepted papers report GRACE as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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