AI Paper Atlas

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

Q-TriM: Question-Guided Tri-Modal Attention for Audio–Visual Question Answering

SungHun Kim, Seung Baek

ECCV 2026 · Vision-language and multimodal › Visual question answering · approach: Multimodal question answering datasets · 0 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 token filtering (ablation)Q-TriM · MUSIC-AVQA-Ravg accuracy (%): 69.21 -> 70.89Improved
add Tri-Modal attention (ablation)Q-TriM · MUSIC-AVQA-Ravg accuracy (%): 68.92 -> 70.89Improved
add modality mixture (ablation)Q-TriM · MUSIC-AVQA-Ravg accuracy (%): 67.85 -> 70.89Improved
add all three modules to the base (ablation)Q-TriM · MUSIC-AVQA-Ravg accuracy (%): 66.96 -> 70.89Improved
add token filtering (ablation)Q-TriM · MUSIC-AVQAavg accuracy (%): 75.58 -> 77.68Improved
add Tri-Modal attention (ablation)Q-TriM · MUSIC-AVQAavg accuracy (%): 75.41 -> 77.68Improved
use Tri-Modal attention instead of plain cross attention (ablation)Q-TriM · MUSIC-AVQAavg accuracy (%): 76.50 -> 77.68Improved
add modality mixture (ablation)Q-TriM · MUSIC-AVQAavg accuracy (%): 75.01 -> 77.68Improved
keep only 10% of patches instead of 50% (ablation)Q-TriM token filtering · MUSIC-AVQA-Ravg accuracy (%): 70.89 -> 70.11Mixed
vary token keep ratio 0.1 to 0.9 (ablation)Q-TriM token filtering · MUSIC-AVQA-Ravg accuracy (%): 70.11 / 70.56 / 70.89 / 70.39 / 70.50No clear effect
use adaptive keep ratio instead of fixed (ablation)Q-TriM token filtering · MUSIC-AVQAaccuracy (%): 77.68 -> 76.02Hurt
use adaptive keep ratio instead of fixed (ablation)Q-TriM token filtering · MUSIC-AVQA-Raccuracy (%): 70.89 -> 70.57Hurt

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