AI Paper Atlas

Research map / Architectures and efficiency / Attention and transformers

Efficient Attention Mechanism for Visual Dialog that can Handle All the Interactions between Multiple Inputs

Van-Quang Nguyen, Masanori Suganuma, Takayuki Okatani

ECCV 2020 · Architectures and efficiency › Attention and transformers · approach: Sparse and structured attention · 55 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 self-attention within each utility (ablation)LTMI · VisDial v1.0 val (ablation)D-NDCG: 61.76 -> 62.72Improved
concatenate attended features instead of adding them (ablation)LTMI · VisDial v1.0 val (ablation)D-NDCG: 60.28 -> 62.72Improved
do not share attention weights across blocks (ablation)LTMI · VisDial v1.0 val (ablation)D-NDCG: 62.50 -> 62.72Improved
use 2 attention blocks instead of 1 (ablation)LTMI · VisDial v1.0 val (ablation)D-NDCG: 62.06 -> 62.72Improved
use 3 blocks instead of 2 (ablation)LTMI · VisDial v1.0 val (ablation)D-NDCG: 62.72 -> 62.48No clear effect
add image context vector c_V to question context (ablation)LTMI · VisDial v1.0 val (ablation)D-NDCG: 61.50 -> 62.72Improved
add dialog-history context c_R (ablation)LTMI · VisDial v1.0 val (ablation)D-NDCG: 62.72 -> 62.37No clear effect
train both decoders jointly vs discriminative only (ablation)LTMI · VisDial v1.0 val (ablation)D-NDCG: 61.80 -> 62.72Improved
train both decoders jointly vs generative only (ablation)LTMI · VisDial v1.0 val (ablation)G-NDCG: 62.35 -> 63.58Improved
use 100 objects instead of 36 (ablation)LTMI · VisDial v1.0 val (ablation)D-NDCG: 62.40 -> 62.72Improved
add positional and spatial embeddings (ablation)LTMI · VisDial v1.0 val (ablation)D-NDCG: 61.84 -> 62.72Improved
replace naive Transformer extension with LTMI attentionLTMI · VisDial v1.0 test-standardNDCG: 55.10 -> 60.92Improved

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