AI Paper Atlas

Research map / Vision-language and multimodal / Human and embodied modeling

Entangled Transformer for Image Captioning

Guang Li, Linchao Zhu, Ping Liu, Yi Yang

ICCV 2019 · Vision-language and multimodal › Human and embodied modeling · approach: Concept-guided image captioning · 412 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 the transformer encoder (self-attention over semantic inputs) (ablation)Transformer semantic · COCO VGG token-levelCIDEr: 93.1 -> 96.2Improved
add the transformer encoder (self-attention over visual inputs) (ablation)Transformer visual · COCO VGG token-levelCIDEr: 91.8 -> 93.8Improved
replace late fusion of Tv and Ts with ETA feature-level fusion (ablation)ETA-Transformer · COCO SCSTCIDEr: 118.7 -> 126.6Improved
add semantic modality to visual Transformer through ETA+GBCTransformer visual · COCO SCSTCIDEr: 124.1 -> 126.6Improved
add preliminary information injection (ETA) to Parallel Attention (ablation)Parallel Attention · COCO SCSTCIDEr: 124.9 -> 126.6Improved
add GBC to Stacked Attention (visual first) (ablation)Stacked Attention visual · COCO SCSTCIDEr: 125.0 -> 126.6Improved
add GBC to Stacked Attention (semantic first) (ablation)Stacked Attention semantic · COCO SCSTCIDEr: 124.5 -> 126.6Improved
use identity function instead of Sigmoid in GBC (ablation)GBC · COCO token-levelCIDEr: 104.9 -> 115.2Improved
use identity function instead of Tanh in GBC (ablation)GBC · COCO token-levelCIDEr: 104.1 -> 115.2Improved
use identity function instead of ReLU in GBC (ablation)GBC · COCO token-levelCIDEr: 114.0 -> 115.2No clear effect
add semantic attributes alongside visual features (ETA vs Transformer visual only)Transformer · COCO cross-entropyCIDEr: 112.2 -> 117.9Improved
replace LSTM-based Up-Down with ETA-Transformer under SCSTUp-Down · COCO SCSTCIDEr: 120.1 -> 126.6Improved

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