AI Paper Atlas

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

Learning Distinct and Representative Modes for Image Captioning

Qi Chen, Chaorui Deng, Qi Wu

NeurIPS 2022 · Vision-language and multimodal › Human and embodied modeling · approach: Neural image captioning · 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 Hungarian mode assignment (vs DML w/o Hungarian) (ablation)Transformer-DML k=64 · COCOoracle CIDEr: 1.211 -> 1.871Improved
add fully non-autoregressive objective (vs w/o NAT) (ablation)Transformer-DML k=64 · COCOoracle CIDEr: 1.207 -> 1.871Improved
increase sampled captions for MIC branch from 1 to 5 at batch 64 (ablation)Transformer-DML · COCOoracle CIDEr: 1.871 -> 1.779Hurt
increase sampled captions from 1 to 2 at batch 64 (ablation)Transformer-DML · COCOoracle CIDEr: 1.871 -> 1.833Hurt
reduce total batch size from 64 to 16 (2 captions) (ablation)Transformer-DML · COCOoracle CIDEr: 1.833 -> 1.760Hurt
increase sampled captions from 2 to 5 at batch 16 (ablation)Transformer-DML · COCOoracle CIDEr: 1.760 -> 1.799Improved
mask probability 0.5 instead of 1.0 in masked decoder (ablation)Transformer-DML · COCOoracle CIDEr: 1.871 -> 1.384Hurt
decay mask probability 1.0 -> 0.0 instead of fixed 1.0 (ablation)Transformer-DML · COCOoracle CIDEr: 1.871 -> 1.579Hurt
mask probability 0.0 instead of 1.0 (ablation)Transformer-DML · COCOoracle CIDEr: 1.871 -> 1.207Hurt
replace Transformer-BS with Transformer-DML modesTransformer · COCO diversityDiv-1: 0.21 -> 0.43Improved
replace Transformer-BS with Transformer-DMLTransformer · COCO diversitymBLEU (lower better): 0.78 -> 0.54Improved
use DML vs COS-CVAETransformer-DML · COCO oracle 20 samplesoracle CIDEr: 1.624 -> 1.704Improved

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