AI Paper Atlas

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

Discriminability Objective for Training Descriptive Captions

Ruotian Luo, Brian Price, Scott Cohen, Gregory Shakhnarovich

CVPR 2018 · Vision-language and multimodal › Human and embodied modeling · approach: Neural image captioning · 196 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 discriminability loss with λ=1 to ATTN+CIDER (ablation)ATTN caption generator · COCO test 1k4 in 5 human accuracy: 55.95% -> 61.67%Improved
Add discriminability loss with λ=1 to ATTN+CIDER (ablation)ATTN caption generator · COCO test 1kBLEU4: 0.3871 -> 0.3971Improved
Increase discriminability-loss weight from λ=1 to λ=10 (ablation)ATTN+CIDER+DISC · COCO test 1k4 in 5 human accuracy: 61.67% -> 64.63%Improved
Increase discriminability-loss weight from λ=1 to λ=10 (ablation)ATTN+CIDER+DISC · COCO test 1kBLEU4: 0.3971 -> 0.3538Hurt
Add discriminability loss with λ=1 to ATTN+CIDER (ablation)ATTN caption generator · COCO standard split testBLEU4: 0.3566 -> 0.3614Improved
Increase discriminability-loss weight from λ=1 to λ=10 (ablation)ATTN+CIDER+DISC · COCO standard split testBLEU4: 0.3614 -> 0.3203Hurt
Add discriminability loss with λ=1 to ATTN+CIDER (ablation)ATTN caption generator · COCO standard split testCIDEr: 1.1345 -> 1.1425Improved
Add discriminability loss with λ=1 to ATTN+CIDER (ablation)ATTN caption generator · COCO MSCOCO test serverBLEU-4: 0.349 -> 0.352Improved
Use discriminability loss with λ=10 instead of λ=1 (ablation)ATTN+CIDER+DISC · COCO standard split valCIDEr: 1.1381 -> 1.0402Hurt
Add discriminability loss with λ=1 to ATTN+CIDER (ablation)ATTN caption generator · COCO validationDistinct captions: 2640 -> 3235Improved

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