AI Paper Atlas

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

Learn from Your Neighbor: Learning Multi-modal Mappings from Sparse Annotations

Ashwin Kalyan, Stefan Lee, Anitha Kannan, Dhruv Batra

ICML 2018 · Vision-language and multimodal › Human and embodied modeling · approach: Vision-language pretraining · 5 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
transfer neighboring examples’ annotations using learned semantic similaritymulti-label attribute prediction · AWA and CUBAverage Precision@10: AWA: 54.18 -> 57.12 (1 observed label); 63.71 -> 66.28 (20% observed labels); CUB: 35.62 -> 38.83 (1 observed label); 39Improved
transfer neighboring annotations with attention selecting relevant sequence portionsimage captioning · Flickr-8k, Flickr-30k, COCORecall5@100: MLE: 0.71 -> 0.95; 1.30 -> 1.63; 1.45 -> 1.84Improved
transfer neighboring annotations with attention selecting relevant sequence portionsvisually grounded question generation · Flickr, COCO, BingRecall5@100: MLE: 0.35 -> 0.51; 0.33 -> 0.49; 0.37 -> 0.55Improved
add attention to select relevant portions of neighboring captions instead of transferring entire captions (ablation)Caption-Transfer · COCORecall5@100: 1.78 -> 1.84Improved
append neighboring captions as full ground-truth training targets rather than softly weighting supervision (ablation)COCO captioning · COCORecall5@100: 1.84 -> 1.49Hurt
refine the representation space jointly for the task instead of holding it fixed (ablation)COCO captioning · COCORecall5@100: 1.63 -> 1.84Improved
transfer neighboring captions without attention rather than using attention (ablation)Caption-Transfer · Flickr-8kRecall5@100: 0.89 -> 0.95Hurt

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