AI Paper Atlas

Research map / Vision-language and multimodal / CLIP and zero-shot

SmallCap: Lightweight Image Captioning Prompted With Retrieval Augmentation

Rita Ramos, Bruno Martins, Desmond Elliott, Yova Kementchedjhieva

CVPR 2023 · Vision-language and multimodal › CLIP and zero-shot · approach: Recaptioning and text-supervised pretraining · 128 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 retrieved-caption prompting (ablation)SmallCap · COCO validationCIDEr: 113.4 -> 117.3Improved
Condition on visual features rather than zeroing them (ablation)SmallCap · COCO validationCIDEr: 90.1Improved
Add web and human-labeled data to the datastore (ablation)SmallCap · nocaps test OutCIDEr: 65.0 -> 84.4Improved
Replace COCO captions with target-domain captions in the datastore (ablation)SmallCap · VizWiz validationCIDEr: 34.5 -> 47.7Improved
Add web data to the in-domain datastore (ablation)SmallCap · Flickr30K validationCIDEr: 55.4 -> 58.6Improved
Add human-labeled data to the in-domain datastore (ablation)SmallCap · MSR-VTT validationCIDEr: 29.2 -> 30.9Improved
Combine web and human-labeled data on top of in-domain data (ablation)SmallCap · Flickr30K, VizWiz, and MSR-VTT validationCIDEr: 57.9 vs 58.6; 48.0 vs 48.0; 30.7 vs 30.9Hurt
Add human-labeled data to the in-domain datastore (ablation)SmallCap · VizWiz validationCIDEr: 47.7 -> 47.5No clear effect
Use video-caption data in the datastore (ablation)SmallCap · MSR-VTT validationCIDEr: 29.2 -> 31.1Improved
Reduce the cross-attention dimension from 16 to 8 (ablation)SmallCap · COCO testCIDEr: 120.7 -> 119.1Hurt
Use GPT-2 Large instead of GPT-2 Base (ablation)SmallCap · COCO testCIDEr: 119.7 -> 121.8Improved
Use CLIP ResNet-50x64 rather than ViT-B/32 as the retrieval encoder (ablation)SmallCap · COCO validationCIDEr: 109.5 -> 117.9Improved

Later papers that compared with this method

4 later accepted papers report SmallCap as a baseline in their results tables; 3 of them report a better number on most of the results they share with it.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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