AI Paper Atlas

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

Oscar: Object-Semantics Aligned Pre-training for Vision-Language Tasks

Xiujun Li, Xi Yin, Chunyuan Li, Pengchuan Zhang, Xiaowei Hu, Lei Zhang, Lijuan Wang, Houdong Hu, Li Dong, Furu Wei, Yejin Choi, Jianfeng Gao

ECCV 2020 · Vision-language and multimodal › CLIP and zero-shot · approach: Recaptioning and text-supervised pretraining · 1,806 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 Visual Genome object tags as anchor points (ablation)Oscar_B · VQA dev (pre-training ablation)accuracy: 70.93 -> 71.70Improved
add Visual Genome object tags (ablation)Oscar_B · COCO 1K text retrievalR@1: 84.4 -> 88.4Improved
add Visual Genome object tags (ablation)Oscar_B · COCO 1K image retrievalR@1: 73.1 -> 75.7Improved
add Visual Genome object tags (ablation)Oscar_B · COCO captioningCIDEr: 115.6 -> 123.4Improved
add Visual Genome object tags (ablation)Oscar_B · COCO captioningBLEU@4: 34.5 -> 36.4Improved
add Open Images object tags (ablation)Oscar_B · VQA dev (pre-training ablation)accuracy: 70.93 -> 71.15Improved
add Open Images object tags (ablation)Oscar_B · COCO 1K image retrievalR@1: 73.1 -> 72.9Hurt
add Open Images object tags (ablation)Oscar_B · COCO captioningCIDEr: 115.6 -> 119.5Improved
use VG-detector tags instead of OI-detector tags (ablation)Oscar_B · COCO captioningCIDEr: 119.5 -> 123.4Improved
add object tags during fine-tuning (learning curves) (ablation)Oscar_B · VQA and retrieval fine-tuningconvergence speedImproved
add object tags as features via attention (w-v vs full attention) (ablation)Oscar_B (no pretraining) · COCO 1K retrievalR@1Improved
scale Oscar from BERT base to BERT largeOscar · COCO captioningCIDEr: 137.6 -> 140.0Improved

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