AI Paper Atlas

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

Unicoder-VL: A Universal Encoder for Vision and Language by Cross-Modal Pre-Training

Gen Li, Nan Duan, Yuejian Fang, Ming Gong, Daxin Jiang

AAAI 2020 · Vision-language and multimodal › Human and embodied modeling · approach: Vision-language pretraining · 821 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
Pre-train on image-caption pairs and fine-tune for retrieval rather than train only on task-specific data (ablation)Unicoder-VL · MSCOCO 1K test, sentence retrievalR@1: 75.1 -> 84.3Improved
Pre-train on image-caption pairs and fine-tune for retrieval rather than train only on task-specific data (ablation)Unicoder-VL · Flickr30k 1K test, sentence retrievalR@1: 73.0 -> 86.2Improved
Pre-train on image-caption pairs and fine-tune for retrieval rather than train only on task-specific data (ablation)Unicoder-VL · Flickr30k 1K test, image retrievalR@1: 57.8 -> 71.5Improved
Use pretrained alignment scoring without task-specific fine-tuningUnicoder-VL · Flickr30k 1K test, sentence retrievalR@1: 64.3 -> 86.2Hurt
Increase the encoder depth from 6 to 12 layers (ablation)Unicoder-VL · Flickr30k 1K test, sentence retrievalR@1: 72.4 -> 86.2Improved
Increase the encoder depth from 12 to 24 layers (ablation)Unicoder-VL · Flickr30k 1K test, image retrievalR@1: 71.5 -> 73.6Improved
Increase pre-training data from 0 to 3M image-text pairs (ablation)Unicoder-VL · Flickr30k 1K test, sentence retrievalR@1: 73.0 -> 82.3Improved
Increase pre-training data from 3M to 3.8M image-text pairs (ablation)Unicoder-VL · Flickr30k 1K test, image retrievalR@1: 68.3 -> 71.5Improved
Use cross-modal pre-training for VCR instead of no pre-training (ablation)Unicoder-VL · VCR val, Q→Aaccuracy: 71.6 -> 72.6Improved
Use cross-modal pre-training for VCR instead of no pre-training (ablation)Unicoder-VL · VCR val, Q→ARaccuracy: 52.3 -> 54.5Improved
Use caption-based cross-modal pre-training for cognitive VCR reasoning (ablation)Unicoder-VL · VCR val, Q→Aaccuracy: 71.6 -> 72.6No clear effect

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