AI Paper Atlas

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

X2-VLM: All-in-One Pre-Trained Model for Vision-Language Tasks

Yan Zeng, Xinsong Zhang, Hang Li, Jiawei Wang, Jipeng Zhang, Wangchunshu Zhou

TPAMI 2023 · Vision-language and multimodal › CLIP and zero-shot · approach: Self-distillation for vision-language pretraining · 123 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 multi-grained aligning (ablation)X2-VLM · Flickr30K TRR@1: 96.6 -> 98.0Improved
add multi-grained aligning (ablation)X2-VLM · OVADmAP: 23.1 -> 27.9Improved
add box prediction loss (ablation)X2-VLM · RefCOCO+ testBaccuracy: 66.0 -> 76.7Improved
add box prediction loss (ablation)X2-VLM · Flickr30K IRR@1: 89.6 -> 89.0Hurt
add region data (ablation)X2-VLM · OVADmAP: 22.3 -> 27.9Improved
add region data (ablation)X2-VLM · RefCOCO+ testBaccuracy: 69.3 -> 76.7Improved
add object data (ablation)X2-VLM · Flickr30K IRR@1: 86.8 -> 89.0Improved
add both multi-grained aligning and box prediction (full X2-VLM) (ablation)X2-VLM · RefCOCO+ testAaccuracy: 78.6 -> 88.6Improved
add both multi-grained aligning and box prediction (full X2-VLM) (ablation)X2-VLM · OVADmAP: 20.6 -> 27.9Improved
use temporal attention instead of average pooling over frames (ablation)X2-VLM · MSRVTT retrievalIR R@1: 47.6 -> 45.6Hurt
use temporal attention instead of average pooling over frames (ablation)X2-VLM · MSVD-QAaccuracy: 52.8 -> 52.1Hurt
replace text encoder with XLM-R, no multilingual pretrainingX2-VLM base · Flickr30K DEavg recall: 93.3 -> 94.0Improved

Later papers that compared with this method

1 later accepted paper reports X2VLM base as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 on most shared results.

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