Papers › X²-VLM: All-In-One Pre-trained Model For Vision-Language Tasks

X²-VLM: All-In-One Pre-trained Model For Vision-Language Tasks

22 Nov 2022arXiv:2211.12402archive 2025-07-28

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

Vision language pre-training aims to learn alignments between vision and language from a large amount of data. Most existing methods only learn image-text alignments. Some others utilize pre-trained object detectors to leverage vision language alignments at the object level. In this paper, we propose to learn multi-grained vision language alignments by a unified pre-training framework that learns multi-grained aligning and multi-grained localization simultaneously. Based on it, we present X²-VLM, an all-in-one model with a flexible modular architecture, in which we further unify image-text pre-training and video-text pre-training in one model. X²-VLM is able to learn unlimited visual concepts associated with diverse text descriptions. Experiment results show that X²-VLM performs the best on base and large scale for both image-text and video-text tasks, making a good trade-off between performance and model scale. Moreover, we show that the modular design of X²-VLM results in high transferability for it to be utilized in any language or domain. For example, by simply replacing the text encoder with XLM-R, X²-VLM outperforms state-of-the-art multilingual multi-modal pre-trained models without any multilingual pre-training. The code and pre-trained models are available at https://github.com/zengyan-97/X2-VLM.

PaperPDFCodeCode Syntology ran

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, on Syntology's MCP service (how to connect):

Code

Syntology 6 code samples harvested from 2 repositories linked to this paper: 3 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 2 where Syntology's instrument failed) · 3 unverified. Where Syntology's instrument failed: 1 where our contract draft was wrong; 1 where the contract check could not run; failures of Syntology's instrument, not of the code.

By repository: official repository: 5 samples from 1 repository, 2 ran (1 with no instrument failure; 1 where Syntology's instrument failed); community (archive-listed): 1 sample from 1 repository, 1 ran (0 with no instrument failure; 1 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

Levels written on 2026-09-25 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed. 2 of the 3 samples that ran have no recorded date.

zengyan-97/x2-vlm officialmentioned in paperpytorchBSD-3-ClauseSyntology: 5 samples, 2 ran (1 instrument-failed) · 3 unverified report
zengyan-97/x-vlm mentioned on GitHubpytorchBSD-3-ClauseSyntology: 1 sample, 1 ran (1 instrument-failed) · 0 unverified report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

6 samples harvested: 3 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 2 where Syntology's instrument failed) · 3 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page has not changed since , the first build that kept a record date for it; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

1ran · our draft was wrong
1ran; the contract check could not run
1ran
3unverified

Licence: 6 of the 6 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from 2 repositories linked to this paper, official or community; the list below is grouped by repository, each group headed by its name and what it is to the paper. 1 sample whose link to this paper records no repository is listed under one of this paper's own repositories, because identical code was harvested from it for another paper. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of six reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity.

zengyan-97/x2-vlm official repository 5 samples, 2 ran (1 instrument-failed) · 3 unverified
get_acc in 22 papers XGQA.py ran BSD-3-Clause recorded; this copy not marked cleared · pointer only · 10aeef229798e5ff · report fixture: signature results:list/1/dict, test_file:list/1/str · returned float · level recorded 2026-09-25
itm_eval in 27 papers Retrieval.py ran; the contract check could not run (RAISES) · metamorphic tier: deterministic BSD-3-Clause recorded; this copy not marked cleared · pointer only · c7f85279d772ea19 · report fixture: signature scores_i2t:ndarray/2/float, scores_t2i:ndarray/2/float, txt2img:dict/0/?, img2txt:dict/0/? · returned dict · level recorded, date not recorded
reinit_scheduler_properties_mysched Pretrain.py unverified BSD-3-Clause recorded; this copy not marked cleared · pointer only · e3d6fbf640a143ee · report
run_image_iter Pretrain.py unverified BSD-3-Clause recorded; this copy not marked cleared · pointer only · 3365d5bf2acc203f · report
run_region_iter Pretrain.py unverified BSD-3-Clause recorded; this copy not marked cleared · pointer only · ee513e1972c86c60 · report
zengyan-97/x-vlm community (archive-listed) 1 sample, 1 ran (1 instrument-failed) · 0 unverified
build_mlp in 74 papers models/xvlm.py identical code Syntology harvested from this repository for another paper; this paper's link records no repository, path or licence of its own ran · our draft was wrong licence of this copy not recorded · 51e7514fd736cca3 · report fixture: signature input_dim:int/0/int, output_dim:int/0/int · returned Sequential · level recorded, date not recorded

Tasks

AllCross-Modal RetrievalImage CaptioningText to Video RetrievalVideo Question AnsweringVideo RetrievalVisual GroundingVisual Question Answering (VQA)Visual ReasoningXLM-R

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Cross-Modal Retrieval COCO 2014 X2-VLM (large) Image-to-text R@1 84.4 #2 of 36 Archive leaderboard report
Cross-Modal Retrieval COCO 2014 X2-VLM (large) Image-to-text R@10 98.5 #2 of 36 Archive leaderboard report
Cross-Modal Retrieval COCO 2014 X2-VLM (large) Image-to-text R@5 96.5 #2 of 36 Archive leaderboard report
Cross-Modal Retrieval COCO 2014 X2-VLM (large) Text-to-image R@1 67.7 #2 of 36 Archive leaderboard report
Cross-Modal Retrieval COCO 2014 X2-VLM (large) Text-to-image R@10 92.5 #2 of 36 Archive leaderboard report
Cross-Modal Retrieval COCO 2014 X2-VLM (large) Text-to-image R@5 87.5 #2 of 36 Archive leaderboard report
Cross-Modal Retrieval COCO 2014 X2-VLM (base) Image-to-text R@1 83.5 #5 of 36 Archive leaderboard report
Cross-Modal Retrieval COCO 2014 X2-VLM (base) Image-to-text R@10 98.5 #5 of 36 Archive leaderboard report
Cross-Modal Retrieval COCO 2014 X2-VLM (base) Image-to-text R@5 96.3 #5 of 36 Archive leaderboard report
Cross-Modal Retrieval COCO 2014 X2-VLM (base) Text-to-image R@1 66.2 #5 of 36 Archive leaderboard report
Cross-Modal Retrieval COCO 2014 X2-VLM (base) Text-to-image R@10 92.2 #5 of 36 Archive leaderboard report
Cross-Modal Retrieval COCO 2014 X2-VLM (base) Text-to-image R@5 87.1 #5 of 36 Archive leaderboard report
Cross-Modal Retrieval Flickr30k X2-VLM (large) Image-to-text R@1 98.8 #1 of 27 Archive leaderboard report
Cross-Modal Retrieval Flickr30k X2-VLM (large) Image-to-text R@10 100 #1 of 27 Archive leaderboard report
Cross-Modal Retrieval Flickr30k X2-VLM (large) Image-to-text R@5 100 #1 of 27 Archive leaderboard report
Cross-Modal Retrieval Flickr30k X2-VLM (large) Text-to-image R@1 91.8 #1 of 27 Archive leaderboard report
Cross-Modal Retrieval Flickr30k X2-VLM (large) Text-to-image R@10 99.5 #1 of 27 Archive leaderboard report
Cross-Modal Retrieval Flickr30k X2-VLM (large) Text-to-image R@5 98.6 #1 of 27 Archive leaderboard report
Cross-Modal Retrieval Flickr30k X2-VLM (base) Image-to-text R@1 98.5 #2 of 27 Archive leaderboard report
Cross-Modal Retrieval Flickr30k X2-VLM (base) Image-to-text R@10 100 #2 of 27 Archive leaderboard report
Cross-Modal Retrieval Flickr30k X2-VLM (base) Image-to-text R@5 100 #2 of 27 Archive leaderboard report
Cross-Modal Retrieval Flickr30k X2-VLM (base) Text-to-image R@1 90.4 #2 of 27 Archive leaderboard report
Cross-Modal Retrieval Flickr30k X2-VLM (base) Text-to-image R@10 99.3 #2 of 27 Archive leaderboard report
Cross-Modal Retrieval Flickr30k X2-VLM (base) Text-to-image R@5 98.2 #2 of 27 Archive leaderboard report
Video Retrieval MSR-VTT-1kA X2-VLM (large) text-to-video R@1 49.6 #18 of 63 Archive leaderboard report
Video Retrieval MSR-VTT-1kA X2-VLM (large) text-to-video R@10 84.2 #18 of 63 Archive leaderboard report
Video Retrieval MSR-VTT-1kA X2-VLM (large) text-to-video R@5 76.7 #18 of 63 Archive leaderboard report
Video Retrieval MSR-VTT-1kA X2-VLM (base) text-to-video R@1 47.6 #28 of 63 Archive leaderboard report
Video Retrieval MSR-VTT-1kA X2-VLM (base) text-to-video R@10 84.2 #28 of 63 Archive leaderboard report
Video Retrieval MSR-VTT-1kA X2-VLM (base) text-to-video R@5 74.1 #28 of 63 Archive leaderboard report
Visual Grounding RefCOCO+ test B X2-VLM (large) Accuracy (%) 81.8 #3 of 6 Archive leaderboard report
Visual Grounding RefCOCO+ test B X2-VLM (base) Accuracy (%) 78.4 #5 of 6 Archive leaderboard report
Visual Grounding RefCOCO+ testA X2-VLM (large) Accuracy (%) 92.1 #3 of 7 Archive leaderboard report
Visual Grounding RefCOCO+ testA X2-VLM (base) Accuracy (%) 90.3 #5 of 7 Archive leaderboard report
Visual Grounding RefCOCO+ val X2-VLM (large) Accuracy (%) 87.6 #3 of 6 Archive leaderboard report
Visual Grounding RefCOCO+ val X2-VLM (base) Accuracy (%) 85.2 #5 of 6 Archive leaderboard report
Visual Question Answering (VQA) MSRVTT-QA X2-VLM (large) Accuracy 0.455 #15 of 34 Archive leaderboard report
Visual Question Answering (VQA) MSRVTT-QA X2-VLM (base) Accuracy 0.45 #16 of 34 Archive leaderboard report
Visual Question Answering (VQA) MSVD-QA X2-VLM (large) Accuracy 0.546 #17 of 36 Archive leaderboard report
Visual Question Answering (VQA) MSVD-QA X2-VLM (base) Accuracy 0.528 #18 of 36 Archive leaderboard report
Visual Question Answering (VQA) VQA v2 test-dev X2-VLM (large) Accuracy 81.9 #7 of 56 Archive leaderboard report
Visual Question Answering (VQA) VQA v2 test-dev X2-VLM (base) Accuracy 80.4 #11 of 56 Archive leaderboard report
Visual Question Answering (VQA) VQA v2 test-std X2-VLM (large) overall 81.8 #4 of 38 Archive leaderboard report
Visual Question Answering (VQA) VQA v2 test-std X2-VLM (base) overall 80.2 #7 of 38 Archive leaderboard report
Visual Reasoning NLVR2 Dev X2-VLM (large) Accuracy 88.7 #2 of 15 Archive leaderboard report
Visual Reasoning NLVR2 Dev X2-VLM (base) Accuracy 86.2 #4 of 15 Archive leaderboard report
Visual Reasoning NLVR2 Test X2-VLM (large) Accuracy 89.4 #2 of 14 Archive leaderboard report
Visual Reasoning NLVR2 Test X2-VLM (base) Accuracy 87.0 #5 of 14 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

BASEXLM-R

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections