Papers › Label Propagation for Zero-shot Classification with Vision-Language Models

Label Propagation for Zero-shot Classification with Vision-Language Models

5 Apr 2024CVPR 2024 1arXiv:2404.04072archive 2025-07-28

Vladan Stojnić, Yannis Kalantidis, Giorgos Tolias

Vision-Language Models (VLMs) have demonstrated impressive performance on zero-shot classification, i.e. classification when provided merely with a list of class names. In this paper, we tackle the case of zero-shot classification in the presence of unlabeled data. We leverage the graph structure of the unlabeled data and introduce ZLaP, a method based on label propagation (LP) that utilizes geodesic distances for classification. We tailor LP to graphs containing both text and image features and further propose an efficient method for performing inductive inference based on a dual solution and a sparsification step. We perform extensive experiments to evaluate the effectiveness of our method on 14 common datasets and show that ZLaP outperforms the latest related works. Code: https://github.com/vladan-stojnic/ZLaP

PaperPDFConference PDFCodeCode 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 2 code samples harvested from 1 repository linked to this paper: 1 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; 0 where Syntology's instrument failed) · 1 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code; here there are none.

By repository: official repository: 2 samples from 1 repository, 1 ran (1 with no instrument failure; 0 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 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

vladan-stojnic/zlap officialmentioned in papermentioned on GitHubpytorchMITSyntology: 2 samples, 1 ran (0 instrument-failed) · 1 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

2 samples harvested: 1 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; 0 where Syntology's instrument failed) · 1 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
1unverified

Licence: 0 of the 2 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 vladan-stojnic/zlap. “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.

vladan-stojnic/zlap official repository 2 samples, 1 ran (0 instrument-failed) · 1 unverified
combine_separate_knns zlap.py ran MIT (permissive) · 6e6c1f2cdab9ec78 · report fixture: signature knn_im2im:ndarray/2/int, sim_im2im:ndarray/2/float, knn_im2text:ndarray/2/int, sim_im2text:ndarray/2/float, num_classes:int/0/int · returned tuple · level recorded 2026-09-24
search_faiss utils.py unverified MIT (permissive) · 7075d41a55223ea5 · report

Tasks

ClassificationImage ClassificationMulti-label zero-shot learningTransductive LearningTransductive Zero-Shot ClassificationZero-Shot Learning

1 archive task tag without a task page not shown.

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Transductive Zero-Shot Classification CIFAR-10 ZLaP Accuracy 93.6 #1 of 1 Archive leaderboard report
Transductive Zero-Shot Classification CIFAR-100 ZLaP Accuarcy 73.3 #1 of 1 Archive leaderboard report
Transductive Zero-Shot Classification CUB-200-2011 ZLaP Accuracy 64.1 #1 of 1 Archive leaderboard report
Transductive Zero-Shot Classification Caltech-101 ZLaP Accuracy 83.7 #1 of 1 Archive leaderboard report
Transductive Zero-Shot Classification DTD ZLaP Accuracy 51.8 #1 of 1 Archive leaderboard report
Transductive Zero-Shot Classification EuroSAT ZLaP Accuracy 60.9 #2 of 2 Archive leaderboard report
Transductive Zero-Shot Classification FGVC-Aircraft ZLaP Accuracy 28.4 #1 of 1 Archive leaderboard report
Transductive Zero-Shot Classification Flowers-102 ZLaP Accuracy 73.4 #1 of 1 Archive leaderboard report
Transductive Zero-Shot Classification Food-101 ZLaP Accuracy 87.9 #1 of 1 Archive leaderboard report
Transductive Zero-Shot Classification ImageNet ZLaP Top 1 Accuracy 72.7 #1 of 1 Archive leaderboard report
Transductive Zero-Shot Classification Oxford-IIIT Pets ZLaP Accuracy 92.8 #1 of 1 Archive leaderboard report
Transductive Zero-Shot Classification SUN397 ZLaP Accuracy 71.9 #1 of 1 Archive leaderboard report
Transductive Zero-Shot Classification Stanford Cars ZLaP Accuracy 72.1 #1 of 1 Archive leaderboard report
Transductive Zero-Shot Classification UCF101 ZLaP Accuracy 77.7 #1 of 1 Archive leaderboard report
Zero-Shot Learning CIFAR-10 ZLaP* Accuracy 93.6 #1 of 2 Archive leaderboard report
Zero-Shot Learning CIFAR-10 ZLaP Accuracy 93.4 #2 of 2 Archive leaderboard report
Zero-Shot Learning CIFAR-100 ZLaP* Accuracy 74.2 #1 of 2 Archive leaderboard report
Zero-Shot Learning CIFAR-100 ZLaP Accuracy 74 #2 of 2 Archive leaderboard report
Zero-Shot Learning CUB-200-2011 ZLaP Accuracy 64.3 #13 of 14 Archive leaderboard report
Zero-Shot Learning CUB-200-2011 ZLaP* Accuracy 64.2 #14 of 14 Archive leaderboard report
Zero-Shot Learning Caltech-101 ZLaP Accuracy 84 #1 of 2 Archive leaderboard report
Zero-Shot Learning Caltech-101 ZLaP* Accuracy 83.1 #2 of 2 Archive leaderboard report
Zero-Shot Learning DTD ZLaP Accuracy 51.2 #1 of 2 Archive leaderboard report
Zero-Shot Learning DTD ZLaP* Accuracy 51 #2 of 2 Archive leaderboard report
Zero-Shot Learning EuroSAT ZLaP* Accuracy 63.2 #1 of 1 Archive leaderboard report
Zero-Shot Learning FGVC-Aircraft ZLaP Accuracy 29.1 #1 of 2 Archive leaderboard report
Zero-Shot Learning FGVC-Aircraft ZLaP* Accuracy 29 #2 of 2 Archive leaderboard report
Zero-Shot Learning Flowers-102 ZLaP Accuracy 75.9 #1 of 2 Archive leaderboard report
Zero-Shot Learning Flowers-102 ZLaP* Accuracy 75.5 #2 of 2 Archive leaderboard report
Zero-Shot Learning Food-101 ZLaP* Accuracy 87.9 #1 of 2 Archive leaderboard report
Zero-Shot Learning Food-101 ZLaP Accuracy 87.8 #2 of 2 Archive leaderboard report
Zero-Shot Learning ImageNet ZLaP Top 1 Accuracy 72.1 #1 of 2 Archive leaderboard report
Zero-Shot Learning ImageNet ZLaP* Top 1 Accuracy 72.1 #2 of 2 Archive leaderboard report
Zero-Shot Learning Oxford-IIIT Pets ZLaP Accuracy 90 #1 of 2 Archive leaderboard report
Zero-Shot Learning Oxford-IIIT Pets ZLaP* Accuracy 89 #2 of 2 Archive leaderboard report
Zero-Shot Learning SUN397 ZLaP* Accuracy 71.4 #1 of 2 Archive leaderboard report
Zero-Shot Learning SUN397 ZLaP Accuracy 71 #2 of 2 Archive leaderboard report
Zero-Shot Learning Stanford Cars ZLaP* Accuracy 71.8 #1 of 2 Archive leaderboard report
Zero-Shot Learning Stanford Cars ZLaP Accuracy 71.2 #2 of 2 Archive leaderboard report
Zero-Shot Learning UCF101 ZLaP Accuracy 76.3 #1 of 2 Archive leaderboard report
Zero-Shot Learning UCF101 ZLaP* Accuracy 76.3 #2 of 2 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

Introduced by this paper: ZLaP, ZLaP*

ZLaPZLaP*

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