Papers › CLIP-KD: An Empirical Study of CLIP Model Distillation

CLIP-KD: An Empirical Study of CLIP Model Distillation

24 Jul 2023CVPR 2024 1arXiv:2307.12732archive 2025-07-28

Chuanguang Yang, Zhulin An, Libo Huang, Junyu Bi, Xinqiang Yu, Han Yang, Boyu Diao, Yongjun Xu

Contrastive Language-Image Pre-training (CLIP) has become a promising language-supervised visual pre-training framework. This paper aims to distill small CLIP models supervised by a large teacher CLIP model. We propose several distillation strategies, including relation, feature, gradient and contrastive paradigms, to examine the effectiveness of CLIP-Knowledge Distillation (KD). We show that a simple feature mimicry with Mean Squared Error loss works surprisingly well. Moreover, interactive contrastive learning across teacher and student encoders is also effective in performance improvement. We explain that the success of CLIP-KD can be attributed to maximizing the feature similarity between teacher and student. The unified method is applied to distill several student models trained on CC3M+12M. CLIP-KD improves student CLIP models consistently over zero-shot ImageNet classification and cross-modal retrieval benchmarks. When using ViT-L/14 pretrained on Laion-400M as the teacher, CLIP-KD achieves 57.5\% and 55.4\% zero-shot top-1 ImageNet accuracy over ViT-B/16 and ResNet-50, surpassing the original CLIP without KD by 20.5\% and 20.1\% margins, respectively. Our code is released on https://github.com/winycg/CLIP-KD.

PaperPDFConference PDFCodeCode Syntology ran

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

For agents, Syntology's MCP tool lists every function and class Syntology harvested from this paper and whether it ran (how to connect): get_harvested_code_for_paper(arxiv_id="2307.12732")

Code

Syntology Ran 8 of 18 code samples harvested from 1 repository linked to this paper; 10 have no recorded run. Of those that ran: 1 ran · violated contract; 4 ran · our draft was wrong; 3 ran with no contract checked.

By repository: official repository: 18 samples from 1 repository, 8 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

winycg/clip-kd officialmentioned in paperpytorch 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

18 samples harvested; 8 ran; 0 honoured the contract we drafted; 10 have no recorded run. Read from Syntology's graph 2026-09-24; that is when this build read the record, not when the samples ran.

1ran · violated contract
4ran · our draft was wrong
3ran
10unverified

Licence: 18 of the 18 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 winycg/clip-kd. “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. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself; “our draft was wrong” and “fixture could not drive it” are failures of Syntology's instrument, not of the code.

Each sample ends with its code_sha256, Syntology's identity for that exact code. An agent fetches the stored sample with Syntology's MCP tool get_code(code_sha256="…") (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 from a repository marked official are listed first. 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 four 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; or the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar). 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.

basic_clean winycg/clip-kd/src/open_clip/tokenizer.py official repository ran · our draft was wrong fingerprinted no licence file found · pointer only · 98f385d847636a3e · report
convert_to_custom_text_state_dict winycg/clip-kd/src/open_clip/model.py official repository ran · violated contract no licence file found · pointer only · f19962ebb134b3d7 · report
get_cast_dtype winycg/clip-kd/src/open_clip/model.py official repository ran · our draft was wrong no licence file found · pointer only · dcd422d66b0581d8 · report
get_grad winycg/clip-kd/src/open_clip/loss.py official repository ran no licence file found · pointer only · ba49abe4c9f896d0 · report
get_pairs winycg/clip-kd/src/open_clip/tokenizer.py official repository ran · our draft was wrong no licence file found · pointer only · d919ae32e5e4e616 · report
image_transform winycg/clip-kd/src/open_clip/transform.py official repository ran no licence file found · pointer only · 181856ae1295b257 · report
load_state_dict winycg/clip-kd/src/open_clip/factory.py official repository ran no licence file found · pointer only · 4b3a4d17e1df3140 · report
whitespace_clean winycg/clip-kd/src/open_clip/tokenizer.py official repository ran · our draft was wrong fingerprinted no licence file found · pointer only · 9542161e9640b858 · report
build_model_from_openai_state_dict winycg/clip-kd/src/open_clip/model.py official repository unverified no licence file found · pointer only · 1ae9f5eab925c792 · report
gather_features winycg/clip-kd/src/open_clip/loss.py official repository unverified no licence file found · pointer only · ddcbd45e940484ee · report
get_model_config winycg/clip-kd/src/open_clip/factory.py official repository unverified no licence file found · pointer only · 5c7cf820e99443ff · report
get_tokenizer winycg/clip-kd/src/open_clip/factory.py official repository unverified no licence file found · pointer only · 9b7c99fcbb58f26f · report
list_pretrained winycg/clip-kd/src/open_clip/pretrained.py official repository unverified no licence file found · pointer only · 70a24b5107b6b553 · report
list_pretrained_models_by_tag winycg/clip-kd/src/open_clip/pretrained.py official repository unverified no licence file found · pointer only · b206733fa0764571 · report
list_pretrained_tags_by_model winycg/clip-kd/src/open_clip/pretrained.py official repository unverified no licence file found · pointer only · 1e11392669acc284 · report
load_openai_model winycg/clip-kd/src/open_clip/openai.py official repository unverified no licence file found · pointer only · a1693b09a14a516f · report
register_pooler winycg/clip-kd/src/open_clip/hf_model.py official repository unverified no licence file found · pointer only · 2a377da4a76a2d44 · report
sim_stats winycg/clip-kd/src/open_clip/loss.py official repository unverified no licence file found · pointer only · 486bb91e77fa750a · report

Tasks

Contrastive LearningCross-Modal RetrievalKnowledge DistillationRetrieval

1 archive task tag without a task page not shown.

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

CLIPContrastive Learning

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