Papers › AltCLIP: Altering the Language Encoder in CLIP for Extended Language Capabilities

AltCLIP: Altering the Language Encoder in CLIP for Extended Language Capabilities

12 Nov 2022arXiv:2211.06679archive 2025-07-28

Zhongzhi Chen, Guang Liu, Bo-Wen Zhang, Fulong Ye, Qinghong Yang, Ledell Wu

In this work, we present a conceptually simple and effective method to train a strong bilingual/multilingual multimodal representation model. Starting from the pre-trained multimodal representation model CLIP released by OpenAI, we altered its text encoder with a pre-trained multilingual text encoder XLM-R, and aligned both languages and image representations by a two-stage training schema consisting of teacher learning and contrastive learning. We validate our method through evaluations of a wide range of tasks. We set new state-of-the-art performances on a bunch of tasks including ImageNet-CN, Flicker30k-CN, COCO-CN and XTD. Further, we obtain very close performances with CLIP on almost all tasks, suggesting that one can simply alter the text encoder in CLIP for extended capabilities such as multilingual understanding. Our models and code are available at https://github.com/FlagAI-Open/FlagAI.

PaperPDFCodeCode Syntology ran

In Syntology View this paper on Syntology: its repositories, every harvested function with whether it ran, its licence and the call to fetch it.

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="2211.06679")

Code

Syntology Ran 10 of 11 code samples harvested from 1 repository linked to this paper; 1 has no recorded run. Of those that ran: 10 ran with no contract checked.

By repository: official repository: 11 samples from 1 repository, 10 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

flagai-open/flagai officialmentioned in papermentioned on GitHubpytorchApache-2.0 report
pwc-1/Paper-8 mindspore 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

11 samples harvested; 10 ran; 0 honoured the contract we drafted; 1 has no recorded run. Read from Syntology's graph 2026-09-25; that is when this build read the record, not when the samples ran.

10ran
1unverified

Licence: 0 of the 11 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 flagai-open/flagai. “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.

accuracy_metric flagai-open/flagai/flagai/metrics.py official repository ran Apache-2.0 (permissive) · e923a6d5aaa37270 · report
change_json_to_cls flagai-open/flagai/flagai/model/base_model.py official repository ran Apache-2.0 (permissive) · 9a58da55dcd68a4b · report
cmd_load_hyperparam flagai-open/flagai/flagai/launch.py official repository ran Apache-2.0 (permissive) · 329c97f25378e517 · report
create_custom_forward flagai-open/flagai/flagai/model/aquila_model.py official repository ran Apache-2.0 (permissive) · fd6777c43a4e9d66 · report
get_args_list flagai-open/flagai/flagai/env_trainer.py official repository ran Apache-2.0 (permissive) · 2774d56117e0ff9f · report
get_args_list flagai-open/flagai/flagai/env_trainer_v1.py official repository ran Apache-2.0 (permissive) · 8977724aa4056e15 · report
save_best flagai-open/flagai/flagai/env_args.py official repository ran Apache-2.0 (permissive) · 5011d61e17e5dcd5 · report
sigmoid flagai-open/flagai/flagai/metrics.py official repository ran Apache-2.0 (permissive) · 98a2b2f174e040a3 · report
str2bool flagai-open/flagai/flagai/env_args.py official repository ran Apache-2.0 (permissive) · 92a5f0d0912c3a91 · report
to_python_float flagai-open/flagai/flagai/fp16/loss_scaler.py official repository ran Apache-2.0 (permissive) · 6b5d363ab4dbf012 · report
should_log_le flagai-open/flagai/flagai/logger.py official repository unverified Apache-2.0 (permissive) · 4e8fb0839eac7d8c · report

Tasks

Contrastive LearningCross-Modal RetrievalImage ClassificationImage RetrievalImage-to-Text RetrievalText-to-Image GenerationXLM-RZero-Shot Cross-Modal RetrievalZero-Shot Image ClassificationZero-Shot Transfer Image ClassificationZero-Shot Transfer Image Classification (CN)Zero-shot Image RetrievalZero-shot Text Retrieval

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Zero-Shot Cross-Modal Retrieval Flickr30k AltCLIP Image-to-text R@1 86 #17 of 22 Archive leaderboard report
Zero-Shot Cross-Modal Retrieval Flickr30k AltCLIP Image-to-text R@10 99.1 #17 of 22 Archive leaderboard report
Zero-Shot Cross-Modal Retrieval Flickr30k AltCLIP Image-to-text R@5 98 #17 of 22 Archive leaderboard report
Zero-Shot Cross-Modal Retrieval Flickr30k AltCLIP Text-to-image R@1 72.5 #17 of 22 Archive leaderboard report
Zero-Shot Cross-Modal Retrieval Flickr30k AltCLIP Text-to-image R@10 95.4 #17 of 22 Archive leaderboard report
Zero-Shot Cross-Modal Retrieval Flickr30k AltCLIP Text-to-image R@5 91.6 #17 of 22 Archive leaderboard report
Zero-Shot Transfer Image Classification CN-ImageNet AltCLIP Accuracy (Private) 59.6 #2 of 2 Archive leaderboard report
Zero-Shot Transfer Image Classification CN-ImageNet V2 AltCLIP Accuracy (Private) 50.9 #1 of 1 Archive leaderboard report
Zero-Shot Transfer Image Classification CN-ImageNet-A AltCLIP Accuracy (Private) 58.5 #1 of 1 Archive leaderboard report
Zero-Shot Transfer Image Classification CN-ImageNet-R AltCLIP Accuracy (Private) 79.9 #1 of 1 Archive leaderboard report
Zero-Shot Transfer Image Classification CN-ImageNet-Sketch AltCLIP Accuracy (Private) 46.5 #1 of 1 Archive leaderboard report
Zero-Shot Transfer Image Classification ImageNet AltCLIP Accuracy (Private) 74.5 #18 of 23 Archive leaderboard report
Zero-Shot Transfer Image Classification ImageNet V2 AltCLIP Accuracy (Private) 68.1 #12 of 13 Archive leaderboard report
Zero-Shot Transfer Image Classification ImageNet-A AltCLIP Accuracy (Private) 69.5 #12 of 13 Archive leaderboard report
Zero-Shot Transfer Image Classification ImageNet-R AltCLIP Accuracy 87.2 #11 of 12 Archive leaderboard report
Zero-Shot Transfer Image Classification ImageNet-Sketch AltCLIP Accuracy (Private) 58.7 #7 of 7 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: AltCLIP, AltDiffusion

AltCLIPAltDiffusionCLIPDiffusionXLM-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