Papers › Massively Multilingual Sentence Embeddings for Zero-Shot Cross-Lingual Transfer and Beyond

Massively Multilingual Sentence Embeddings for Zero-Shot Cross-Lingual Transfer and Beyond

26 Dec 2018TACL 2019 3arXiv:1812.10464archive 2025-07-28

Mikel Artetxe, Holger Schwenk

We introduce an architecture to learn joint multilingual sentence representations for 93 languages, belonging to more than 30 different families and written in 28 different scripts. Our system uses a single BiLSTM encoder with a shared BPE vocabulary for all languages, which is coupled with an auxiliary decoder and trained on publicly available parallel corpora. This enables us to learn a classifier on top of the resulting embeddings using English annotated data only, and transfer it to any of the 93 languages without any modification. Our experiments in cross-lingual natural language inference (XNLI dataset), cross-lingual document classification (MLDoc dataset) and parallel corpus mining (BUCC dataset) show the effectiveness of our approach. We also introduce a new test set of aligned sentences in 112 languages, and show that our sentence embeddings obtain strong results in multilingual similarity search even for low-resource languages. Our implementation, the pre-trained encoder and the multilingual test set are available at https://github.com/facebookresearch/LASER

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

No sample from the archive's official repository (facebookresearch/LASER) is linked to this paper; the 10 samples below come from 4 community repositories. Syntology harvested code from facebookresearch/LASER for 2 other papers.

facebookresearch/LASER: Syntology's graph links these samples to Exploiting Sentence Order in Document Alignment (arXiv:2004.14523; that paper's own run record: 7 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 0 honoured, 0 violated, 5 with no contract checked; 2 where Syntology's instrument failed) · 1 unverified) and Better Quality Estimation for Low Resource Corpus Mining (arXiv:2203.08259; that paper's own run record: 0 ran · 1 unverified); nothing checks that they implement this paper's method.

Syntology 10 code samples harvested from 4 repositories linked to this paper: 10 ran (of which 0 constructed an object rather than computing a result; 8 with no instrument failure: 2 honoured, 0 violated, 6 with no contract checked; 2 where Syntology's instrument failed) · 0 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: community (archive-listed): 10 samples from 4 repositories, 10 ran (8 with no instrument failure; 2 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.

No sample is linked to this paper from 3 of its repositories that Syntology harvested for other papers:

Levels written between 2026-09-22 and 2026-09-26 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

13 repositories listed; official and paper-mentioned ones first.

facebookresearch/LASER officialmentioned in papermentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Exploiting Sentence Order in Document Alignment (arXiv:2004.14523; that paper's own run record: 7 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 0 honoured, 0 violated, 5 with no contract checked; 2 where Syntology's instrument failed) · 1 unverified) and Better Quality Estimation for Low Resource Corpus Mining (arXiv:2203.08259; that paper's own run record: 0 ran · 1 unverified); nothing checks that they implement this paper's method. report
LawrenceDuan/myLASER mentioned on GitHubpytorchSyntology: 2 samples, 2 ran (1 instrument-failed) · 0 unverified report
Tony4469/laser-agir mentioned on GitHubpytorchNOASSERTIONSyntology: not harvested report
Unbabel/COMET mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding (arXiv:1810.04805; that paper's own run record: 300 ran (of which 75 constructed an object rather than computing a result; 234 with no instrument failure: 18 honoured, 5 violated, 211 with no contract checked; 66 where Syntology's instrument failed) · 359 unverified) and Unsupervised Cross-lingual Representation Learning at Scale (arXiv:1911.02116; that paper's own run record: 40 ran (of which 13 constructed an object rather than computing a result; 35 with no instrument failure: 4 honoured, 1 violated, 30 with no contract checked; 5 where Syntology's instrument failed) · 19 unverified); nothing checks that they implement this paper's method. report
facebookresearch/vizseq mentioned on GitHubSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to BERTScore: Evaluating Text Generation with BERT (arXiv:1904.09675; that paper's own run record: 44 ran (of which 4 constructed an object rather than computing a result; 10 with no instrument failure: 3 honoured, 2 violated, 5 with no contract checked; 34 where Syntology's instrument failed) · 32 unverified); nothing checks that they implement this paper's method. report
imamathcat/LASER_Dependencies mentioned on GitHubpytorchNOASSERTIONSyntology: not harvested report
jeongukjae/smaller-labse mentioned on GitHubtfApache-2.0Syntology: not harvested report
jiamingkong/infoxlm_paddle mentioned on GitHubpaddleSyntology: not harvested report
kmkwon94/ainize-laser mentioned on GitHubpytorchNOASSERTIONSyntology: not harvested report
prabhakar267/LASER-improved mentioned on GitHubpytorchNOASSERTIONSyntology: not harvested report
raymondhs/fairseq-laser mentioned on GitHubpytorchMITSyntology: 2 samples, 2 ran (0 instrument-failed) · 0 unverified report
transducens/LASERtrain mentioned on GitHubpytorchSyntology: 2 samples, 2 ran (1 instrument-failed) · 0 unverified report
yannvgn/laserembeddings mentioned on GitHubpytorchBSD-3-ClauseSyntology: 4 samples, 4 ran (0 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.

A paper named beside a repository with no sample linked to this paper is shown with that paper's own run record, not this paper's: “ran” means executed on a synthesized input, not that the code is correct, and “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Papers are listed in arXiv-id order, at most three per repository.

Code Syntology ran Syntology

10 samples harvested: 10 ran (of which 0 constructed an object rather than computing a result; 8 with no instrument failure: 2 honoured, 0 violated, 6 with no contract checked; 2 where Syntology's instrument failed) · 0 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.

2ran · honoured contract
1ran · our draft was wrong
1ran; the contract check could not run
6ran

Licence: 4 of the 10 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 4 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. “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.

LawrenceDuan/myLASER community (archive-listed) 2 samples, 2 ran (1 instrument-failed) · 0 unverified
BuccExtract tasks/bucc/bucc.py ran · our draft was wrong licence not identified · pointer only · 5e3abc8abc7e1ba3 · report fixture: signature cand2score:dict/0/?, th:float/0/float, fname:NoneType/0/? · returned list · level recorded 2026-09-22
BuccOptimize tasks/bucc/bucc.py ran · honoured contract licence not identified · pointer only · ff90acbe13627868 · report fixture: signature candidate2score:dict/0/?, gold:set/0/? · returned float · level recorded 2026-09-22
raymondhs/fairseq-laser community (archive-listed) 2 samples, 2 ran (0 instrument-failed) · 0 unverified
add_decoder_language laser/laser_dataset.py ran MIT (permissive) · ccb4f61614cf3799 · report fixture: signature batch:dict/0/?, lang_pair:str/0/str · returned dict · level recorded 2026-09-26
uniform_sampler in 5 papers laser/laser_dataset.py ran MIT (permissive) · 50e8df121d8ca072 · report fixture: signature x:ndarray/1/int · returned int · level recorded 2026-09-26
transducens/LASERtrain community (archive-listed) 2 samples, 2 ran (1 instrument-failed) · 0 unverified
buffered_arange fairseq-modules/multilingual_lstm_laser.py ran · honoured contract no licence file found · pointer only · b635d11e8b643e36 · report fixture: signature max:int/0/int · returned Tensor · level recorded 2026-09-22
convert_padding_direction fairseq-modules/multilingual_lstm_laser.py ran; the contract check could not run (RAISES) · metamorphic tier: invariant fingerprinted no licence file found · pointer only · 6031a95f9f326ac7 · report fixture: signature src_tokens:Tensor/2/int, padding_idx:int/0/int, right_to_left:bool/0/bool, left_to_right:bool/0/bool · returned Tensor · level recorded 2026-09-22
yannvgn/laserembeddings community (archive-listed) 4 samples, 4 ran (0 instrument-failed) · 0 unverified
adapt_bpe_codes laserembeddings/utils.py ran BSD-3-Clause (permissive) · 13a068cfdda041fe · report fixture: signature bpe_codes_f:StringIO/0/? · returned StringIO · level recorded 2026-09-26
buffered_arange laserembeddings/encoder.py ran BSD-3-Clause (permissive) · b2ad3933e6341902 · report fixture: signature max:int/0/int · returned Tensor · level recorded 2026-09-26
convert_padding_direction laserembeddings/encoder.py ran BSD-3-Clause (permissive) · 19403a190ad0b624 · report fixture: signature src_tokens:Tensor/2/int, padding_idx:int/0/int, right_to_left:bool/0/bool, left_to_right:bool/0/bool · returned Tensor · level recorded 2026-09-26
non_win_string laserembeddings/__main__.py ran fingerprinted BSD-3-Clause (permissive) · 9bb263aefc7737cf · report fixture: signature s:str/0/str · returned str · level recorded 2026-09-26

Tasks

Cross-Lingual Bitext MiningCross-Lingual Document ClassificationCross-Lingual Natural Language InferenceCross-Lingual TransferDecoderDocument ClassificationJoint Multilingual Sentence RepresentationsNatural Language InferenceParallel Corpus MiningSentenceSentence EmbeddingsZero-Shot Cross-Lingual Transfer

Datasets

Introduced by this paper, per the archive.

Tatoeba

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Cross-Lingual Bitext Mining BUCC Chinese-to-English Massively Multilingual Sentence Embeddings F1 score 92.27 #1 of 1 Archive leaderboard report
Cross-Lingual Bitext Mining BUCC French-to-English Massively Multilingual Sentence Embeddings F1 score 93.91 #1 of 3 Archive leaderboard report
Cross-Lingual Bitext Mining BUCC German-to-English Massively Multilingual Sentence Embeddings F1 score 96.19 #1 of 3 Archive leaderboard report
Cross-Lingual Bitext Mining BUCC Russian-to-English Massively Multilingual Sentence Embeddings F1 score 93.3 #1 of 1 Archive leaderboard report
Cross-Lingual Document Classification MLDoc Zero-Shot English-to-Chinese Massively Multilingual Sentence Embeddings Accuracy 71.93 #5 of 5 Archive leaderboard report
Cross-Lingual Document Classification MLDoc Zero-Shot English-to-French Massively Multilingual Sentence Embeddings Accuracy 77.95 #3 of 6 Archive leaderboard report
Cross-Lingual Document Classification MLDoc Zero-Shot English-to-German Massively Multilingual Sentence Embeddings Accuracy 84.78% #3 of 5 Archive leaderboard report
Cross-Lingual Document Classification MLDoc Zero-Shot English-to-Italian Massively Multilingual Sentence Embeddings Accuracy 69.43 #2 of 4 Archive leaderboard report
Cross-Lingual Document Classification MLDoc Zero-Shot English-to-Japanese Massively Multilingual Sentence Embeddings Accuracy 60.3 #3 of 3 Archive leaderboard report
Cross-Lingual Document Classification MLDoc Zero-Shot English-to-Russian Massively Multilingual Sentence Embeddings Accuracy 67.78 #3 of 5 Archive leaderboard report
Cross-Lingual Document Classification MLDoc Zero-Shot English-to-Spanish Massively Multilingual Sentence Embeddings Accuracy 77.33 #3 of 6 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

BPEBiLSTMLSTMSigmoid ActivationTanh Activation

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