Papers › Fast, Effective, and Self-Supervised: Transforming Masked Language Models into...

Fast, Effective, and Self-Supervised: Transforming Masked Language Models into Universal Lexical and Sentence Encoders

16 Apr 2021EMNLP 2021 11arXiv:2104.08027archive 2025-07-28

Fangyu Liu, Ivan Vulić, Anna Korhonen, Nigel Collier

Pretrained Masked Language Models (MLMs) have revolutionised NLP in recent years. However, previous work has indicated that off-the-shelf MLMs are not effective as universal lexical or sentence encoders without further task-specific fine-tuning on NLI, sentence similarity, or paraphrasing tasks using annotated task data. In this work, we demonstrate that it is possible to turn MLMs into effective universal lexical and sentence encoders even without any additional data and without any supervision. We propose an extremely simple, fast and effective contrastive learning technique, termed Mirror-BERT, which converts MLMs (e.g., BERT and RoBERTa) into such encoders in 20-30 seconds without any additional external knowledge. Mirror-BERT relies on fully identical or slightly modified string pairs as positive (i.e., synonymous) fine-tuning examples, and aims to maximise their similarity during identity fine-tuning. We report huge gains over off-the-shelf MLMs with Mirror-BERT in both lexical-level and sentence-level tasks, across different domains and different languages. Notably, in the standard sentence semantic similarity (STS) tasks, our self-supervised Mirror-BERT model even matches the performance of the task-tuned Sentence-BERT models from prior work. Finally, we delve deeper into the inner workings of MLMs, and suggest some evidence on why this simple approach can yield effective universal lexical and sentence encoders.

PaperPDFConference PDFCodeCode Syntology ranMethod as code

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 1 code sample harvested from 1 repository linked to this paper: 0 ran · 1 unverified.

By repository: official repository: 1 sample from 1 repository, 0 ran (0 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.

cambridgeltl/mirror-bert officialmentioned in papermentioned on GitHubpytorchSyntology: 1 sample, 0 ran · 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

1 sample harvested: 0 ran · 1 unverified. 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.

1unverified

Licence: 0 of the 1 sample 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 cambridgeltl/mirror-bert. “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 “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.

cambridgeltl/mirror-bert official repository 1 sample, 0 ran · 1 unverified
MirrorBERT src/mirror_bert.py unverified MIT (permissive) · b50c6b3ff155dbd1 · report

Method as code Syntology

What Syntology's graph records as this paper's claimed method, lifted into code: each assembly names its bricks (functions and classes), each with its own role and its own verification level. An assembly is not a check that the code implements the method correctly or reproduces the paper's results. Levels are each brick's own verification level in Syntology's graph (level 0: no recorded run), listed brick by brick and never combined into one level; “ran” means the brick executed on a synthesized input, not that its output is correct, and the instrument figure counts failures of Syntology's instrument, not of the code. Where a model read the code against the paper's claim, each assembly's label verdict is one model's judgement, kept in this page's JSON twin; it is not calibrated, so no card prints it.

1 assembly on this paper: 1 harvested, 0 generated.

harvested Assembly for Mirror-BERT (graph method: Mirror-BERT). 1 brick; individually on a fixture: 0 ran · 1 unverified. The assembly has not been run end-to-end.

  1. MirrorBERT entry · level 0 · unverified · b50c6b3ff155dbd1

Each brick ends with its code_sha256, Syntology's identity for that exact code; click one to copy the get_code call that returns it. For agents, on Syntology's MCP service: assembly(arxiv_id="2104.08027") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

Contrastive LearningCross-Lingual Semantic Textual SimilarityEntity LinkingSTSSemantic SimilaritySemantic Textual SimilaritySentenceSentence Similarity

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Semantic Textual Similarity SICK Mirror-RoBERTa-base (unsup.) Spearman Correlation 0.706 #16 of 22 Archive leaderboard report
Semantic Textual Similarity SICK Mirror-BERT-base (unsup.) Spearman Correlation 0.703 #17 of 22 Archive leaderboard report
Semantic Textual Similarity STS Benchmark Mirror-RoBERTa-base (unsup.) Spearman Correlation 0.787 #54 of 66 Archive leaderboard report
Semantic Textual Similarity STS Benchmark Mirror-BERT-base (unsup.) Spearman Correlation 0.764 #59 of 66 Archive leaderboard report
Semantic Textual Similarity STS12 Mirror-BERT-base (unsup.) Spearman Correlation 0.674 #17 of 20 Archive leaderboard report
Semantic Textual Similarity STS12 Mirror-RoBERTa-base (unsup.) Spearman Correlation 0.648 #19 of 20 Archive leaderboard report
Semantic Textual Similarity STS13 Mirror-RoBERTa-base (unsup.) Spearman Correlation 0.819 #16 of 22 Archive leaderboard report
Semantic Textual Similarity STS13 Mirror-BERT-base (unsup.) Spearman Correlation 0.796 #19 of 22 Archive leaderboard report
Semantic Textual Similarity STS14 Mirror-RoBERTa-base (unsup.) Spearman Correlation 0.732 #17 of 21 Archive leaderboard report
Semantic Textual Similarity STS14 Mirror-BERT-base (unsup.) Spearman Correlation 0.713 #18 of 21 Archive leaderboard report
Semantic Textual Similarity STS15 Mirror-BERT-base (unsup.) Spearman Correlation 0.814 #16 of 20 Archive leaderboard report
Semantic Textual Similarity STS15 Mirror-RoBERTa-base (unsup.) Spearman Correlation 0.798 #18 of 20 Archive leaderboard report
Semantic Textual Similarity STS16 Mirror-RoBERTa-base (unsup.) Spearman Correlation 0.78 #15 of 20 Archive leaderboard report
Semantic Textual Similarity STS16 Mirror-BERT-base (unsup.) Spearman Correlation 0.743 #19 of 20 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: Mirror-BERT

AdamAttentionAttention DropoutBERTContrastive LearningDense ConnectionsDropoutLayer NormalizationLinear LayerLinear Warmup With Linear DecayMirror-BERTMulti-Head AttentionResidual ConnectionSoftmaxWeight DecayWordPiece

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