Papers › RuCoLA: Russian Corpus of Linguistic Acceptability

RuCoLA: Russian Corpus of Linguistic Acceptability

23 Oct 2022arXiv:2210.12814archive 2025-07-28

Vladislav Mikhailov, Tatiana Shamardina, Max Ryabinin, Alena Pestova, Ivan Smurov, Ekaterina Artemova

Linguistic acceptability (LA) attracts the attention of the research community due to its many uses, such as testing the grammatical knowledge of language models and filtering implausible texts with acceptability classifiers. However, the application scope of LA in languages other than English is limited due to the lack of high-quality resources. To this end, we introduce the Russian Corpus of Linguistic Acceptability (RuCoLA), built from the ground up under the well-established binary LA approach. RuCoLA consists of $9.8$k in-domain sentences from linguistic publications and $3.6$k out-of-domain sentences produced by generative models. The out-of-domain set is created to facilitate the practical use of acceptability for improving language generation. Our paper describes the data collection protocol and presents a fine-grained analysis of acceptability classification experiments with a range of baseline approaches. In particular, we demonstrate that the most widely used language models still fall behind humans by a large margin, especially when detecting morphological and semantic errors. We release RuCoLA, the code of experiments, and a public leaderboard (rucola-benchmark.com) to assess the linguistic competence of language models for Russian.

PaperPDFCode

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

Code

russiannlp/rucola officialmentioned in papermentioned on GitHubpytorchApache-2.0 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

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Linguistic AcceptabilityText Generation

Datasets

Introduced by this paper, per the archive.

RuCoLA

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Linguistic Acceptability CoLA RemBERT MCC 0.6 #43 of 43 Archive leaderboard report
Linguistic Acceptability ItaCoLA XLM-R MCC 0.52 #2 of 4 Archive leaderboard report
Linguistic Acceptability ItaCoLA mBERT MCC 0.36 #4 of 4 Archive leaderboard report
Linguistic Acceptability RuCoLA ruRoBERTa Accuracy 79.34 #2 of 9 Archive leaderboard report
Linguistic Acceptability RuCoLA ruRoBERTa MCC 0.53 #2 of 9 Archive leaderboard report
Linguistic Acceptability RuCoLA RemBERT Accuracy 75.06 #4 of 9 Archive leaderboard report
Linguistic Acceptability RuCoLA RemBERT MCC 0.44 #4 of 9 Archive leaderboard report
Linguistic Acceptability RuCoLA ruBERT Accuracy 74.3 #5 of 9 Archive leaderboard report
Linguistic Acceptability RuCoLA ruBERT MCC 0.42 #5 of 9 Archive leaderboard report
Linguistic Acceptability RuCoLA ruGPT-3 Accuracy 53.82 #6 of 9 Archive leaderboard report
Linguistic Acceptability RuCoLA ruGPT-3 MCC 0.30 #6 of 9 Archive leaderboard report
Linguistic Acceptability RuCoLA ruT5 Accuracy 68.41 #7 of 9 Archive leaderboard report
Linguistic Acceptability RuCoLA ruT5 MCC 0.25 #7 of 9 Archive leaderboard report
Linguistic Acceptability RuCoLA mBERT MCC 0.15 #8 of 9 Archive leaderboard report
Linguistic Acceptability RuCoLA XLM-R Accuracy 61.13 #9 of 9 Archive leaderboard report
Linguistic Acceptability RuCoLA XLM-R MCC 0.13 #9 of 9 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.

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