Papers › Acceptability Judgements via Examining the Topology of Attention Maps

Acceptability Judgements via Examining the Topology of Attention Maps

19 May 2022arXiv:2205.09630archive 2025-07-28

Daniil Cherniavskii, Eduard Tulchinskii, Vladislav Mikhailov, Irina Proskurina, Laida Kushnareva, Ekaterina Artemova, Serguei Barannikov, Irina Piontkovskaya, Dmitri Piontkovski, Evgeny Burnaev

The role of the attention mechanism in encoding linguistic knowledge has received special interest in NLP. However, the ability of the attention heads to judge the grammatical acceptability of a sentence has been underexplored. This paper approaches the paradigm of acceptability judgments with topological data analysis (TDA), showing that the geometric properties of the attention graph can be efficiently exploited for two standard practices in linguistics: binary judgments and linguistic minimal pairs. Topological features enhance the BERT-based acceptability classifier scores by $8$%-$24$% on CoLA in three languages (English, Italian, and Swedish). By revealing the topological discrepancy between attention maps of minimal pairs, we achieve the human-level performance on the BLiMP benchmark, outperforming nine statistical and Transformer LM baselines. At the same time, TDA provides the foundation for analyzing the linguistic functions of attention heads and interpreting the correspondence between the graph features and grammatical phenomena.

PaperPDFCode

Code

danchern97/tda4la officialmentioned in papermentioned on GitHub 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 AcceptabilitySentenceTopological Data Analysis

1 archive task tag without a task page not shown.

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Linguistic Acceptability CoLA En-BERT + TDA + PCA Accuracy 88.6% #1 of 43 Archive leaderboard report
Linguistic Acceptability CoLA En-BERT + TDA Accuracy 82.1% #8 of 43 Archive leaderboard report
Linguistic Acceptability CoLA En-BERT + TDA MCC 0.565 #8 of 43 Archive leaderboard report
Linguistic Acceptability CoLA Dev En-BERT + TDA Accuracy 88.6 #1 of 6 Archive leaderboard report
Linguistic Acceptability CoLA Dev En-BERT + TDA MCC 0.725 #1 of 6 Archive leaderboard report
Linguistic Acceptability CoLA Dev XLM-R (pre-trained) + TDA Accuracy 73 #2 of 6 Archive leaderboard report
Linguistic Acceptability CoLA Dev En-BERT (pre-trained) + TDA MCC 0.420 #6 of 6 Archive leaderboard report
Linguistic Acceptability DaLAJ Sw-BERT + H0M Accuracy 76.9 #1 of 1 Archive leaderboard report
Linguistic Acceptability DaLAJ Sw-BERT + H0M MCC 0.542 #1 of 1 Archive leaderboard report
Linguistic Acceptability ItaCoLA XLM-R + TDA Accuracy 92.8 #1 of 4 Archive leaderboard report
Linguistic Acceptability ItaCoLA XLM-R + TDA MCC 0.683 #1 of 4 Archive leaderboard report
Linguistic Acceptability ItaCoLA It-BERT (pre-trained) + TDA Accuracy 89.2 #3 of 4 Archive leaderboard report
Linguistic Acceptability ItaCoLA It-BERT (pre-trained) + TDA MCC 0.478 #3 of 4 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

Absolute Position EncodingsAdamAttentionBPECOLADense ConnectionsDropoutLabel SmoothingLayer NormalizationLinear LayerMulti-Head AttentionPosition-Wise Feed-Forward LayerResidual ConnectionSoftmaxTransformer

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