Papers › WiC-TSV: An Evaluation Benchmark for Target Sense Verification of Words in Context

WiC-TSV: An Evaluation Benchmark for Target Sense Verification of Words in Context

30 Apr 2020EACL 2021 2arXiv:2004.15016archive 2025-07-28

Anna Breit, Artem Revenko, Kiamehr Rezaee, Mohammad Taher Pilehvar, Jose Camacho-Collados

We present WiC-TSV, a new multi-domain evaluation benchmark for Word Sense Disambiguation. More specifically, we introduce a framework for Target Sense Verification of Words in Context which grounds its uniqueness in the formulation as a binary classification task thus being independent of external sense inventories, and the coverage of various domains. This makes the dataset highly flexible for the evaluation of a diverse set of models and systems in and across domains. WiC-TSV provides three different evaluation settings, depending on the input signals provided to the model. We set baseline performance on the dataset using state-of-the-art language models. Experimental results show that even though these models can perform decently on the task, there remains a gap between machine and human performance, especially in out-of-domain settings. WiC-TSV data is available at https://competitions.codalab.org/competitions/23683

PaperPDFConference PDFCode

Code

semantic-web-company/wic-tsv officialmentioned on GitHubpytorch 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

Binary ClassificationEntity LinkingWord Sense Disambiguation

Datasets

Introduced by this paper, per the archive.

WiC-TSV

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Entity Linking WiC-TSV Bert-base Task 1 Accuracy: all 75.3 #4 of 8 Archive leaderboard report
Entity Linking WiC-TSV Bert-base Task 1 Accuracy: domain specific 77.9 #4 of 8 Archive leaderboard report
Entity Linking WiC-TSV Bert-base Task 1 Accuracy: general purpose 73.3 #4 of 8 Archive leaderboard report
Entity Linking WiC-TSV Bert-base Task 2 Accuracy: all 71.7 #4 of 8 Archive leaderboard report
Entity Linking WiC-TSV Bert-base Task 2 Accuracy: domain specific 74.7 #4 of 8 Archive leaderboard report
Entity Linking WiC-TSV Bert-base Task 2 Accuracy: general purpose 68.6 #4 of 8 Archive leaderboard report
Entity Linking WiC-TSV Bert-base Task 3 Accuracy: all 76.6 #4 of 8 Archive leaderboard report
Entity Linking WiC-TSV Bert-base Task 3 Accuracy: domain specific 80.4 #4 of 8 Archive leaderboard report
Entity Linking WiC-TSV Bert-base Task 3 Accuracy: general purpose 73.5 #4 of 8 Archive leaderboard report
Entity Linking WiC-TSV Unsupervised Bert Task 1 Accuracy: all 54.4 #5 of 8 Archive leaderboard report
Entity Linking WiC-TSV Unsupervised Bert Task 1 Accuracy: domain specific 60.6 #5 of 8 Archive leaderboard report
Entity Linking WiC-TSV Unsupervised Bert Task 1 Accuracy: general purpose 49.2 #5 of 8 Archive leaderboard report
Entity Linking WiC-TSV Unsupervised Bert Task 2 Accuracy: all 62.8 #5 of 8 Archive leaderboard report
Entity Linking WiC-TSV Unsupervised Bert Task 2 Accuracy: domain specific 69.1 #5 of 8 Archive leaderboard report
Entity Linking WiC-TSV Unsupervised Bert Task 2 Accuracy: general purpose 57.6 #5 of 8 Archive leaderboard report
Entity Linking WiC-TSV Unsupervised Bert Task 3 Accuracy: all 60.5 #5 of 8 Archive leaderboard report
Entity Linking WiC-TSV Unsupervised Bert Task 3 Accuracy: domain specific 67.9 #5 of 8 Archive leaderboard report
Entity Linking WiC-TSV Unsupervised Bert Task 3 Accuracy: general purpose 54.4 #5 of 8 Archive leaderboard report
Entity Linking WiC-TSV FastText Task 1 Accuracy: all 53.7 #6 of 8 Archive leaderboard report
Entity Linking WiC-TSV FastText Task 1 Accuracy: domain specific 50.6 #6 of 8 Archive leaderboard report
Entity Linking WiC-TSV FastText Task 1 Accuracy: general purpose 56.2 #6 of 8 Archive leaderboard report
Entity Linking WiC-TSV FastText Task 2 Accuracy: all 52.7 #6 of 8 Archive leaderboard report
Entity Linking WiC-TSV FastText Task 2 Accuracy: domain specific 47.7 #6 of 8 Archive leaderboard report
Entity Linking WiC-TSV FastText Task 2 Accuracy: general purpose 56.8 #6 of 8 Archive leaderboard report
Entity Linking WiC-TSV FastText Task 3 Accuracy: all 53.4 #6 of 8 Archive leaderboard report
Entity Linking WiC-TSV FastText Task 3 Accuracy: domain specific 49.0 #6 of 8 Archive leaderboard report
Entity Linking WiC-TSV FastText Task 3 Accuracy: general purpose 57.1 #6 of 8 Archive leaderboard report
Entity Linking WiC-TSV All true Task 1 Accuracy: all 50.8 #7 of 8 Archive leaderboard report
Entity Linking WiC-TSV All true Task 1 Accuracy: domain specific 47.0 #7 of 8 Archive leaderboard report
Entity Linking WiC-TSV All true Task 1 Accuracy: general purpose 53.8 #7 of 8 Archive leaderboard report
Entity Linking WiC-TSV All true Task 2 Accuracy: all 50.8 #7 of 8 Archive leaderboard report
Entity Linking WiC-TSV All true Task 2 Accuracy: domain specific 47.0 #7 of 8 Archive leaderboard report
Entity Linking WiC-TSV All true Task 2 Accuracy: general purpose 53.8 #7 of 8 Archive leaderboard report
Entity Linking WiC-TSV All true Task 3 Accuracy: all 50.8 #7 of 8 Archive leaderboard report
Entity Linking WiC-TSV All true Task 3 Accuracy: domain specific 47.0 #7 of 8 Archive leaderboard report
Entity Linking WiC-TSV All true Task 3 Accuracy: general purpose 53.8 #7 of 8 Archive leaderboard report
Entity Linking WiC-TSV Human Task 3 Accuracy: all 85.3 #8 of 8 Archive leaderboard report
Entity Linking WiC-TSV Human Task 3 Accuracy: domain specific 89.2 #8 of 8 Archive leaderboard report
Entity Linking WiC-TSV Human Task 3 Accuracy: general purpose 82.1 #8 of 8 Archive leaderboard report
Word Sense Disambiguation WiC-TSV Bert-base Task 1 Accuracy: all 75.3 #4 of 8 Archive leaderboard report
Word Sense Disambiguation WiC-TSV Bert-base Task 1 Accuracy: domain specific 77.9 #4 of 8 Archive leaderboard report
Word Sense Disambiguation WiC-TSV Bert-base Task 1 Accuracy: general purpose 73.3 #4 of 8 Archive leaderboard report
Word Sense Disambiguation WiC-TSV Bert-base Task 2 Accuracy: all 71.7 #4 of 8 Archive leaderboard report
Word Sense Disambiguation WiC-TSV Bert-base Task 2 Accuracy: domain specific 74.7 #4 of 8 Archive leaderboard report
Word Sense Disambiguation WiC-TSV Bert-base Task 2 Accuracy: general purpose 68.6 #4 of 8 Archive leaderboard report
Word Sense Disambiguation WiC-TSV Bert-base Task 3 Accuracy: all 76.6 #4 of 8 Archive leaderboard report
Word Sense Disambiguation WiC-TSV Bert-base Task 3 Accuracy: domain specific 80.4 #4 of 8 Archive leaderboard report
Word Sense Disambiguation WiC-TSV Bert-base Task 3 Accuracy: general purpose 73.5 #4 of 8 Archive leaderboard report
Word Sense Disambiguation WiC-TSV Unsupervised Bert Task 1 Accuracy: all 54.4 #5 of 8 Archive leaderboard report
Word Sense Disambiguation WiC-TSV Unsupervised Bert Task 1 Accuracy: domain specific 60.6 #5 of 8 Archive leaderboard report
Word Sense Disambiguation WiC-TSV Unsupervised Bert Task 1 Accuracy: general purpose 49.2 #5 of 8 Archive leaderboard report
Word Sense Disambiguation WiC-TSV Unsupervised Bert Task 2 Accuracy: all 62.8 #5 of 8 Archive leaderboard report
Word Sense Disambiguation WiC-TSV Unsupervised Bert Task 2 Accuracy: domain specific 69.1 #5 of 8 Archive leaderboard report
Word Sense Disambiguation WiC-TSV Unsupervised Bert Task 2 Accuracy: general purpose 57.6 #5 of 8 Archive leaderboard report
Word Sense Disambiguation WiC-TSV Unsupervised Bert Task 3 Accuracy: all 60.5 #5 of 8 Archive leaderboard report
Word Sense Disambiguation WiC-TSV Unsupervised Bert Task 3 Accuracy: domain specific 67.9 #5 of 8 Archive leaderboard report
Word Sense Disambiguation WiC-TSV Unsupervised Bert Task 3 Accuracy: general purpose 54.4 #5 of 8 Archive leaderboard report
Word Sense Disambiguation WiC-TSV FastText Task 1 Accuracy: all 53.7 #6 of 8 Archive leaderboard report
Word Sense Disambiguation WiC-TSV FastText Task 1 Accuracy: domain specific 50.6 #6 of 8 Archive leaderboard report
Word Sense Disambiguation WiC-TSV FastText Task 1 Accuracy: general purpose 56.2 #6 of 8 Archive leaderboard report
Word Sense Disambiguation WiC-TSV FastText Task 2 Accuracy: all 52.7 #6 of 8 Archive leaderboard report
Word Sense Disambiguation WiC-TSV FastText Task 2 Accuracy: domain specific 47.7 #6 of 8 Archive leaderboard report
Word Sense Disambiguation WiC-TSV FastText Task 2 Accuracy: general purpose 56.8 #6 of 8 Archive leaderboard report
Word Sense Disambiguation WiC-TSV FastText Task 3 Accuracy: all 53.4 #6 of 8 Archive leaderboard report
Word Sense Disambiguation WiC-TSV FastText Task 3 Accuracy: domain specific 49.0 #6 of 8 Archive leaderboard report
Word Sense Disambiguation WiC-TSV FastText Task 3 Accuracy: general purpose 57.1 #6 of 8 Archive leaderboard report
Word Sense Disambiguation WiC-TSV All true Task 1 Accuracy: all 50.8 #7 of 8 Archive leaderboard report
Word Sense Disambiguation WiC-TSV All true Task 1 Accuracy: domain specific 47.0 #7 of 8 Archive leaderboard report
Word Sense Disambiguation WiC-TSV All true Task 1 Accuracy: general purpose 53.8 #7 of 8 Archive leaderboard report
Word Sense Disambiguation WiC-TSV All true Task 2 Accuracy: all 50.8 #7 of 8 Archive leaderboard report
Word Sense Disambiguation WiC-TSV All true Task 2 Accuracy: domain specific 47.0 #7 of 8 Archive leaderboard report
Word Sense Disambiguation WiC-TSV All true Task 2 Accuracy: general purpose 53.8 #7 of 8 Archive leaderboard report
Word Sense Disambiguation WiC-TSV All true Task 3 Accuracy: all 50.8 #7 of 8 Archive leaderboard report
Word Sense Disambiguation WiC-TSV All true Task 3 Accuracy: domain specific 47.0 #7 of 8 Archive leaderboard report
Word Sense Disambiguation WiC-TSV All true Task 3 Accuracy: general purpose 53.8 #7 of 8 Archive leaderboard report
Word Sense Disambiguation WiC-TSV Human Task 3 Accuracy: all 85.3 #8 of 8 Archive leaderboard report
Word Sense Disambiguation WiC-TSV Human Task 3 Accuracy: domain specific 89.2 #8 of 8 Archive leaderboard report
Word Sense Disambiguation WiC-TSV Human Task 3 Accuracy: general purpose 82.1 #8 of 8 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

AdamAttentionAttention DropoutBERTDense ConnectionsDropoutLayer NormalizationLinear LayerLinear Warmup With Linear DecayMulti-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