SOTAVerified

Word Sense Disambiguation

The task of Word Sense Disambiguation (WSD) consists of associating words in context with their most suitable entry in a pre-defined sense inventory. The de-facto sense inventory for English in WSD is WordNet.. For example, given the word “mouse” and the following sentence:

“A mouse consists of an object held in one's hand, with one or more buttons.”

we would assign “mouse” with its electronic device sense (the 4th sense in the WordNet sense inventory).

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 1–50 of 1035 papers

TitleStatusHype
Training Compute-Optimal Large Language ModelsCode6
Galactica: A Large Language Model for ScienceCode4
N-Grammer: Augmenting Transformers with latent n-gramsCode4
Language Models are Few-Shot LearnersCode3
ST-MoE: Designing Stable and Transferable Sparse Expert ModelsCode3
DeBERTa: Decoding-enhanced BERT with Disentangled AttentionCode2
Knowledge-Design: Pushing the Limit of Protein Design via Knowledge RefinementCode2
LaMini-LM: A Diverse Herd of Distilled Models from Large-Scale InstructionsCode2
The CoT Collection: Improving Zero-shot and Few-shot Learning of Language Models via Chain-of-Thought Fine-TuningCode2
Exploring the Limits of Transfer Learning with a Unified Text-to-Text TransformerCode2
AlexaTM 20B: Few-Shot Learning Using a Large-Scale Multilingual Seq2Seq ModelCode2
Scaling Language Models: Methods, Analysis & Insights from Training GopherCode2
Hungry Hungry Hippos: Towards Language Modeling with State Space ModelsCode2
Fietje: An open, efficient LLM for DutchCode2
Try to Substitute: An Unsupervised Chinese Word Sense Disambiguation Method Based on HowNetCode1
The MuCoW Test Suite at WMT 2019: Automatically Harvested Multilingual Contrastive Word Sense Disambiguation Test Sets for Machine TranslationCode1
UL2: Unifying Language Learning ParadigmsCode1
RussianSuperGLUE: A Russian Language Understanding Evaluation BenchmarkCode1
Multiple Object Tracking Challenge Technical Report for Team MT_IoTCode1
Nibbling at the Hard Core of Word Sense DisambiguationCode1
ViHealthBERT: Pre-trained Language Models for Vietnamese in Health Text MiningCode1
Latin BERT: A Contextual Language Model for Classical PhilologyCode1
Adapting BERT for Word Sense Disambiguation with Gloss Selection Objective and Example SentencesCode1
3AM: An Ambiguity-Aware Multi-Modal Machine Translation DatasetCode1
Non-Parametric Few-Shot Learning for Word Sense DisambiguationCode1
Potential Idiomatic Expression (PIE)-English: Corpus for Classes of IdiomsCode1
SubjQA: A Dataset for Subjectivity and Review ComprehensionCode1
Language Modelling Makes Sense: Propagating Representations through WordNet for Full-Coverage Word Sense DisambiguationCode1
The MUCOW word sense disambiguation test suite at WMT 2020Code1
Learning to Learn to Disambiguate: Meta-Learning for Few-Shot Word Sense DisambiguationCode1
Guess the Instruction! Flipped Learning Makes Language Models Stronger Zero-Shot LearnersCode1
Fine-Tuning Pre-trained Language Model with Weak Supervision: A Contrastive-Regularized Self-Training ApproachCode1
Improved Word Sense Disambiguation with Enhanced Sense RepresentationsCode1
LIAAD at SemDeep-5 Challenge: Word-in-Context (WiC)Code1
Don't Neglect the Obvious: On the Role of Unambiguous Words in Word Sense DisambiguationCode1
ConSeC: Word Sense Disambiguation as Continuous Sense ComprehensionCode1
Conception: Multilingually-Enhanced, Human-Readable Concept Vector RepresentationsCode1
Context-Aware Semantic Similarity Measurement for Unsupervised Word Sense DisambiguationCode1
ESC: Redesigning WSD with Extractive Sense ComprehensionCode1
Can a Fruit Fly Learn Word Embeddings?Code1
Breaking Through the 80\% Glass Ceiling: Raising the State of the Art in Word Sense Disambiguation by Incorporating Knowledge Graph InformationCode1
ChatGPT: Jack of all trades, master of noneCode1
An Evaluation Benchmark for Testing the Word Sense Disambiguation Capabilities of Machine Translation SystemsCode1
An Incremental Parser for Abstract Meaning RepresentationCode1
Analysis and Evaluation of Language Models for Word Sense DisambiguationCode1
Combating the Curse of Multilinguality in Cross-Lingual WSD by Aligning Sparse Contextualized Word RepresentationsCode1
Exploring the Benefits of Training Expert Language Models over Instruction TuningCode1
Moving Down the Long Tail of Word Sense Disambiguation with Gloss-Informed BiencodersCode1
LMMS Reloaded: Transformer-based Sense Embeddings for Disambiguation and BeyondCode1
Addressing Problems across Linguistic Levels in SMT: Combining Approaches to Model Morphology, Syntax and Lexical Choice—0
Show:102550
← PrevPage 1 of 21Next →
All datasetsWords in ContextSupervised:RUSSESemEval 2013 Task 12Senseval-2SensEval 3 Task 1SemEval 2007 Task 7SemEval 2007 Task 17FEWSWiC-TSVBIG-bench (Anachronisms)Knowledge-based:

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1COSINE + Transductive LearningAccuracy85.3—Unverified
2PaLM 540B (finetuned)Accuracy78.8—Unverified
3ST-MoE-32B 269B (fine-tuned)Accuracy77.7—Unverified
4DeBERTa-EnsembleAccuracy77.5—Unverified
5Vega v2 6B (fine-tuned)Accuracy77.4—Unverified
6UL2 20B (fine-tuned)Accuracy77.3—Unverified
7Turing NLR v5 XXL 5.4B (fine-tuned)Accuracy77.1—Unverified
8T5-XXL 11BAccuracy76.9—Unverified
9DeBERTa-1.5BAccuracy76.4—Unverified
10ST-MoE-L 4.1B (fine-tuned)Accuracy74—Unverified
#ModelMetricClaimedVerifiedStatus
1SANDWiCHSenseval 287.8—Unverified
2GlossGPTSenseval 286.1—Unverified
3ConSeC+WNGCSenseval 282.7—Unverified
4ESR+WNGCSenseval 282.5—Unverified
5ConSeCSenseval 282.3—Unverified
6ESCHER SemCorSenseval 281.7—Unverified
7ESRSenseval 281.3—Unverified
8EWISER+WNGCSenseval 280.8—Unverified
9SemCor+WNGC, hypernymsSenseval 279.7—Unverified
10SparseLMMS+WNGCSenseval 279.6—Unverified
#ModelMetricClaimedVerifiedStatus
1Human BenchmarkAccuracy0.81—Unverified
2ruT5-large-finetuneAccuracy0.74—Unverified
3RuBERT conversationalAccuracy0.73—Unverified
4RuBERT plainAccuracy0.73—Unverified
5ruRoberta-large finetuneAccuracy0.72—Unverified
6ruBert-base finetuneAccuracy0.71—Unverified
7Multilingual BertAccuracy0.69—Unverified
8ruT5-base-finetuneAccuracy0.68—Unverified
9ruBert-large finetuneAccuracy0.68—Unverified
10SBERT_Large_mt_ru_finetuningAccuracy0.66—Unverified
#ModelMetricClaimedVerifiedStatus
1SemCor+WNGC, hypernymsF178.7—Unverified
2SemCor+WNGT, vocabulary reduced, ensembleF172.63—Unverified
3LSTMLP (T:SemCor, U:1K)F169.5—Unverified
4LSTMLP (T:OMSTI, U:1K)F168.1—Unverified
5LSTMLP (T:SemCor, U:OMSTI)F167.9—Unverified
6LSTM (T:OMSTI)F167.3—Unverified
7GASext (Concatenation)F167.2—Unverified
8GASext (Linear)F167.1—Unverified
9GAS (Concatenation)F167—Unverified
10LSTM (T:SemCor)F167—Unverified
#ModelMetricClaimedVerifiedStatus
1SemCor+WNGC, hypernymsF179.7—Unverified
2SemCor+WNGT, vocabulary reduced, ensembleF175.15—Unverified
3LSTMLP (T:OMSTI, U:1K)F174.4—Unverified
4LSTMLP (T:SemCor, U:OMSTI)F173.9—Unverified
5LSTMLP (T:SemCor, U:1K)F173.8—Unverified
6LSTM (T:SemCor)F173.6—Unverified
7GASext (Linear)F172.4—Unverified
8LSTM (T:OMSTI)F172.4—Unverified
9GASext (Concatenation)F172.2—Unverified
10GAS (Concatenation)F172.1—Unverified
#ModelMetricClaimedVerifiedStatus
1SemCor+WNGC, hypernymsF177.8—Unverified
2LSTMLP (T:SemCor, U:1K)F171.8—Unverified
3LSTMLP (T:SemCor, U:OMSTI)F171.1—Unverified
4LSTMLP (T:OMSTI, U:1K)F171—Unverified
5GASext (Concatenation)F170.5—Unverified
6GAS (Concatenation)F170.2—Unverified
7SemCor+WNGT, vocabulary reduced, ensembleF170.11—Unverified
8GASext (Linear)F170.1—Unverified
9GAS (Linear)F170—Unverified
10LSTM (T:SemCor)F169.2—Unverified
#ModelMetricClaimedVerifiedStatus
1SemCor+WNGC, hypernymsF190.4—Unverified
2SemCor+WNGT, vocabulary reduced, ensembleF186.02—Unverified
3kNN-BERT + POS (training corpus: WNGT)F185.32—Unverified
4LSTMLP (T:SemCor, U:OMSTI)F184.3—Unverified
5LSTMLP (T:SemCor, U:1K)F183.6—Unverified
6LSTMLP (T:OMSTI, U:1K)F183.3—Unverified
7LSTM (T:SemCor)F182.8—Unverified
8ShotgunWSD 2.0F181.22—Unverified
9kNN-BERTF181.2—Unverified
10LSTM (T:OMSTI)F181.1—Unverified
#ModelMetricClaimedVerifiedStatus
1SemCor+WNGC, hypernymsF173.4—Unverified
2SemCor+WNGT, vocabulary reduced, ensembleF166.81—Unverified
3LSTM (T:SemCor)F164.2—Unverified
4LSTMLP (T:SemCor, U:OMSTI)F163.7—Unverified
5LSTMLP (T:SemCor, U:1K)F163.5—Unverified
6LSTMLP (T:OMSTI, U:1K)F163.3—Unverified
7kNN-BERT + POS (training corpus: SemCor)F163.17—Unverified
8kNN-BERTF160.94—Unverified
9LSTM (T:OMSTI)F160.7—Unverified
#ModelMetricClaimedVerifiedStatus
1GlossGPTF1 (Zeroshot Dev)81.8—Unverified
2ESR LargeF1 (Zeroshot Dev)77.4—Unverified
3ESR baseF1 (Zeroshot Dev)73.9—Unverified
4SEMEq LargeF1 (Zeroshot Dev)73.7—Unverified
5SEMeq baseF1 (Zeroshot Dev)71.5—Unverified
6RTWE largeF1 (Zero shot test)69.9—Unverified
7LeskF1 (Zeroshot Dev)40.1—Unverified
8MFSF1 (Zeroshot Dev)0—Unverified
#ModelMetricClaimedVerifiedStatus
1HumanTask 3 Accuracy: all85.3—Unverified
2transformersTask 1 Accuracy: all77.8—Unverified
3CTLRTask 1 Accuracy: all76.8—Unverified
4GlossBert-wsTask 1 Accuracy: all75.9—Unverified
5Bert-baseTask 1 Accuracy: all75.3—Unverified
6Unsupervised BertTask 1 Accuracy: all54.4—Unverified
7FastTextTask 1 Accuracy: all53.7—Unverified
8All trueTask 1 Accuracy: all50.8—Unverified
#ModelMetricClaimedVerifiedStatus
1Chinchilla-70B (few-shot, k=5)Accuracy69.1—Unverified
2Gopher-280B (few-shot, k=5)Accuracy56.4—Unverified
3OPT 175BAccuracy49.1—Unverified
4GAL 120B (few-shot, k=5)Accuracy48.7—Unverified
5GAL 30B (few-shot, k=5)Accuracy47—Unverified
6BLOOM 176BAccuracy1.3—Unverified
#ModelMetricClaimedVerifiedStatus
1UKBppr_w2wSenseval 268.8—Unverified
2KEFAll68—Unverified
3WSD-TMAll66.9—Unverified
4BabelfyAll65.5—Unverified
5WN 1st sense baselineAll65.2—Unverified
6UKBppr_w2w-nfAll57.5—Unverified
#ModelMetricClaimedVerifiedStatus
1SemCor+WNGC, hypernymsF182.6—Unverified
2SemCor+WNGT, vocabulary reduced, ensembleF174.46—Unverified
3GASext (Concatenation)F172.6—Unverified
4GASext (Linear)F172.1—Unverified
5GAS (Concatenation)F171.8—Unverified
6GAS (Linear)F171.6—Unverified
#ModelMetricClaimedVerifiedStatus
1kNN-BERTF180.12—Unverified
2IMS + adapted CWF173.4—Unverified
3BiLSTM with GloVeF173.4—Unverified
4Single BiLSTMF172.5—Unverified
#ModelMetricClaimedVerifiedStatus
1kNN-BERTF176.52—Unverified
2BiLSTM with GloVeF166.9—Unverified
3IMS + adapted CWF166.2—Unverified
#ModelMetricClaimedVerifiedStatus
1SPINSequence Recovery %(All)30.3—Unverified