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 Added Most Hyped Most Active Needs Verification Most Verified

Showing 301–325 of 1035 papers

Title	Date	Tasks	Status
Disambiguated skip-gram model	Oct 1, 2018	Image Captioningmodel	—Unverified
xSense: Learning Sense-Separated Sparse Representations and Textual Definitions for Explainable Word Sense Networks	Sep 10, 2018	Word EmbeddingsWord Sense Disambiguation	CodeCode Available
A Novel Neural Sequence Model with Multiple Attentions for Word Sense Disambiguation	Sep 4, 2018	DecoderSentence	—Unverified
WiC: the Word-in-Context Dataset for Evaluating Context-Sensitive Meaning Representations	Aug 28, 2018	Word EmbeddingsWord Sense Disambiguation	—Unverified
KDSL: a Knowledge-Driven Supervised Learning Framework for Word Sense Disambiguation	Aug 28, 2018	Word Sense Disambiguation	—Unverified
Why Self-Attention? A Targeted Evaluation of Neural Machine Translation Architectures	Aug 27, 2018	Machine TranslationTranslation	—Unverified
Learning Graph Embeddings from WordNet-based Similarity Measures	Aug 16, 2018	Graph EmbeddingSemantic Similarity	—Unverified
Synonymy in Bilingual Context: The CzEngClass Lexicon	Aug 1, 2018	Word Sense Disambiguation	—Unverified
Using Word Embeddings for Unsupervised Acronym Disambiguation	Aug 1, 2018	Image CaptioningWord Embeddings	—Unverified
Model-Free Context-Aware Word Composition	Aug 1, 2018	Dimensionality ReductionLearning Word Embeddings	—Unverified
Word Sense Disambiguation Based on Word Similarity Calculation Using Word Vector Representation from a Knowledge-based Graph	Aug 1, 2018	Information RetrievalMachine Translation	CodeCode Available
GenSense: A Generalized Sense Retrofitting Model	Aug 1, 2018	modelSemantic Similarity	CodeCode Available
A New Approach to Animacy Detection	Aug 1, 2018	Coreference ResolutionSemantic Role Labeling	—Unverified
A Deep Dive into Word Sense Disambiguation with LSTM	Aug 1, 2018	Language ModelingLanguage Modelling	—Unverified
A Rule-Based System for Disambiguating French Locative Verbs and Their Translation into Arabic	Aug 1, 2018	FormMachine Translation	—Unverified
WordNet-Based Information Retrieval Using Common Hypernyms and Combined Features	Jul 15, 2018	Information RetrievalRetrieval	—Unverified
Towards Enhancing Lexical Resource and Using Sense-annotations of OntoSenseNet for Sentiment Analysis	Jul 9, 2018	Sentiment AnalysisWord Sense Disambiguation	CodeCode Available
Exploiting Common Characters in Chinese and Japanese to Learn Cross-Lingual Word Embeddings via Matrix Factorization	Jul 1, 2018	Cross-Lingual Word EmbeddingsMachine Translation	—Unverified
EmotionX-SmartDubai\_NLP: Detecting User Emotions In Social Media Text	Jul 1, 2018	Feature EngineeringGeneral Classification	—Unverified
Sense-Aware Neural Models for Pun Location in Texts	Jul 1, 2018	Word Sense Disambiguation	—Unverified
Using pseudo-senses for improving the extraction of synonyms from word embeddings	Jul 1, 2018	Dimensionality ReductionSemantic Similarity	—Unverified
Ranking-Based Automatic Seed Selection and Noise Reduction for Weakly Supervised Relation Extraction	Jul 1, 2018	RelationRelation Extraction	CodeCode Available
Unsupervised Detection of Metaphorical Adjective-Noun Pairs	Jun 1, 2018	ClusteringMachine Translation	—Unverified
SemEval-2018 Task 10: Capturing Discriminative Attributes	Jun 1, 2018	AttributeSemantic Similarity	—Unverified
NLP\_HZ at SemEval-2018 Task 9: a Nearest Neighbor Approach	Jun 1, 2018	Hypernym DiscoveryInformation Retrieval	—Unverified

Show:10 25 50

← PrevPage 13 of 42Next →

All datasets Words in Context Supervised:RUSSE SemEval 2013 Task 12 Senseval-2 SensEval 3 Task 1 SemEval 2007 Task 7 SemEval 2007 Task 17 FEWS WiC-TSV BIG-bench (Anachronisms)Knowledge-based:

Benchmark Results

#	Model	Metric	Claimed	Verified	Status
1	COSINE + Transductive Learning	Accuracy	85.3	—	Unverified
2	PaLM 540B (finetuned)	Accuracy	78.8	—	Unverified
3	ST-MoE-32B 269B (fine-tuned)	Accuracy	77.7	—	Unverified
4	DeBERTa-Ensemble	Accuracy	77.5	—	Unverified
5	Vega v2 6B (fine-tuned)	Accuracy	77.4	—	Unverified
6	UL2 20B (fine-tuned)	Accuracy	77.3	—	Unverified
7	Turing NLR v5 XXL 5.4B (fine-tuned)	Accuracy	77.1	—	Unverified
8	T5-XXL 11B	Accuracy	76.9	—	Unverified
9	DeBERTa-1.5B	Accuracy	76.4	—	Unverified
10	ST-MoE-L 4.1B (fine-tuned)	Accuracy	74	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	SANDWiCH	Senseval 2	87.8	—	Unverified
2	GlossGPT	Senseval 2	86.1	—	Unverified
3	ConSeC+WNGC	Senseval 2	82.7	—	Unverified
4	ESR+WNGC	Senseval 2	82.5	—	Unverified
5	ConSeC	Senseval 2	82.3	—	Unverified
6	ESCHER SemCor	Senseval 2	81.7	—	Unverified
7	ESR	Senseval 2	81.3	—	Unverified
8	EWISER+WNGC	Senseval 2	80.8	—	Unverified
9	SemCor+WNGC, hypernyms	Senseval 2	79.7	—	Unverified
10	SparseLMMS+WNGC	Senseval 2	79.6	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	Human Benchmark	Accuracy	0.81	—	Unverified
2	ruT5-large-finetune	Accuracy	0.74	—	Unverified
3	RuBERT conversational	Accuracy	0.73	—	Unverified
4	RuBERT plain	Accuracy	0.73	—	Unverified
5	ruRoberta-large finetune	Accuracy	0.72	—	Unverified
6	ruBert-base finetune	Accuracy	0.71	—	Unverified
7	Multilingual Bert	Accuracy	0.69	—	Unverified
8	ruT5-base-finetune	Accuracy	0.68	—	Unverified
9	ruBert-large finetune	Accuracy	0.68	—	Unverified
10	SBERT_Large_mt_ru_finetuning	Accuracy	0.66	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	SemCor+WNGC, hypernyms	F1	78.7	—	Unverified
2	SemCor+WNGT, vocabulary reduced, ensemble	F1	72.63	—	Unverified
3	LSTMLP (T:SemCor, U:1K)	F1	69.5	—	Unverified
4	LSTMLP (T:OMSTI, U:1K)	F1	68.1	—	Unverified
5	LSTMLP (T:SemCor, U:OMSTI)	F1	67.9	—	Unverified
6	LSTM (T:OMSTI)	F1	67.3	—	Unverified
7	GASext (Concatenation)	F1	67.2	—	Unverified
8	GASext (Linear)	F1	67.1	—	Unverified
9	GAS (Concatenation)	F1	67	—	Unverified
10	LSTM (T:SemCor)	F1	67	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	SemCor+WNGC, hypernyms	F1	79.7	—	Unverified
2	SemCor+WNGT, vocabulary reduced, ensemble	F1	75.15	—	Unverified
3	LSTMLP (T:OMSTI, U:1K)	F1	74.4	—	Unverified
4	LSTMLP (T:SemCor, U:OMSTI)	F1	73.9	—	Unverified
5	LSTMLP (T:SemCor, U:1K)	F1	73.8	—	Unverified
6	LSTM (T:SemCor)	F1	73.6	—	Unverified
7	GASext (Linear)	F1	72.4	—	Unverified
8	LSTM (T:OMSTI)	F1	72.4	—	Unverified
9	GASext (Concatenation)	F1	72.2	—	Unverified
10	GAS (Concatenation)	F1	72.1	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	SemCor+WNGC, hypernyms	F1	77.8	—	Unverified
2	LSTMLP (T:SemCor, U:1K)	F1	71.8	—	Unverified
3	LSTMLP (T:SemCor, U:OMSTI)	F1	71.1	—	Unverified
4	LSTMLP (T:OMSTI, U:1K)	F1	71	—	Unverified
5	GASext (Concatenation)	F1	70.5	—	Unverified
6	GAS (Concatenation)	F1	70.2	—	Unverified
7	SemCor+WNGT, vocabulary reduced, ensemble	F1	70.11	—	Unverified
8	GASext (Linear)	F1	70.1	—	Unverified
9	GAS (Linear)	F1	70	—	Unverified
10	LSTM (T:SemCor)	F1	69.2	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	SemCor+WNGC, hypernyms	F1	90.4	—	Unverified
2	SemCor+WNGT, vocabulary reduced, ensemble	F1	86.02	—	Unverified
3	kNN-BERT + POS (training corpus: WNGT)	F1	85.32	—	Unverified
4	LSTMLP (T:SemCor, U:OMSTI)	F1	84.3	—	Unverified
5	LSTMLP (T:SemCor, U:1K)	F1	83.6	—	Unverified
6	LSTMLP (T:OMSTI, U:1K)	F1	83.3	—	Unverified
7	LSTM (T:SemCor)	F1	82.8	—	Unverified
8	ShotgunWSD 2.0	F1	81.22	—	Unverified
9	kNN-BERT	F1	81.2	—	Unverified
10	LSTM (T:OMSTI)	F1	81.1	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	SemCor+WNGC, hypernyms	F1	73.4	—	Unverified
2	SemCor+WNGT, vocabulary reduced, ensemble	F1	66.81	—	Unverified
3	LSTM (T:SemCor)	F1	64.2	—	Unverified
4	LSTMLP (T:SemCor, U:OMSTI)	F1	63.7	—	Unverified
5	LSTMLP (T:SemCor, U:1K)	F1	63.5	—	Unverified
6	LSTMLP (T:OMSTI, U:1K)	F1	63.3	—	Unverified
7	kNN-BERT + POS (training corpus: SemCor)	F1	63.17	—	Unverified
8	kNN-BERT	F1	60.94	—	Unverified
9	LSTM (T:OMSTI)	F1	60.7	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	GlossGPT	F1 (Zeroshot Dev)	81.8	—	Unverified
2	ESR Large	F1 (Zeroshot Dev)	77.4	—	Unverified
3	ESR base	F1 (Zeroshot Dev)	73.9	—	Unverified
4	SEMEq Large	F1 (Zeroshot Dev)	73.7	—	Unverified
5	SEMeq base	F1 (Zeroshot Dev)	71.5	—	Unverified
6	RTWE large	F1 (Zero shot test)	69.9	—	Unverified
7	Lesk	F1 (Zeroshot Dev)	40.1	—	Unverified
8	MFS	F1 (Zeroshot Dev)	0	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	Human	Task 3 Accuracy: all	85.3	—	Unverified
2	transformers	Task 1 Accuracy: all	77.8	—	Unverified
3	CTLR	Task 1 Accuracy: all	76.8	—	Unverified
4	GlossBert-ws	Task 1 Accuracy: all	75.9	—	Unverified
5	Bert-base	Task 1 Accuracy: all	75.3	—	Unverified
6	Unsupervised Bert	Task 1 Accuracy: all	54.4	—	Unverified
7	FastText	Task 1 Accuracy: all	53.7	—	Unverified
8	All true	Task 1 Accuracy: all	50.8	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	Chinchilla-70B (few-shot, k=5)	Accuracy	69.1	—	Unverified
2	Gopher-280B (few-shot, k=5)	Accuracy	56.4	—	Unverified
3	OPT 175B	Accuracy	49.1	—	Unverified
4	GAL 120B (few-shot, k=5)	Accuracy	48.7	—	Unverified
5	GAL 30B (few-shot, k=5)	Accuracy	47	—	Unverified
6	BLOOM 176B	Accuracy	1.3	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	UKBppr_w2w	Senseval 2	68.8	—	Unverified
2	KEF	All	68	—	Unverified
3	WSD-TM	All	66.9	—	Unverified
4	Babelfy	All	65.5	—	Unverified
5	WN 1st sense baseline	All	65.2	—	Unverified
6	UKBppr_w2w-nf	All	57.5	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	SemCor+WNGC, hypernyms	F1	82.6	—	Unverified
2	SemCor+WNGT, vocabulary reduced, ensemble	F1	74.46	—	Unverified
3	GASext (Concatenation)	F1	72.6	—	Unverified
4	GASext (Linear)	F1	72.1	—	Unverified
5	GAS (Concatenation)	F1	71.8	—	Unverified
6	GAS (Linear)	F1	71.6	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	kNN-BERT	F1	80.12	—	Unverified
2	IMS + adapted CW	F1	73.4	—	Unverified
3	BiLSTM with GloVe	F1	73.4	—	Unverified
4	Single BiLSTM	F1	72.5	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	kNN-BERT	F1	76.52	—	Unverified
2	BiLSTM with GloVe	F1	66.9	—	Unverified
3	IMS + adapted CW	F1	66.2	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	SPIN	Sequence Recovery %(All)	30.3	—	Unverified