datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
wordnet-definitions-en-2021
Wordnet definitions for English
Dataset by Princeton WordNet and the Open English WordNet team
https://github.com/globalwordnet/english-wordnet
This dataset contains every entry in wordnet that has a definition and an example.
Be aware that the word "null" can be misinterpreted as a null value if loading it in with e.g. pandas
CEFR-Annotated-WordNet
CEFR-Annotated WordNet: LLM-Based Proficiency-Guided Semantic Database for Language Learning
Paper: CEFR-Annotated WordNet: LLM-Based Proficiency-Guided Semantic Database for Language Learning
Authors: Masato Kikuchi, Masatsugu Ono, Toshioki Soga, Tetsu Tanabe, Tadachika Ozono
Overview
CEFR-Annotated WordNet is a comprehensive semantic database that maps WordNet 3.0 senses and SemCor 3.0 instances to CEFR (Common European Framework of Reference for Languages)… See the full description on the dataset page: https://huggingface.co/datasets/star092304/CEFR-Annotated-WordNet.open-english-wordnet-synset-2023
open-english-wordnet-synset-2023
Open English WordNet (2023)
Dataset Details
Dataset Description
Open English WordNet is a lexical network of the English language grouping words into synsets and linking them according to relationships such as hypernymy, antonymy and meronymy. It is intended to be used in natural language processing applications and provides deep lexical information about the English language as a graph.
Dataset Sources
Repository:… See the full description on the dataset page: https://huggingface.co/datasets/jon-tow/open-english-wordnet-synset-2023.WordNetNoun
Dataset Card for WordNetNoun
This dataset is a collection of Multi-hop Inference and Mixed-hop Prediction datasets created from WordNet's subsumption (hypernym) hierarchy of noun entities for training and evaluating hierarchy embedding models.
Multi-hop Inference: This task aims to evaluate the model’s ability in deducing indirect, multi-hop subsumptions from direct, one-hop subsumptions, so as to simulate transitive inference.
Mixed-hop Prediction: This task aims to evaluate the… See the full description on the dataset page: https://huggingface.co/datasets/Hierarchy-Transformers/WordNetNoun.wordnet-lexical-topology
WordNet Lexical Topology Dataset
Dataset Summary
The WordNet Lexical Topology Dataset provides comprehensive n-gram frequency analysis from multiple sources:
NLTK WordNet: Original Princeton WordNet with 117,659 synsets
HF WordNet: Frequency-weighted definitions from 864,894 entries with cardinality data
Unicode: Character names from 143,041 Unicode codepoints
This dataset preserves sequential information crucial for language modeling and text generation, with over 12… See the full description on the dataset page: https://huggingface.co/datasets/AbstractPhil/wordnet-lexical-topology.wordnet-multiple-definitions
9/4/2025 the wordnet baseline prep is complete
You can access the prepared and optimized wordnet definitions here;
https://huggingface.co/datasets/AbstractPhil/wordnet-definitions
As they are refined and curated forms of these definitions.
Warning
This dataset is imperfect and not ready for production use yet.
I advise using the wordnet-definitions repo above.
It requires truncation removal for multiple truncated definitions, bias removal for obvious GPT-nano biases, and… See the full description on the dataset page: https://huggingface.co/datasets/AbstractPhil/wordnet-multiple-definitions.adapted-wordnetword_net_synset_lemma
Dataset Card for "word_net_synset_lemma"
More Information needed
task1508_wordnet_antonyms
Dataset Card for Natural Instructions (https://github.com/allenai/natural-instructions) Task: task1508_wordnet_antonyms
Additional Information
Citation Information
The following paper introduces the corpus in detail. If you use the corpus in published work, please cite it:
@misc{wang2022supernaturalinstructionsgeneralizationdeclarativeinstructions,
title={Super-NaturalInstructions: Generalization via Declarative Instructions on 1600+ NLP Tasks}… See the full description on the dataset page: https://huggingface.co/datasets/Lots-of-LoRAs/task1508_wordnet_antonyms.WordnetCitation:
@article{liu2026lmlexiconimprovingdefinitionmodeling,
title={LM-Lexicon: Improving Definition Modeling via Harmonizing Semantic Experts},
author={Yang Liu and Jiaye Yang and Weikang Li and Jiahui Liang and Yang Li and Lingyong Yan},
year={2026},
eprint={2602.14060},
archivePrefix={arXiv},
primaryClass={cs.CL},
url={https://arxiv.org/abs/2602.14060},
}
WordNetNoun
WordNetNoun (Disambiguated Version)
This is a disambiguated version of the WordNet Noun hierarchy dataset, where entity names are formatted as name: definition to resolve polysemy issues.
Disambiguation
Original format (ambiguous):
child: "bank"
parent: "slope"
New format (disambiguated):
child: "bank: sloping land (especially the slope beside a body of water)"
parent: "slope: an elevated geological formation"
Problem Solved
In the original dataset, the word… See the full description on the dataset page: https://huggingface.co/datasets/Jinrui/WordNetNoun.wordnet_datasetwordnet-definitions-en-2021
Wordnet definitions for English
Dataset by Princeton WordNet and the Open English WordNet team
https://github.com/globalwordnet/english-wordnet
This dataset contains every entry in wordnet that has a definition and an example.
Be aware that the word "null" can be misinterpreted as a null value if loading it in with e.g. pandas
wordnet
WordNet RDF
Dataset Description
Lexical database of semantic relations between words (English WordNet 2024)
Original Source: https://en-word.net/static/english-wordnet-2024.ttl.gz
Dataset Summary
This dataset contains RDF triples from WordNet RDF converted to HuggingFace
dataset format for easy use in machine learning pipelines.
Format: Originally turtle, converted to HuggingFace Dataset
Size: 0.21 GB (extracted)
Entities: ~120K synsets
Triples: ~2M
Original… See the full description on the dataset page: https://huggingface.co/datasets/CleverThis/wordnet.OCW_wordnet
Only Connect Wall (OCW) Dataset
The Only Connect Wall (OCW) dataset contains 618 "Connecting Walls" from the Round 3: Connecting Wall segment of the Only Connect quiz show, collected from 15 seasons' worth of episodes. Each wall contains the ground-truth groups and connections as well as recorded human performance. Please see our paper and GitHub repo for more details about the dataset and its motivations.
Usage
# pip install datasets
from datasets import load_dataset… See the full description on the dataset page: https://huggingface.co/datasets/TaatiTeam/OCW_wordnet.Chinese-Wordnet-SemCor
Chinese Wordnet SemCor
Dataset Summary
This dataset is designed for the task of Word Sense Disambiguation (WSD) for common Chinese words, specifically focusing on words identified as "difficult" (having more than 10 senses) within Chinese Wordnet (CWN) 2.0. It originates from the annotation dataset described in Section 3.1 of the paper "Resolving Regular Polysemy in Named Entities."
The original dataset consisted of 28,836 example sentences where a target "difficult" word… See the full description on the dataset page: https://huggingface.co/datasets/lopentu/Chinese-Wordnet-SemCor.english-words-wordnet
English Words WordNet Dataset
This dataset contains a comprehensive collection of English words paired with their detailed definitions from WordNet. Includes words between 2 and 20 characters in length.
Each line represents a single record containing the following structure:
{
"word": "zwiebacks",
"index_0": {
"pos": "n",
"definition":"slice of sweet raised bread baked again until it is brown and hard and crisp",
"examples": []
}
}
WordNet-TaxoLLaMA
Dataset card for WordNet-TaxoLLaMA
TaxoLLaMA is a model capable of solving Lexical Semantics task with SoTA metrics.
The model was fine-tuned on instructive dataset WordNet-TaxoLLaMA. It consists of hypernym-hyponym pairs sampled from WordNet 3.0. As well, it contains definitions, that were used during training to help model disambiguate senses.
Input Format
The TaxoLLaMA model was trained to use the following format :
<s>[INST] <<SYS>> You are a helpfull assistant. List… See the full description on the dataset page: https://huggingface.co/datasets/VityaVitalich/WordNet-TaxoLLaMA.wordnet_steaming_download_2hierarchical_QA_wordnet_extendedwordnet-definitions
WordNet Multiple Definitions - Columnar Format
Overview
This dataset is an optimized columnar version of WordNet multiple definitions, designed for high-performance queries and rapid extraction.
Each definition was sourced by GPT-5 Nano. I may update this to include additional definitions in the future, but I will not break the format.
The original dataset has a more unabridged and noisy set of data; so I'm definitely going to leave it intact. Noisy training is important… See the full description on the dataset page: https://huggingface.co/datasets/AbstractPhil/wordnet-definitions.WordNet-synthetic-example-sentences
I accidentally deleted the existing data, so I re-upload it.
For the data contained in nltk wordnet, example sentences were generated using gemini-2.5-pro-03-25.
I've corrected some typos along the way (“oberson” -> “oberon”, etc.), so some ids may not match their word.
myriade_noun_aligned_with_wordnet_v2
Dataset Card for "myriade_noun_aligned_with_wordnet_v2"
More Information needed
thai-wordnet-instruction
Thai Wordnet instruction
license: cc-by-4.0
Created by Wannaphong Phatthiyaphaibun
wordnet-test
WordNet RDF
Dataset Description
Lexical database of semantic relations between words (English WordNet 2024)
Original Source: https://en-word.net/static/english-wordnet-2024.ttl.gz
Dataset Summary
This dataset contains RDF triples from WordNet RDF converted to HuggingFace
dataset format for easy use in machine learning pipelines.
Format: Originally turtle, converted to HuggingFace Dataset
Size: 0.21 GB (extracted)
Entities: ~120K synsets
Triples: ~2M
Original… See the full description on the dataset page: https://huggingface.co/datasets/sasukae/wordnet-test.wordnet-test-2
WordNet RDF
Dataset Description
Lexical database of semantic relations between words (English WordNet 2024)
Original Source: https://en-word.net/static/english-wordnet-2024.ttl.gz
Dataset Summary
This dataset contains RDF triples from WordNet RDF converted to HuggingFace
dataset format for easy use in machine learning pipelines.
Format: Originally turtle, converted to HuggingFace Dataset
Size: 0.21 GB (extracted)
Entities: ~120K synsets
Triples: ~2M
Original… See the full description on the dataset page: https://huggingface.co/datasets/sasukae/wordnet-test-2.wordnet-v1
WordNet RDF
Dataset Description
Lexical database of semantic relations between words (English WordNet 2024)
Original Source: https://en-word.net/static/english-wordnet-2024.ttl.gz
Dataset Summary
This dataset contains RDF triples from WordNet RDF converted to
HuggingFace dataset format for easy use in machine learning pipelines.
Format: Originally turtle, converted to HuggingFace
Dataset
Size: 0.21 GB (extracted)
Entities: ~120K synsets
Triples: ~2M
Original… See the full description on the dataset page: https://huggingface.co/datasets/Dabbu19/wordnet-v1.qualc-wordnet-en
QualC WordNet English
Dataset Summary
QualC WordNet English is a clean, structured English lexical dataset derived from the Princeton WordNet lexical database through the NLTK corpus interface.
Each record represents a unique lexical sense (synset) and includes definitions, example sentences, synonyms, antonyms, semantic relationships, and derivationally related forms.
The dataset is intended for Natural Language Processing (NLP), semantic analysis, lexical… See the full description on the dataset page: https://huggingface.co/datasets/nandhakumarms/qualc-wordnet-en.myriade_noun_aligned_with_wordnet_noun_sens
Dataset Card for "myriade_noun_aligned_with_wordnet_noun_sens"
More Information needed
wordnet_987
WordNet RDF
Dataset Description
Lexical database of semantic relations between words (English WordNet 2024)
Original Source: https://en-word.net/static/english-wordnet-2024.ttl.gz
Dataset Summary
This dataset contains RDF triples from WordNet RDF converted to HuggingFace
dataset format for easy use in machine learning pipelines.
Format: Originally turtle, converted to HuggingFace Dataset
Size: 0.21 GB (extracted)
Entities: ~120K synsets
Triples: ~2M
Original… See the full description on the dataset page: https://huggingface.co/datasets/vipinpalhugging/wordnet_987.