datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
poem_sentiment
Dataset Card for Gutenberg Poem Dataset
Dataset Summary
Poem Sentiment is a sentiment dataset of poem verses from Project Gutenberg.
This dataset can be used for tasks such as sentiment classification or style transfer for poems.
Supported Tasks and Leaderboards
[More Information Needed]
Languages
The text in the dataset is in English (en).
Dataset Structure
Data Instances
Example of one instance in the dataset.
{'id': 0… See the full description on the dataset page: https://huggingface.co/datasets/google-research-datasets/poem_sentiment.POEM-v2Poetry-Foundation-PoemsFrom: https://www.kaggle.com/datasets/tgdivy/poetry-foundation-poems
Poetry Foundation Poems Dataset
Overview
This dataset contains a collection of 13.9k poems sourced from the Poetry Foundation website. Each poem entry includes its title, author, and associated tags (if available). The dataset provides a robust resource for exploring poetry, analyzing thematic trends, or creating applications such as poem generators.
Dataset Structure
The dataset consists of the following columns:
1. Title:… See the full description on the dataset page: https://huggingface.co/datasets/suayptalha/Poetry-Foundation-Poems.50k_persian_poem_authorpoemsPersian_poemPoemSentimentClassification
PoemSentimentClassification
An MTEB dataset
Massive Text Embedding Benchmark
Poem Sentiment is a sentiment dataset of poem verses from Project Gutenberg.
Task category
t2c
Domains
Reviews, Written
Reference
https://arxiv.org/abs/2011.02686
How to evaluate on this task
You can evaluate an embedding model on this dataset using the following code:
import mteb
task = mteb.get_tasks(["PoemSentimentClassification"])
evaluator = mteb.MTEB(task)
model =… See the full description on the dataset page: https://huggingface.co/datasets/mteb/PoemSentimentClassification.SerendibLLM-PoemSong-Dataset
SerendibLLM Poem and Song Dataset
Sinhala poem and song lyrics dataset for fine-tuning Sinhala LLMs on creative generation.
Built as part of the Serendib LLM Honours project (UCLan 2025-2026).
19,184 instruction-response entries
Sources: kawmuthu.blogspot.com (poems) + lyrics-lk.com (songs)
6 instruction variants per entry (Sinhala + English prompts)
8 themes: love, nature, sadness, joy, spring, country, religion, general
instruction-poemsPoem dataset to be used with instruction fine tuning
Poem_APCD_text_onlyWe used the APCD dataset cited hereafter for pretraining the model. The dataset has been cleaned and only the main text column was kept:
@Article{Yousef2019LearningMetersArabicEnglish-arxiv,
author = {Yousef, Waleed A. and Ibrahime, Omar M. and Madbouly, Taha M. and Mahmoud,
Moustafa A.},
title = {Learning Meters of Arabic and English Poems With Recurrent Neural Networks: a Step
Forward for Language Understanding and Synthesis},
journal =… See the full description on the dataset page: https://huggingface.co/datasets/Yah216/Poem_APCD_text_only.poemspoem_sentimentArabic_Poem_Comprehensive_Dataset_APCDpersian_poems_stanzas_vocabbasepoemsExtract of data from https://www.poetryfoundation.org/
Randomly shuffled with 10% of data as validation set
task833_poem_sentiment_classification
Dataset Card for Natural Instructions (https://github.com/allenai/natural-instructions) Task: task833_poem_sentiment_classification
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/task833_poem_sentiment_classification.poem-generation-1k
poem-generation-1k
poem-generation-1k is a synthetic dataset of 1,149 Persian poem-request instructions paired with generated poems, spanning 11 classical and modern Persian poetic forms and 35 topic categories. It is the base dataset in the Poem Generation Collection, and forms the source pairs for artindnr/poem-reasoning-1k, which adds reasoning traces on top of these same (question, poem) pairs.
How it was built
Both the poem-request instructions and the poems… See the full description on the dataset page: https://huggingface.co/datasets/artindnr/poem-generation-1k.ukrainian-poems
Dataset Card for "ukrainian-poems"
More Information needed
aya-telugu-poems
Summary
aya-telugu-poems is an open source dataset of instruct-style records generated by webscraping a Telugu poems website. This was created as part of Aya Open Science Initiative from Cohere For AI.
This dataset can be used for any purpose, whether academic or commercial, under the terms of the Apache 2.0 License.
Supported Tasks:
Training LLMs
Synthetic Data Generation
Data Augmentation
Languages: Telugu Version: 1.0
Dataset Overview
aya-telugu-poems is a corpus… See the full description on the dataset page: https://huggingface.co/datasets/SuryaKrishna02/aya-telugu-poems.genius_poems_annotationspoem-instructions-datasetArabic_PoemsArabic Poems
it's a filtered version for (https://huggingface.co/datasets/arbml/ashaar) with only al-diwan data
Overview:
The "Arabic Poems" dataset is a comprehensive collection of Arabic poetry, comprising 8,875 entries. Each entry contains detailed information about individual poems and their respective poets, making it a valuable resource for researchers, developers, and enthusiasts of Arabic literature.
Columns:
Unnamed: 0: An index column.
poem_title: The title of the poem.
poem_meter:… See the full description on the dataset page: https://huggingface.co/datasets/alwalid54321/Arabic_Poems.khayyam-challenge-poems
Khayyam Challenge - Poems
The 20 classical Persian source poems (Hafez, Ferdowsi, and others) used in the
Khayyam Challenge benchmark, split
low / medium / long by length (low: 10, medium: 7, long: 3 poems).
This is a standalone slice of the full benchmark repo, published separately so it
can be loaded directly:
from datasets import load_dataset
poems = load_dataset("artindnr/khayyam-challenge-poems")
poem = poems["low"][0]
For the paired AI-generated prose descriptions of… See the full description on the dataset page: https://huggingface.co/datasets/artindnr/khayyam-challenge-poems.instruction-turkish-poems
Turkish poems for fine-tuning LLMs with instructions.
Instructions created with Google's Gemini-Pro.
For a dataset that has variety of instructions check: beratcmn/rephrased-instruction-turkish-poems
Base dataset:
beratcmn/turkish-poems-cleaned
poemma-10k
Poemma 10K
Dataset for naturalisations of language model outputs in the domain of poems.
poemas
Spanish poems for GPT finetuning
Collection of Spanish poems from www.poemas-del-alma.com.
Chinese-PoemsChinese-Poems这个数据集基于 larryvrh/Chinese-Poems ，使用 硅基流动 的 Qwen/Qwen2.5-14B-Instruct 生成提示词，手动修复部分提示词后生成，估计仍有错漏，欢迎反馈。
poem-metreGong-Poem-Dataset
龚诗完整全文数据集
本仓库发布经过规范化与逐条校验的龚诗全文。当前版本为 v1.0.1-fulltext，校验日期为 2026-07-14。
数据规模
配置
文件
条目数
内容
canonical_poems
data/poems_full.parquet
17
规范诗作全文
editions
data/editions_full.parquet
24
不同来源、版本见证全文
fragments
data/fragments_full.parquet
4
可核验散句
每个配置均同时提供 Parquet、CSV 和 JSONL；Hugging Face 查看器直接读取带显式字段类型的 Parquet。所有 45 条全文记录均完成 Unicode NFC 规范化，并通过目录记录的行数、非空白字符数及 SHA-256 一致性校验；结果见 data/full_text_validation.json。
使用方式
from datasets import… See the full description on the dataset page: https://huggingface.co/datasets/Sakanaction/Gong-Poem-Dataset.