datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
ReviewRebuttal
Introduction
This dataset is the largest real-world consistency-ensured dataset for peer review, which features the widest range of conferences and the most complete review stages, including initial submissions, reviews, ratings and confidence, aspect ratings, rebuttals, discussions, score changes, meta-reviews, and final decisions.
Paper: https://arxiv.org/abs/2505.07920
If our dataset can help you, please consider include the following citation in your publications:… See the full description on the dataset page: https://huggingface.co/datasets/Daoze/ReviewRebuttal.amazon_reviews_multiWe provide an Amazon product reviews dataset for multilingual text classification. The dataset contains reviews in English, Japanese, German, French, Chinese and Spanish, collected between November 1, 2015 and November 1, 2019. Each record in the dataset contains the review text, the review title, the star rating, an anonymized reviewer ID, an anonymized product ID and the coarse-grained product category (e.g. ‘books’, ‘appliances’, etc.) The corpus is balanced across stars, so each star rating constitutes 20% of the reviews in each language.
For each language, there are 200,000, 5,000 and 5,000 reviews in the training, development and test sets respectively. The maximum number of reviews per reviewer is 20 and the maximum number of reviews per product is 20. All reviews are truncated after 2,000 characters, and all reviews are at least 20 characters long.
Note that the language of a review does not necessarily match the language of its marketplace (e.g. reviews from amazon.de are primarily written in German, but could also be written in English, etc.). For this reason, we applied a language detection algorithm based on the work in Bojanowski et al. (2017) to determine the language of the review text and we removed reviews that were not written in the expected language.TTCW-Based-Review
TTCW Creative Writing Evaluation Dataset
If you use this dataset in your research, please cite our paper — it helps support ongoing academic work. Citation details are at the bottom of this page.
Dataset Description
Summary
A supervised fine-tuning (SFT) dataset for training LLMs to act as creative writing evaluators. Each example contains a creative story and four message-format columns representing different evaluation objectives — from… See the full description on the dataset page: https://huggingface.co/datasets/VibrantVista/TTCW-Based-Review.Paper-Review-Dataset
Dataset Card for Paper Review Dataset (ICLR 2023-2025)
Dataset Description
This dataset contains paper submissions and review data from the International Conference on Learning Representations (ICLR) for the years 2023, 2024, and 2025. The data is sourced from OpenReview, an open peer review platform that hosts the review process for top ML conferences.
Focus on Review Data
This dataset emphasizes the peer review ecosystem surrounding academic papers. Each… See the full description on the dataset page: https://huggingface.co/datasets/AgentAlphaAGI/Paper-Review-Dataset.SWE-Review-Chat
SWE-Review-Chat: A Dataset of Code Review Conversations and Human-AI Collaboration in Agentic Code Review
Paper: https://arxiv.org/abs/2607.13196
GitHub: https://github.com/suzhenxzhong/SWE-Review-Chat
SWE-Review-Chat is a large-scale dataset of real-world code review conversations from pull requests of 207 popular GitHub projects, spanning the transition from human-centric to LLM-assisted and agentic code review by AI agents.
📊 Dataset Overview
Field… See the full description on the dataset page: https://huggingface.co/datasets/Suzhen/SWE-Review-Chat.reviewsFollowing packages are necessary to compile the model in C:
sudo apt install gcc-7
sudo apt-get install build-essential
for i in range(1,21):
!wget https://huggingface.co/datasets/teragron/reviews/resolve/main/chunk_{i}.bin
git clone https://github.com/karpathy/llama2.c.git
cd llama2.c
pip install -r requirements.txt
Path: data/TinyStories_all_data
amazon_us_reviewsAmazon Customer Reviews (a.k.a. Product Reviews) is one of Amazons iconic products. In a period of over two decades since the first review in 1995, millions of Amazon customers have contributed over a hundred million reviews to express opinions and describe their experiences regarding products on the Amazon.com website. This makes Amazon Customer Reviews a rich source of information for academic researchers in the fields of Natural Language Processing (NLP), Information Retrieval (IR), and Machine Learning (ML), amongst others. Accordingly, we are releasing this data to further research in multiple disciplines related to understanding customer product experiences. Specifically, this dataset was constructed to represent a sample of customer evaluations and opinions, variation in the perception of a product across geographical regions, and promotional intent or bias in reviews.
Over 130+ million customer reviews are available to researchers as part of this release. The data is available in TSV files in the amazon-reviews-pds S3 bucket in AWS US East Region. Each line in the data files corresponds to an individual review (tab delimited, with no quote and escape characters).
Each Dataset contains the following columns:
- marketplace: 2 letter country code of the marketplace where the review was written.
- customer_id: Random identifier that can be used to aggregate reviews written by a single author.
- review_id: The unique ID of the review.
- product_id: The unique Product ID the review pertains to. In the multilingual dataset the reviews for the same product in different countries can be grouped by the same product_id.
- product_parent: Random identifier that can be used to aggregate reviews for the same product.
- product_title: Title of the product.
- product_category: Broad product category that can be used to group reviews (also used to group the dataset into coherent parts).
- star_rating: The 1-5 star rating of the review.
- helpful_votes: Number of helpful votes.
- total_votes: Number of total votes the review received.
- vine: Review was written as part of the Vine program.
- verified_purchase: The review is on a verified purchase.
- review_headline: The title of the review.
- review_body: The review text.
- review_date: The date the review was written.amazon_us_reviewsAmazon Customer Reviews (a.k.a. Product Reviews) is one of Amazons iconic products. In a period of over two decades since the first review in 1995, millions of Amazon customers have contributed over a hundred million reviews to express opinions and describe their experiences regarding products on the Amazon.com website. This makes Amazon Customer Reviews a rich source of information for academic researchers in the fields of Natural Language Processing (NLP), Information Retrieval (IR), and Machine Learning (ML), amongst others. Accordingly, we are releasing this data to further research in multiple disciplines related to understanding customer product experiences. Specifically, this dataset was constructed to represent a sample of customer evaluations and opinions, variation in the perception of a product across geographical regions, and promotional intent or bias in reviews.
Over 130+ million customer reviews are available to researchers as part of this release. The data is available in TSV files in the amazon-reviews-pds S3 bucket in AWS US East Region. Each line in the data files corresponds to an individual review (tab delimited, with no quote and escape characters).
Each Dataset contains the following columns:
- marketplace: 2 letter country code of the marketplace where the review was written.
- customer_id: Random identifier that can be used to aggregate reviews written by a single author.
- review_id: The unique ID of the review.
- product_id: The unique Product ID the review pertains to. In the multilingual dataset the reviews for the same product in different countries can be grouped by the same product_id.
- product_parent: Random identifier that can be used to aggregate reviews for the same product.
- product_title: Title of the product.
- product_category: Broad product category that can be used to group reviews (also used to group the dataset into coherent parts).
- star_rating: The 1-5 star rating of the review.
- helpful_votes: Number of helpful votes.
- total_votes: Number of total votes the review received.
- vine: Review was written as part of the Vine program.
- verified_purchase: The review is on a verified purchase.
- review_headline: The title of the review.
- review_body: The review text.
- review_date: The date the review was written.reviewarena
ReviewArena
ReviewArena accompanies the NeurIPS Evaluations & Datasets submission ReviewArena: A Large-Scale Cross-Conference Dataset and Benchmark for LLM Peer Review.
This release is a large, multi-conference corpus of peer-reviewed papers + their reviews + author rebuttals + acceptance decisions, harvested from OpenReview and aligned with OCR'd full-text markdown of each paper PDF where available.
51,529 papers
196,099 reviews
558,785 OCR'd PDF pages (markdown inlined… See the full description on the dataset page: https://huggingface.co/datasets/Samarth0710/reviewarena.Paper-Review-Dataset
Dataset Card for Paper Review Dataset (ICLR 2023-2025)
Dataset Description
This dataset contains paper submissions and review data from the International Conference on Learning Representations (ICLR) for the years 2023, 2024, and 2025. The data is sourced from OpenReview, an open peer review platform that hosts the review process for top ML conferences.
Focus on Review Data
This dataset emphasizes the peer review ecosystem surrounding academic papers. Each… See the full description on the dataset page: https://huggingface.co/datasets/WENJINLIU/Paper-Review-Dataset.SWE-Review-Bench
SWE-Review-Bench
A benchmark for evaluating code review agents on real-world GitHub issues with executable verification. Each instance pairs a GitHub issue with an AI-generated pull request; the reviewer must decide whether the PR resolves the issue and, if not, provide structured feedback to guide revision.
Project Page | Paper | Code | Benchmark | Training Data | Claude Code Plugin
About SWE-Review
SWE-Review is a framework for closing the issue-resolution… See the full description on the dataset page: https://huggingface.co/datasets/Lego-X/SWE-Review-Bench.SWE-Review-Traj
SWE-Review-Traj
8,914 decision-correct agentic code review trajectories for training open-source code review models. Each trajectory captures a full multi-turn review session where an AI agent explores a repository, traces the root cause of a bug, and produces a structured review decision — all verified against executable test suites.
Project Page | Paper | Code | Benchmark | Training Data | Claude Code Plugin
About SWE-Review
SWE-Review is a framework for… See the full description on the dataset page: https://huggingface.co/datasets/Lego-X/SWE-Review-Traj.reviewbench
ReviewBench
A large, multi-conference corpus of peer-reviewed papers + their reviews + author rebuttals + acceptance decisions, harvested from OpenReview and aligned with OCR'd full-text markdown of every paper.
51,529 papers
196,099 reviews
558,785 OCR'd PDF pages (markdown inlined per row)
7 conferences, 22 venue/year combinations, 2020 – 2026
from datasets import load_dataset
ds = load_dataset("/reviewbench")
print(ds)
# DatasetDict({
# neurips: Dataset(num_rows=...)# iclr:… See the full description on the dataset page: https://huggingface.co/datasets/Samarth0710/reviewbench.amazon_reviews_multiPlease refer to https://huggingface.co/datasets/defunct-datasets/amazon_reviews_multi.hotel_reviews
Dataset Card for "hotel_reviews"
More Information needed
SciCode-Runnable-Benchmark-ReviewedReviewer2_PGE_cleaned
Cleaned Review Dataset for Reviewer2
This is a cleaned version of our dataset and can be directly used for fine-tuning. The raw data files including metadata for each paper is in this directory.
venue: venue of the paper;
paper_content: content of the paper divided into sections
prompt: prompt generated for the review based on our PGE pipeline
format: the format of the review
review: human-written review for the paper
Dataset Sources
We incorporate parts of the… See the full description on the dataset page: https://huggingface.co/datasets/GitBag/Reviewer2_PGE_cleaned.pi-diff-review
Coding agent session traces for badlogicgames/pi-diff-review
This dataset contains redacted coding agent session traces collected while working on https://github.com/badlogic/pi-diff-review.git. The traces were exported with pi-share-hf from a local pi workspace and filtered to keep only sessions that passed deterministic redaction and LLM review.
Data description
Each *.jsonl file is a redacted pi session. Sessions are stored as JSON Lines files where each line… See the full description on the dataset page: https://huggingface.co/datasets/badlogicgames/pi-diff-review.task929_products_reviews_classification
Dataset Card for Natural Instructions (https://github.com/allenai/natural-instructions) Task: task929_products_reviews_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… See the full description on the dataset page: https://huggingface.co/datasets/Lots-of-LoRAs/task929_products_reviews_classification.public_domain_review_filtered
Public Domain Review
Description
The Public Domain Review is an online journal dedicated to exploration of works of art and literature that have aged into the public domain.
We collect all articles published in the Public Domain Review under a CC BY-SA license.
Dataset Statistics
Documents
UTF-8 GB
1,406
0.007
License Issues
While we aim to produce datasets with completely accurate licensing information, license laundering and… See the full description on the dataset page: https://huggingface.co/datasets/common-pile/public_domain_review_filtered.Paper-Reviewing-Exam
Paper-Reviewing-Exam
Harbor tasks for collecting peer reviews of scientific manuscripts. Each
task hands an agent a complete paper and asks it to write a review; the review
and its structured companion are archived for human experts to assess afterwards.
These tasks do not score reviews. The verifier checks only that
review.md was written and is substantive and that review.json follows the
published schema. There is no LLM judge: deciding whether findings are correct
or useful… See the full description on the dataset page: https://huggingface.co/datasets/Jack-Jieke-Wu/Paper-Reviewing-Exam.app_reviews_moddedThe modded version of https://huggingface.co/datasets/app_reviews
Dataset Card for app-reviews-modded
Dataset Summary
It is a large dataset of Android applications belonging to 23 different app categories, which provides an overview of the types of feedback users report on the apps and documents the evolution of the related code metrics. The dataset contains about 395 applications of the F-Droid repository, including around 600 versions, 280,000 user reviews (extracted… See the full description on the dataset page: https://huggingface.co/datasets/Sharathhebbar24/app_reviews_modded.task1577_amazon_reviews_multi_japanese_language_classification
Dataset Card for Natural Instructions (https://github.com/allenai/natural-instructions) Task: task1577_amazon_reviews_multi_japanese_language_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… See the full description on the dataset page: https://huggingface.co/datasets/Lots-of-LoRAs/task1577_amazon_reviews_multi_japanese_language_classification.goodreads-reviews
Goodreads Reviews (deduplicated)
~15,739,967 book reviews scraped from Goodreads, deduplicated.
Columns
Column
Type
Description
user_id
string
Anonymised user hash
book_id
string
Goodreads book ID
review_id
string
Unique review ID
rating
int8
1–5 star rating (0 = no rating)
review_text
string
Full review text
date_added
string
Date added to shelf
date_updated
string
Date last updated
read_at
string
Date finished reading
started_at
string
Date… See the full description on the dataset page: https://huggingface.co/datasets/vngclinh/goodreads-reviews.pi-diff-review
Coding agent session traces for badlogicgames/pi-diff-review
This dataset contains redacted coding agent session traces collected while working on https://github.com/badlogic/pi-diff-review.git. The traces were exported with pi-share-hf from a local pi workspace and filtered to keep only sessions that passed deterministic redaction and LLM review.
Data description
Each sessions/*.jsonl file is a redacted pi session. Sessions are stored as JSON Lines files where… See the full description on the dataset page: https://huggingface.co/datasets/cfahlgren1/pi-diff-review.PeerPrism
PeerPrism
PeerPrism is a benchmark for idea–text provenance disentanglement in peer
review. Instead of treating authorship as binary (human vs. AI), PeerPrism separates
two independent axes:
Idea provenance (idea_origin) — who contributed the evaluative reasoning.
Text provenance (text_origin) — who produced the surface form of the text.
This lets the dataset cover fully-human, fully-LLM, and multiple hybrid regimes,
so that LLM-detection methods can be studied in realistic… See the full description on the dataset page: https://huggingface.co/datasets/Reviewerly/PeerPrism.public_domain_review
Public Domain Review
Description
The Public Domain Review is an online journal dedicated to exploration of works of art and literature that have aged into the public domain.
We collect all articles published in the Public Domain Review under a CC BY-SA license.
Dataset Statistics
Documents
UTF-8 GB
1,412
0.007
License Issues
While we aim to produce datasets with completely accurate licensing information, license laundering and… See the full description on the dataset page: https://huggingface.co/datasets/common-pile/public_domain_review.Booking-Hotel-Reviews-Dataset
Booking.com Hotel Reviews Dataset – 4.3K Sample
A rich, structured dataset of hotel reviews collected from Booking.com, featuring a unique split of positive and negative review text, reviewer country, stay dates, traveler tags, and hotel location data. Ideal for sentiment analysis, aspect-based opinion mining, travel AI, hospitality recommendation systems, and LLM fine-tuning on real-world review data.
Dataset Overview
Field
Details
Source… See the full description on the dataset page: https://huggingface.co/datasets/crawlfeeds/Booking-Hotel-Reviews-Dataset.Paper-Reviewing-Exam-Trails
Paper-Reviewing-Exam-Trails
Review-run trail archive for the
Paper-Reviewing-Exam
benchmark. Each trail captures one agent review run: the submitted review.md and review.json,
the agent's brain/, manifests, logs, and verifier output, so a human expert
can assess the review against the exact materials it was written from.
Trails contain review content and are public. Review text can be read from
this dataset; uploading a trail publishes it. Confirm you accept that exposure
before… See the full description on the dataset page: https://huggingface.co/datasets/Jack-Jieke-Wu/Paper-Reviewing-Exam-Trails.ai-reviewer-diagnostic-data
AI Reviewer Diagnostic Data
Dataset for Where Do LLMs Go Wrong? Diagnosing Automated Peer Review via Aspect-Guided Multi-Level Perturbation.Jiatao Li, Yanheng Li, Xinyu Hu, Mingqi Gao, Xiaojun Wan. CIKM 2025. DOI: https://doi.org/10.1145/3746252.3761274
Companion resources
Resource
Link
Use
Paper
https://doi.org/10.1145/3746252.3761274
CIKM 2025 publication
Code
https://github.com/PKU-ONELab/where-do-llms-go-wrong
scripts, prompts, docs, and reproduction… See the full description on the dataset page: https://huggingface.co/datasets/PKU-ONELab/ai-reviewer-diagnostic-data.