datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
Feedback-Collection
Dataset Card
Dataset Summary
The Feedback Collection is a dataset designed to induce fine-grained evaluation capabilities into language models.\
Recently, proprietary LLMs (e.g., GPT-4) have been used to evaluate long-form responses. In our experiments, we found that open-source LMs are not capable of evaluating long-form responses, showing low correlation with both human evaluators and GPT-4.\
In our paper, we found that by (1) fine-tuning feedback generated by GPT-4… See the full description on the dataset page: https://huggingface.co/datasets/prometheus-eval/Feedback-Collection.wb-feedbacks
Dataset Card for Wildberries products
Dataset Summary
The dataset contains product reviews from the Russian marketplace Wildberries, collected by mining about The dataset was collected by bruteforcing possible product identifiers (about 230 million) and querying all available feedbacks for them. The data are stored in zstd-archives containing jsonl-files. The 'nmId' in the dataset usually corresponds to the valid product article on the site, but sometimes reviews are… See the full description on the dataset page: https://huggingface.co/datasets/nyuuzyou/wb-feedbacks.ultra_feedback_dutch_cleaned
Ultra Feedback Dutch Cleaned
This is a cleaned version of BramVanroy/ultra_feedback_dutch, based on the cleaning done by Argilla on the original Ultra Feedback dataset. Another difference is that we only include GEITje 7B Ultra and GPT-4-Turbo. GEITje chat, which was used in the original dataset, is not used.
After cleaning I also generated replies for other models (like TowerInstruct, Mistral), but the results were too poor (in Dutch) to include so we only kept the GEITje Ultra and… See the full description on the dataset page: https://huggingface.co/datasets/BramVanroy/ultra_feedback_dutch_cleaned.openhands-feedback
OpenHands Feedback Dataset 🙌
Dataset Description
What is OpenHands Feedback?
The OpenHands Feedback Dataset is a collection of user interactions and feedback with the OpenHands AI coding assistant. This dataset contains real-world examples of how users interact with AI coding assistants, including both successful and unsuccessful interactions, along with user feedback on the quality and helpfulness of the responses.
The dataset currently contains 275 examples… See the full description on the dataset page: https://huggingface.co/datasets/OpenHands/openhands-feedback.smar-feedback
SMAR AQI 3B -- Sovereign Feedback Dataset
User-curated quality signals from the SMAR AQI 3B chat workspace (smar.neugi.com/chat).
What Is This?
Every time a user clicks thumbs up/down on a chatbot response, that feedback flows here automatically. This dataset powers the continuous improvement loop for SMAR AQI 3B -- a sovereign AI model trained from scratch on Indonesian, English, and Arabic.
Lifecycle (A-Z)
1. Collection (Automatic)… See the full description on the dataset page: https://huggingface.co/datasets/atharia179/smar-feedback.ultra-feedback-pairedmultiturn-feedback
MultiTurn Feedback Dataset
Multi-turn conversation feedback dataset with sparse and dense annotations.
Dataset Description
This dataset contains human feedback annotations for paper "User Feedback in Human-LLM Dialogues:
A Lens to Understand Users But Noisy as a Learning Signal". It includes two evaluation subsets:
Sparse: 75 conversations from LMSYS-Chat-1M with sparse feedback
Dense: 74 conversations from LMSYS-Chat-1M + 34 WildChat with dense feedback
Labels… See the full description on the dataset page: https://huggingface.co/datasets/yuhan-nlp/multiturn-feedback.Code-Feedback-decontaminationA decontaminated version of m-a-p/Code-Feedback.
The excluded (28) files are "contaminated" with only two code segments:
simple GCD function: while b: a, b = b, a % b return a
sum_to_n solution: return sum(range(n + 1))
And reformated to sharegpt.
Decontamination is done in the same way as Magicoder (ie., bigcode decontamination process), which uses a substring-match-finding method to find overlaps between a target dataset and the following standard benchmarks:
HumanEval
MBPP… See the full description on the dataset page: https://huggingface.co/datasets/Leon-Leee/Code-Feedback-decontamination.ultra_feedback_dutch
Dataset Card for Ultra Feedback Dutch
[!WARNING]
It is recommended to use the cleaned version for your experiments.
Citation
If you use this dataset, GEITje 7B Ultra (SFT) or any of its derivatives or quantizations, place cite the following paper:
@misc{vanroy2024geitje7bultraconversational,
title={GEITje 7B Ultra: A Conversational Model for Dutch},
author={Bram Vanroy},
year={2024},
eprint={2412.04092},
archivePrefix={arXiv}… See the full description on the dataset page: https://huggingface.co/datasets/BramVanroy/ultra_feedback_dutch.rubric-feedback-bench
Rubric Feedback Bench
Paper | Code
A dataset designed for optimizing LLMs with an LLM evaluator that can provide quantitative scoring and/or qualitative textual feedback across diverse task categories.
Overview
Rubric Feedback Bench is a specialized dataset containing prompts paired with detailed, task-specific rubrics for LLM evaluation. Unlike traditional evaluation datasets that focus solely on scoring, this benchmark emphasizes the evaluator's ability to provide… See the full description on the dataset page: https://huggingface.co/datasets/vicgalle/rubric-feedback-bench.2026_08_26_omni_math_train_feedback_adherence_gemma3_12b_gemma4_31b_candidates
Omni-MATH train feedback-adherence candidates
Production candidate data for studying whether a student follows teacher feedback.
Student: google/gemma-3-12b-it
Teacher and adherence judge: google/gemma-4-31B-it
Source problems: LLParallax/Omni-MATH-filtered, train partition after a fixed 512-problem test split
Source trajectories: LLParallax/2026_07_16_collect_omni_math_gemma3_12b_gemma4_31b
Collection config:… See the full description on the dataset page: https://huggingface.co/datasets/1337xyz1337xyz/2026_08_26_omni_math_train_feedback_adherence_gemma3_12b_gemma4_31b_candidates.Feedback_Friction_Dataset
Feedback Friction Dataset
This dataset contains the LLaMA-4 Maverick results from the iterative feedback experiments described in the paper: FEEDBACK FRICTION: LLMs Struggle to Fully Incorporate External Feedback.
Github Repository: https://github.com/JHU-CLSP/Feedback-Friction
Note: While the paper evaluated multiple frontier models including LLaMA-3.3-70B-Instruct, LLaMA-4-Scout-17B-16E-Instruct, Claude 3.7 Sonnet, and Claude 3.7 Sonnet with Extended Thinking, this dataset… See the full description on the dataset page: https://huggingface.co/datasets/Dongwei/Feedback_Friction_Dataset.zero-shot-teacher-feedbackTLDR: Classification + text generation feedback on classroom transcripts.
Is ChatGPT a Good Teacher Coach? Measuring Zero-Shot Performance For Scoring and Providing Actionable Insights on Classroom Instruction
Paper •
Project Page •
Code
Authors: Rose E. Wang and Dorottya Demszky
In the Proceedings of Innovative Use of NLP for Building Educational Applications 2023
Selected as the Ambassador Paper for BEA 2023! 🎉 To be presented at AIED 2024.
If you find… See the full description on the dataset page: https://huggingface.co/datasets/rose-e-wang/zero-shot-teacher-feedback.Feedback-Collection-ru
Feedback-Collection-ru
This is russian version of prometheus-eval/Feedback-Collection translated using Google Translate.
Feedback-Collection
Dataset Card
Dataset Summary
The Feedback Collection is a dataset designed to induce fine-grained evaluation capabilities into language models.\
Recently, proprietary LLMs (e.g., GPT-4) have been used to evaluate long-form responses. In our experiments, we found that open-source LMs are not capable of evaluating long-form responses, showing low correlation with both human evaluators and GPT-4.\
In our paper, we found that by (1) fine-tuning feedback generated by GPT-4… See the full description on the dataset page: https://huggingface.co/datasets/pharaouk/Feedback-Collection.socratic-cybernetics-textbook
De Balans-Code: Cybernetische Synthese van Hardware, Software en Kosmische Ankers
Het Geconsolideerde Systeem-Protocol — Candor-Modus Editie
Systeembasis: Decentrale Validatienetwerken ($TAO) vanaf 2026
Hoofdstuk 1: Systeemarchitectuur & Fysieke Infrastructuur
Socio-Economische Centralisatie, Cybernetische Balans en Empirische Analyse van Megalithische Logistiek
1.1 INLEIDING: SYSTEMISCHE ENTROPIE EN HET… See the full description on the dataset page: https://huggingface.co/datasets/Feedbackloop369/socratic-cybernetics-textbook.ultra_feedback_dutch_cleaned_multi
Ultra Feedback Dutch Cleaned
This dataset should not be used unless you are interest in all model generations. Instead, refer to the rated and further filtered version.
This is a cleaned version of BramVanroy/ultra_feedback_dutch, based on the cleaning done by Argilla on the original Ultra Feedback dataset.
It contains multiple LM responses from:
GEITje-7B-ultra
TowerInstruct-13B-v0.1
TowerInstruct-7B-v0.2
GEITje-7B-chat
gpt-4-turbo
llama-2-13b-chat-dutch
sl_code_feedback
Slovenian Code Feedback Instruction Translated Dataset
This dataset is a 40000-sample subset of code-instruction pairs, derived from existing instruction datasets. It is primarily based on the CodeFeedback-Filtered-Instruction dataset (https://huggingface.co/datasets/m-a-p/CodeFeedback-Filtered-Instruction), which contains curated code instruction–response pairs grouped by programming language (e.g., Python, SQL, Java, JavaScript, Rust, C#) and by source (e.g.… See the full description on the dataset page: https://huggingface.co/datasets/cjvt/sl_code_feedback.animal-alignment-feedback
Open Paws Animal Alignment Feedback
🐾 Human feedback and preference data for aligning AI with animal advocacy values
Overview
This dataset is part of the Open Paws initiative to develop AI training data aligned with animal liberation and advocacy principles. Created to train AI systems that understand and promote animal welfare, rights, and liberation.
Dataset Details
Dataset Type: Feedback Data
Format: CSV (Comma-separated values)
Languages: Multilingual… See the full description on the dataset page: https://huggingface.co/datasets/open-paws/animal-alignment-feedback.customer-feedback-action-plans
Customer Feedback → Action Plans
A small, practical dataset that maps raw customer feedback (e.g., restaurant reviews) to actionable recommendations with optional aspect annotations and reasoning. Useful for training instruction-following models, aspect-aware summarizers, or classification heads that support the generation task.
Files & Splits
train.csv — main training split for generation.
validation.csv — validation split for generation.
train_aux_classification.csv —… See the full description on the dataset page: https://huggingface.co/datasets/Programmer-RD-AI/customer-feedback-action-plans.ultra-feedback-js-instruct
Ultra Feedback JS Instruct
A subset of the wonderful and elegant Ultra Feedback dataset. Ratings are 1-5 ( inclusive )
Generation Notebook
Example
{
"inst": "Refine the subsequent JavaScript code snippet to obtain the cumulative total of elements within an array:\nlet numbers = [1, 2, 3]",
"author": "codellama-34b-instruct",
"fun": "let numbers = [1, 2, 3];\nlet total = numbers.reduce((a, b) => a + b);\nconsole.log(total); // Output: 6",
"rating": "5"
}
rubric-feedback-bench
Rubric Feedback Bench
A dataset designed for optimizing LLMs with an LLM evaluator that can provide quantitative scoring and/or qualitative textual feedback across diverse task categories.
Overview
Rubric Feedback Bench is a specialized dataset containing prompts paired with detailed, task-specific rubrics for LLM evaluation. Unlike traditional evaluation datasets that focus solely on scoring, this benchmark emphasizes the evaluator's ability to provide detailed textual… See the full description on the dataset page: https://huggingface.co/datasets/anon590/rubric-feedback-bench.Giving_and_Receiving_Feedback_Content_1
Giving and Receiving Feedback Content 1
This corpus was automatically generated by the Deku Corpus Builder for use in RAG-based AI applications.
Dataset Structure
Each record contains:
text: The content text
source_url: Original source URL
source_title: Title of the source document
source_domain: Domain of the source
license_type: License classification (e.g. public_domain, cc_by, cc_by_sa)
attribution_required: Boolean — True for CC BY / CC BY-SA and other… See the full description on the dataset page: https://huggingface.co/datasets/PhillyMac/Giving_and_Receiving_Feedback_Content_1.Giving_and_Receiving_Feedback__Content_2
Giving and Receiving Feedback Content 2
This corpus was automatically generated by the Deku Corpus Builder for use in RAG-based AI applications.
Dataset Structure
Each record contains:
text: The content text
source_url: Original source URL
source_title: Title of the source document
source_domain: Domain of the source
license_type: License classification (e.g. public_domain, cc_by, cc_by_sa)
attribution_required: Boolean — True for CC BY / CC BY-SA and other… See the full description on the dataset page: https://huggingface.co/datasets/PhillyMac/Giving_and_Receiving_Feedback__Content_2.user-feedback-dpo
📝 ChatTune-DPO (farabi-lab/user-feedback-dpo)
📌 Dataset Summary
ChatTune-DPO is a dataset containing structured user interactions with a Large Language Model (LLM), focusing on user feedback for Direct Preference Optimization (DPO) training. The dataset includes dialogues where users interact with the model, provide likes/dislikes, and corrections to responses.
This dataset is designed to enhance model alignment with human preferences by leveraging both explicit and… See the full description on the dataset page: https://huggingface.co/datasets/farabi-lab/user-feedback-dpo.human-feedback-mentor-sessions-india
license: other
task_categories:
audio-classification
automatic-speech-recognition
text-generation
language:
en
hi
bn
ta
te
ml
mr
or
as
pa
pretty_name: Human Feedback Mentor Sessions India
size_categories:
1K<n<10K
Human Feedback Mentor Sessions India
Dataset Description
A rare and high-value collection of recorded aspirant-mentor interactions
capturing real guidance, feedback, and reasoning corrections in the context
of Indian government exam preparation. Produced by… See the full description on the dataset page: https://huggingface.co/datasets/DataOrigin/human-feedback-mentor-sessions-india.