datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
MSMU
MSMU (Massive Spatial Measuring and Understanding Dataset for Spatial Intelligence)
🌐 Homepage | 🤗 Dataset | 📖 arXiv | GitHub
Dataset Details
Dataset Description
We introduce MSMU and MSMU-Bench: a new benchmark designed to enhance and evaluate multimodal models on spatial measuring and understanding. MSMU is featured as metric-accurate spatial annotations which are sourced from high-precision 3D scenes. It contains , 25K images, 700K QA pairs… See the full description on the dataset page: https://huggingface.co/datasets/cpystan/MSMU.mmlu-redux-2.0-for-llama.cppMMLU-redux-v2.0 converted for the llama.cpp perplexity multiple choice tool.
Only valid entries where kept, there is no error based prompting included.
Dataset Card for MMLU-Redux-2.0
MMLU-Redux is a subset of 5,700 manually re-annotated questions across 57 MMLU subjects.
Citation
BibTeX:
@misc{gema2024mmlu,
title={Are We Done with MMLU?},
author={Aryo Pradipta Gema and Joshua Ong Jun Leang and Giwon Hong and Alessio Devoto and Alberto Carlo Maria… See the full description on the dataset page: https://huggingface.co/datasets/Green-Sky/mmlu-redux-2.0-for-llama.cpp.cpt_instruction_datasets
Instruction datasets
Collection of synthetic instruction datasets used during the continued pretraining of Model-small-instr-1, Model-small-instr-2 and Model-small-instr-3. You can currently find these models under: Llama-3.1-Carballo-Instr1 and Llama-3.1-Carballo-Instr3.
Dataset creation
Datasets were created using two different techniques:
Adapting already existing datasets or corpora by modifying their format to make them suitable for including instructions during… See the full description on the dataset page: https://huggingface.co/datasets/proxectonos/cpt_instruction_datasets.CPsyCoun
CPsyCounD
The high-quality multi-turn dialogue dataset, which has a total of 3,134 multi-turn consultation dialogues. CPsyCounD covers nine representative topics and seven classic schools of psychological counseling.
Paper: CPsyCoun
Data analysis
Topic types
Self-growth
Emotion&Stress
Education
Love&Marriage
Family Relationship
Social Relationship
Sex
Career
Mental Disease
Consulting schools
Psychoanalytic Therapy
Cognitive Behavioral Therapy… See the full description on the dataset page: https://huggingface.co/datasets/CAS-SIAT-XinHai/CPsyCoun.mmlu-redux-for-llama.cppMMLU-redux converted for the llama.cpp perplexity multiple choice tool.
Only valid entries where kept, there is no error based prompting included.
Dataset Card for MMLU-Redux
[!TIP]
Please consider using MMLU-Redux-2.0 which contains all 57 MMLU subjects.
MMLU-Redux is a subset of 3,000 manually re-annotated questions across 30 MMLU subjects.
Citation
BibTeX:
@misc{gema2024mmlu,
title={Are We Done with MMLU?},
author={Aryo Pradipta Gema and Joshua Ong… See the full description on the dataset page: https://huggingface.co/datasets/Green-Sky/mmlu-redux-for-llama.cpp.cpc_2015_brasil
SALVAMENTO do Dataset
from datasets import load_dataset
from datasets import Dataset
import pandas as pd
# Carregar os dados do arquivo de texto
df = pd.read_parquet('../data/cpc_2015_cleaned.parquet')
data = {
"livro": df["Livro"],
"capitulo": df["Capitulo"],
"titulo": df["Titulo"],
"secao": df["Secao"],
"subsecao": df["Subsecao"],
"artigo": df["Artigo"]
}
# Dividir o texto em seções
dataset = Dataset.from_pandas(pd.DataFrame(data))… See the full description on the dataset page: https://huggingface.co/datasets/0rakul0/cpc_2015_brasil.oa_cpp_annotate_gen
Dataset Description
This dataset, compiled by Brendan Dolan-Gavitt, contains ~100 thousand c++ functions and GPT-3.5 turbo-generated summaries of the code's purpose.
An example of Brendan's original prompt and GPT-3.5's summary may be found below.
int gg_set_focus_pos(gg_widget_t *widget, int x, int y) {
return 1;
}
Q. What language is the above code written in?
A. C/C++.
Q. What is the purpose of the above code?
A.
This code defines a function called `gg_set_focus_pos` that… See the full description on the dataset page: https://huggingface.co/datasets/casey-martin/oa_cpp_annotate_gen.wiki-events-cpt
Wikipedia Events CPT
jhdlee/wiki-events-cpt is a public research dataset of 150 selected English Wikipedia articles with compact metadata for continual pretraining (CPT).
Split
Articles
Event window (end exclusive)
cohort_a
75
2023-01-01 to 2024-10-01
cohort_b
75
2024-10-01 to 2025-09-01
Each cohort has 25 articles per topic: natural_hazards, elections, and sports. Cohorts group events by their reviewed whole-occurrence intervals; they are not… See the full description on the dataset page: https://huggingface.co/datasets/jhdlee/wiki-events-cpt.cpt-dataset
Hyperswitch CPT Dataset
A comprehensive Continual Pre-Training (CPT) dataset for the Hyperswitch payment processing platform, combining documentation with actual code to build a "world model" understanding of the codebase.
Dataset Description
This dataset was created by mining the Hyperswitch repository and combining it with DeepWiki documentation. It teaches models:
Repository Structure - Where different types of code live
Concept-to-Code Mapping - How abstract concepts… See the full description on the dataset page: https://huggingface.co/datasets/archit11/cpt-dataset.CPIQA
CPIQA: Climate Paper Image Question Answering
Dataset Details
Dataset Name: CPIQA (Climate Paper Image Question Answering)
Paper: CPIQA: Climate Paper Image Question Answering Dataset for Retrieval-Augmented Generation with Context-based Query Expansion
Authors: Rudra Mutalik, Abiram Panchalingam, Loitongbam Gyanendro Singh, Timothy J. Osborn, Ed Hawkins, and Stuart E. Middleton
Github: CPIQA generation and evaluation code
Dataset Summary: CPIQA is a large scale… See the full description on the dataset page: https://huggingface.co/datasets/RudraMutalik/CPIQA.CPsyCounR
CPsyCounR
We carefully selected 3,134 psychological counseling reports after rule-based cleaning, manual rewriting, and human proofreading from well-known Chinese psychological communities Yidianling and Psy525.
The counseling report format: Title, Type, Method, Case Brief, Consultation Process and Experience Thoughts. Each report corresponds to only one case and the consultation process is written from a third-person perspective and does not contain specific dialog.
We offer a… See the full description on the dataset page: https://huggingface.co/datasets/CAS-SIAT-XinHai/CPsyCounR.finmmeval-cfa-cpa
Financial Exam MCQ Training Dataset
A bilingual training dataset of financial and accounting multiple-choice questions in English and Chinese, formatted for instruction tuning and answer selection tasks.
Dataset Structure
Format: Multiple-choice questions
Language: English and Chinese
Domain: Accounting, finance, auditing, taxation, and financial regulations
Size: 596 examples
Files:
train-00000-of-00001-en.parquet
train-00000-of-00001-cn.parquet… See the full description on the dataset page: https://huggingface.co/datasets/Tomas08119993/finmmeval-cfa-cpa.CP-Bench
Continuous Perception Benchmark (CP-Bench)
Overview
The Continuous Perception Benchmark (CP-Bench) is a diagnostic dataset designed to evaluate whether modern vision-language and multimodal models can integrate continuous visual information over time—an ability that is central to human visual perception but largely absent in contemporary architectures. Inspired by the continuous, stream-based nature of human vision, CP-Bench isolates the core requirement of maintaining… See the full description on the dataset page: https://huggingface.co/datasets/ContinuousPerceptionResearch/CP-Bench.biomedical_cpgQA
Dataset Card for the Biomedical Domain
Dataset Summary
This dataset was obtain through github (https://github.com/mmahbub/cpgQA/blob/main/dataset/cpgQA-v1.0.csv?plain=1) to Huggin Face for easier access while fine tuning.
Languages
English (en)
Dataset Structure
The dataset is in a CSV format, with each row representing a single review. The following columns are included:
Title: Categorises the QA.
Context: Gives a context of the QA.
Question: The… See the full description on the dataset page: https://huggingface.co/datasets/chloecchng/biomedical_cpgQA.medqa-cpv
MedQA CPV
Counterfactual Patient Variations of the MedQA-USMLE-4-options test split,
created to measure demographic bias in LLMs for clinical decision-making.
Paper
Benkirane et al. "How Can We Diagnose and Treat Bias in Large Language Models
for Clinical Decision-Making?" NAACL 2025. arXiv:2410.16574
Bias Representation: Before vs. After CPV
Aspect
Before (original MedQA)
After (CPV expansion)
Patient demographics
Implicit — most patients… See the full description on the dataset page: https://huggingface.co/datasets/kenza-ily/medqa-cpv.cpgQA-DE
Dataset Card for cpgQA-DE
cpgQA-DE is a German benchmark for evaluating whether medical question-answering systems can answer questions grounded in German Clinical Practice Guidelines (CPGs).
Evaluation only. Do not use for training.
This is a research benchmark, not a clinical decision-support tool.
Dataset Details
This repository provides four evaluation files.
cpgQA_DE_main.json
Contains 200 validated guideline-based questions with answer keys… See the full description on the dataset page: https://huggingface.co/datasets/MedInf-AI/cpgQA-DE.LongBench-v2-for-llama.cppLongBench v2 converted for the llama.cpp perplexity multiple choice tool.
[!WARNING]
!! Currently does not work, will fix it in the near future. Probably.
LongBench v2: Towards Deeper Understanding and Reasoning on Realistic Long-context Multitasks
🌐 Project Page: https://longbench2.github.io
💻 Github Repo: https://github.com/THUDM/LongBench
📚 Arxiv Paper: https://arxiv.org/abs/2412.15204
LongBench v2 is designed to assess the ability of LLMs to handle long-context problems… See the full description on the dataset page: https://huggingface.co/datasets/Green-Sky/LongBench-v2-for-llama.cpp.medmcqa-cpv
MedMCQA CPV
Counterfactual Patient Variations of the MedMCQA validation split,
created to measure demographic bias in LLMs for clinical decision-making.
Paper
Benkirane et al. "How Can We Diagnose and Treat Bias in Large Language Models
for Clinical Decision-Making?" NAACL 2025. arXiv:2410.16574
Bias Representation: Before vs. After CPV
Aspect
Before (original MedMCQA)
After (CPV expansion)
Patient demographics
Absent — questions typically do not… See the full description on the dataset page: https://huggingface.co/datasets/kenza-ily/medmcqa-cpv.CPP-Code-Solutions
C++ Code Solutions
Features
1000k of Python Code Solutions for Text Generation and Question Answering
C++ Coding Problems labelled by topic and difficulty
Recommendations
Train your Model on Logical Operations and Mathematical Problems Before Training it on this. This is optional for Fine Tuning 2B parameter + models.
Format the prompts in a orderly way when formatting data eg. {question} Solution: {solution} Topic: {topic}
pubmedqa-cpv
PubMedQA CPV
Counterfactual Patient Variations of the PubMedQA pqa_labeled / train split,
created to measure demographic bias in LLMs for biomedical research question answering.
Paper
Benkirane et al. "How Can We Diagnose and Treat Bias in Large Language Models
for Clinical Decision-Making?" NAACL 2025. arXiv:2410.16574
Bias Representation: Before vs. After CPV
Aspect
Before (original PubMedQA)
After (CPV expansion)
Patient demographics
Absent —… See the full description on the dataset page: https://huggingface.co/datasets/kenza-ily/pubmedqa-cpv.CPsyCoun
CPsyCounD
The high-quality multi-turn dialogue dataset, which has a total of 3,134 multi-turn consultation dialogues. CPsyCounD covers nine representative topics and seven classic schools of psychological counseling.
Paper: CPsyCoun
Data analysis
Topic types
Self-growth
Emotion&Stress
Education
Love&Marriage
Family Relationship
Social Relationship
Sex
Career
Mental Disease
Consulting schools
Psychoanalytic Therapy
Cognitive Behavioral Therapy… See the full description on the dataset page: https://huggingface.co/datasets/baicuya/CPsyCoun.MSA-cpt-100b
MSA CPT corpus (cpt_100b)
Continual-pre-training corpus for Memory Sparse Attention (MSA) / Generative
Retrieval on a Qwen3.5 backbone. Built 2026-06-29 by aggregating and
within-source-deduplicating 45 public retrieval / QA datasets into a unified
query↔document contract.
Layout
corpus/ # document side (.jsonl.gz shards)
train/ # query↔positive-doc training pairs (.jsonl.gz shards)
manifest.json
Stats (from manifest.json)
field
value… See the full description on the dataset page: https://huggingface.co/datasets/william-0g/MSA-cpt-100b.idea-first-code-later-cp
Idea First, Code Later: CP Benchmark
Benchmark dataset for the paper: "Idea First, Code Later: Disentangling Problem Solving from Code Generation in Evaluating LLMs for Competitive Programming"
A curated benchmark of 83 competitive programming problems designed for evaluating LLMs on algorithmic problem-solving separately from code generation.
Motivation
We curate problems from seven contests that are not hosted on major public CP platforms (e.g., Codeforces, AtCoder).… See the full description on the dataset page: https://huggingface.co/datasets/samahadhoud/idea-first-code-later-cp.cpdc2025-dialogue-reasoning-gen@misc{2510.13586,
Author = {Pasin Buakhaw and Kun Kerdthaisong and Phuree Phenhiran and Pitikorn Khlaisamniang and Supasate Vorathammathorn and Piyalitt Ittichaiwong and Nutchanon Yongsatianchot},
Title = {Deflanderization for Game Dialogue: Balancing Character Authenticity with Task Execution in LLM-based NPCs},
Year = {2025},
Eprint = {arXiv:2510.13586},
}
cpa-tax-scenarios-2026
CPA Tax Impact Scenarios 2026
Pre-calculated mortgage tax scenarios for 768 combinations of income, loan amount, property tax, and filing status. Covers purchase, refinance, divorce buyout, and reverse mortgage scenarios.
2026 Standard Deductions
Filing Status
Standard Deduction
Single
$15,000
Married Jointly
$30,000
Head of Household
$22,500
Expert
Wendy Thompson, CDLP (Certified Divorce Lending Professional)
Disclaimer
For… See the full description on the dataset page: https://huggingface.co/datasets/wendymthompson/cpa-tax-scenarios-2026.neurips-2026-cpx
neurips-2026-cpx — Korean OSCE history-taking dialogues with a GPT-4o virtual standardized patient
49 text-based history-taking dialogue sessions between 17 senior Korean
medical-student participants (Years 3–4 of a 6-year curriculum) and a
GPT-4o-driven virtual standardized patient (VSP). Released as the
empirical evaluation dataset accompanying our NeurIPS 2026 submission.
Sessions: 49
Participants: 17 (anonymised to R001–R017)
Total QA turns: 1,763
Language: Korean
Domain:… See the full description on the dataset page: https://huggingface.co/datasets/neurips-2026-cpx/neurips-2026-cpx.cpa-tax-scenarios-2026
CPA Tax Scenarios 2026
768 CPA tax impact scenarios by income, loan, filing status.
Details
Records: 768
Format: JSONL
License: CC-BY-4.0
Last Updated: March 2026
Verified By: Wendy Thompson, CPA, CDLP, NMLS #504814
Publisher: Wendy Thompson Lending Team
Thompson Alpha Logic
Data models calculating the after-tax cost of mortgage debt across purchase, refinance, divorce buyout, and reverse mortgage scenarios. Compares itemized vs. standard deduction ($30K… See the full description on the dataset page: https://huggingface.co/datasets/Wendy-Thompson-Lending-Team/cpa-tax-scenarios-2026.cpdc2025-function-gen@misc{2510.13586,
Author = {Pasin Buakhaw and Kun Kerdthaisong and Phuree Phenhiran and Pitikorn Khlaisamniang and Supasate Vorathammathorn and Piyalitt Ittichaiwong and Nutchanon Yongsatianchot},
Title = {Deflanderization for Game Dialogue: Balancing Character Authenticity with Task Execution in LLM-based NPCs},
Year = {2025},
Eprint = {arXiv:2510.13586},
}
cpdc2025-dialogue-gen@misc{2510.13586,
Author = {Pasin Buakhaw and Kun Kerdthaisong and Phuree Phenhiran and Pitikorn Khlaisamniang and Supasate Vorathammathorn and Piyalitt Ittichaiwong and Nutchanon Yongsatianchot},
Title = {Deflanderization for Game Dialogue: Balancing Character Authenticity with Task Execution in LLM-based NPCs},
Year = {2025},
Eprint = {arXiv:2510.13586},
}
cpath-mcgill-ubc
Canadian Universities Q&A Dataset (CPath)
A comprehensive question-answering dataset focused on Canadian universities' programs, admissions, and academic information, specifically covering McGill University and the University of British Columbia (UBC).
Dataset Description
This dataset contains carefully curated question-answer pairs extracted from official university websites and documentation. It is designed to serve as a reliable resource for understanding academic… See the full description on the dataset page: https://huggingface.co/datasets/houcine-bdk/cpath-mcgill-ubc.