datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
financial-qa-10KFinQA_TAT-QA_financial_finetuning_dataset
Dataset Summary
This dataset provides a unified, flattened context / question / answer format for
question answering over financial documents that combine tabular and textual data. It is
built to support training and evaluating models on numerical and discrete reasoning
tasks in the finance domain, drawing on the structure and style of established
finance-QA benchmarks such as TAT-QA and FinQA.
Each example pairs a passage of financial context (derived from a table and/or… See the full description on the dataset page: https://huggingface.co/datasets/hellotayssir/FinQA_TAT-QA_financial_finetuning_dataset.adaption-financial-tat-qa-pairs
This dataset is a remastered version prepared using Adaption's Adaptive Data platform.
adaption-financial_tat_qa_pairs
This dataset consists of question-and-answer pairs derived from corporate financial reports, specifically Form 10-K filings. The prompts inquire about specific numerical metrics, year-over-year comparisons, and qualitative explanations for financial trends or organizational changes. Completions provide precise values, percentages, or direct textual excerpts… See the full description on the dataset page: https://huggingface.co/datasets/dipanjann/adaption-financial-tat-qa-pairs.adaption-financial-qa-pairs
This dataset is a remastered version prepared using Adaption's Adaptive Data platform.
adaption-financial_qa_pairs
This dataset consists of question-and-answer pairs focused on extracting specific financial metrics from corporate reports. The prompts inquire about percentages, monetary values, growth rates, and operational statistics such as store counts or debt changes. Each completion provides a precise numerical answer derived from financial statements or related… See the full description on the dataset page: https://huggingface.co/datasets/dipanjann/adaption-financial-qa-pairs.financial_regulatory_qafinancial-qa
Financial-QA Dataset Card
Dataset Summary
The Financial-QA dataset is a collection of 50 financial questions created using Llama 3, accompanied by detailed ground truth answers. The dataset also includes two additional prompts providing varying context to the questions. Each entry in the dataset consists of a question, a ground truth answer, and the expected response. The dataset is publicly available on Hugging Face.
Dataset Structure
Data Instances… See the full description on the dataset page: https://huggingface.co/datasets/zeitgeist-ai/financial-qa.QA-Dataset-Financial-Informationeu-financial-regulatory-qa
EU Financial Regulatory QA Dataset (Multi-Format)
📊 Dataset Summary
A high-quality question-answering dataset derived from public regulatory and policy documents published by Irish and European financial authorities. This dataset addresses a critical challenge in fine-tuning language models: response formatting inconsistency.
Sources:
🇮🇪 Central Bank of Ireland (regulatory documents, policy letters, annual reports)
🇪🇺 ESMA (European Securities and Markets Authority)… See the full description on the dataset page: https://huggingface.co/datasets/Infiniaai/eu-financial-regulatory-qa.adaption-financial-table-qa
This dataset is a remastered version prepared using Adaption's Adaptive Data platform.
adaption-financial_table_qa
This dataset contains question-answer pairs derived from corporate financial statements, including balance sheets and income statements for various companies. The prompts require extracting specific line items, calculating year-over-year percentage changes, or converting scaled values (millions/thousands) to absolute dollar amounts. Each completion provides the… See the full description on the dataset page: https://huggingface.co/datasets/RaffiArdhi/adaption-financial-table-qa.financial-qa-1Kfinancial-qa-10KQA_FinancialNepali_Commercial_Bank_Financial_Indicators_QA_Dataset
Nepali Commercial Bank Financial Indicators — Grounded QA Dataset
A Nepali-language, grounded single-metric question–answering dataset built from the quarterly Key Financial Indicators of Commercial Banks published for Nepal's commercial banking sector. Each example is a single-turn human↔assistant conversation (ShareGPT / Hermes style) in which a question about one specific bank, one specific quarter, and one specific financial metric is answered with the exact value drawn from… See the full description on the dataset page: https://huggingface.co/datasets/sabin1234/Nepali_Commercial_Bank_Financial_Indicators_QA_Dataset.financial-qa-10K-modifiedFinancial_Reasoning_QA_Arabic_Dataset
Financial Reasoning QA Arabic Dataset | مجموعة بيانات الاستدلال المالي باللغة العربية
Dataset Description
This dataset is a professionally translated Arabic version of the original English FinQA dataset. It is designed to facilitate research and development of question-answering and numerical reasoning models for the financial domain in Arabic.
The original FinQA dataset consists of question-answering pairs over financial reports, including both text and tables from S&P… See the full description on the dataset page: https://huggingface.co/datasets/Gheras/Financial_Reasoning_QA_Arabic_Dataset.adaption-financial-math-qa
This dataset is a remastered version prepared using Adaption's Adaptive Data platform.
adaption-financial_math_qa
This dataset contains question-and-answer pairs focused on personal finance calculations, including compound interest, loan amortization, tax brackets, and retirement planning. Each sample provides a specific financial scenario in the prompt and a detailed, step-by-step mathematical derivation in the completion. The responses explain the underlying formulas… See the full description on the dataset page: https://huggingface.co/datasets/uditjain/adaption-financial-math-qa.fpa_financial_qa_trainfinancial-qa-10K-modified2Financial_Reasoning_QA_Arabic_Dataset_V2Kor-financial-qa-7Kvirattt-financial-llama-qa-v2-10Karabic-financial-qa_eval
Arabic Financial Q&A Evaluation Dataset
Validation and test splits for evaluating models on Arabic Financial Q&A with analytical and causal reasoning.
Dataset Structure
Format: Simple prompt-answer pairs
Language: Arabic
Domain: Financial reports analysis
Task: Analytical question answering
Fields
id: Unique identifier
prompt: Full prompt with report and question
question: The analytical question
report: The financial report content
answer:… See the full description on the dataset page: https://huggingface.co/datasets/SahmBenchmark/arabic-financial-qa_eval.slm-bilanco-financial-qa-tr
slm-bilanco-financial-qa-tr
Turkce finansal analiz sorulari ve detayli cevaplarindan olusan instruction-tuning veri seti.
BIST sirketlerinin bilanco verileri uzerinden olusturulmustur.
Onemli Uyarilar
Bu veri seti yatirim tavsiyesi icermez. Analizler yalnizca egitim ve arastirma amaclidir.
Icerideki bilgiler ve bilanco degerleri gercek bilgiler olmayabilir. Yapay zeka tarafindan uretilmis veya sentetik olarak olusturulmus veriler icerebilir.
Herhangi bir yatirim karari almak icin… See the full description on the dataset page: https://huggingface.co/datasets/mrcuren/slm-bilanco-financial-qa-tr.financial-qa-s1decontaminate-v1.0_chutesai-Mistral_inferencearabic-financial-qa_train
Arabic Financial Q&A Training Dataset
Training split of the Arabic Financial Q&A dataset in conversational format.
Dataset Structure
Format: Conversational (human-agent pairs)
Language: Arabic
Domain: Financial reports analysis and causal reasoning
Task: Analytical question answering based on financial documents
Features
id: Unique identifier
conversations: Human prompt (report + question) and agent answer
report_type: Type of financial report… See the full description on the dataset page: https://huggingface.co/datasets/SahmBenchmark/arabic-financial-qa_train.financial-qa-filtered-v1.0-deepseek-tokenized
Dataset Information
Generate Deepseek R1 thinking traces for dataset akftam/financial-qa-s1decontaminate-filtered-v1.0
Verified attempt by rules for numeric answer and gemini-2.0-flash for choice answer
Correct: 6447/8605 (74.92%)
Incorrect: 2158/8605 (25.08%)
This dataset only keeps those correct examples
A "text" column is added by using the script similar to s1 tokenization
Dataset background refers to akftam/financial-qa-s1decontaminate-filtered-v1.0
financial-qa-s1decontaminate-v1.0_Qwen-Qwen2.5_inferencefinancial_QA_optimizedadaption-financial-qa-samples
This dataset is a remastered version prepared using Adaption's Adaptive Data platform.
adaption-financial_qa_samples
This dataset contains prompt-completion pairs focused on financial market analysis, including currency exchange rates, stock trends, and commodity prices. The samples demonstrate handling real-time data limitations, analyzing news headlines for temporal and comparative insights, and providing educational frameworks for investment strategies. Responses range from… See the full description on the dataset page: https://huggingface.co/datasets/sidddd625/adaption-financial-qa-samples.adaption-india-financial-rights-qa
This dataset is a remastered version prepared using Adaption's Adaptive Data platform.
adaption-india_financial_rights_qa
This dataset contains question-answer pairs addressing common financial literacy challenges, fraud scenarios, and regulatory rights specific to India. The content covers topics such as investment scams, banking complaints, loan defaults, insurance claims, and government schemes, providing actionable legal and procedural guidance. Each entry explains relevant… See the full description on the dataset page: https://huggingface.co/datasets/sidddd625/adaption-india-financial-rights-qa.