datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
liars-dividend-GPT-image-2-photorealliars-dividend-flux2-images
Liars Dividend FLUX.2 Images
100 prompt-matched synthetic images generated with black-forest-labs/FLUX.2-dev. Images are stored under FLUX.2/fake/.
Generation settings: seed 42, 1024x1024, 50 inference steps, guidance scale 4.0.
schd-dividend-payments-and-holdings
SCHD Dividend Payments and Dated Holdings
Two small, source-attributed historical tables for learning reproducible dividend and portfolio-exposure calculations.
The worked explanations accompany DividendSteps.
Contents
Configuration
Records
Observation scope
payments
40
Ex-dividend dates September 19, 2016 through June 24, 2026; original source-check date September 9, 2026.
holdings
102
September 8, 2026 disclosure, including cash, money-market and… See the full description on the dataset page: https://huggingface.co/datasets/Holaclea/schd-dividend-payments-and-holdings.unpredictable_dividend-comThe UnpredicTable dataset consists of web tables formatted as few-shot tasks for fine-tuning language models to improve their few-shot performance. For more details please see the accompanying dataset card.daily-paper-2026-09-23-validator-dividend-tool-call-consensus
The Validator's Dividend: Measuring the Cost-Quality Frontier of Validation-Gated Resampling versus k-Sample Consensus for Agentic Tool-Call Output on Self-Hosted H200
TL;DR — On a self-hosted quantized model, a free deterministic validator is the cheapest tool-call reliability lever: validate-then-retry recovers the detectable error fraction at marginal cost c/q per quality point, is bounded by a silent floor that no retry budget can cross, and beats both k-sample consensus… See the full description on the dataset page: https://huggingface.co/datasets/thaki-AI/daily-paper-2026-09-23-validator-dividend-tool-call-consensus.NEPSE_Dividend_FAQ
📈 NEPSE Dividend FAQ (Nepali) — Dataset README
"How much dividend did it give this time?" — the question every stock market watcher asks, now turned into 2,000 question-answer pairs.
A simple, clean, 100% Nepali-language financial FAQ dataset that builds question-answer (Q&A) pairs about the total dividend and bonus share dividend of companies listed on NEPSE. This file is a full dissection of that exact dataset — every field, every pattern, everything.
🔖 TL;DR… See the full description on the dataset page: https://huggingface.co/datasets/sabin1234/NEPSE_Dividend_FAQ.stocks-dividendsdividend-common-question
Dividend Common Question Dataset
📖 개요
이 데이터셋은 배당주 투자 관련 자주 묻는 질문과 답변을 Alpaca 포맷(instruction, input, output)으로 구성한 학습용 자료입니다.총 10여 개의 샘플이 포함되어 있으며, 파인튜닝 실습이나 자연어 처리 모델 학습에 활용할 수 있습니다.
📂 데이터 구조
데이터는 CSV 파일(dividend-common-question.csv)로 제공되며, 다음과 같은 열을 포함합니다:
instruction: 모델에게 주는 지시문 (예: "배당성향이 높아야 좋은 건가요?")
input: 지시문을 수행하는 데 필요한 추가 입력 (없으면 빈칸)
output: 모델이 생성해야 하는 답변 (예: "꼭 그렇다고 할 수는 없습니다...")
예시
instruction,input,output
"시가배당수익률이 높으면 좋은… See the full description on the dataset page: https://huggingface.co/datasets/ycryu/dividend-common-question.stock_dividendsbist-dividend-yield-projections-2026NEPSE_Dividend_FAQ_Dataset_Romanized_Nepali_Questions
NEPSE Dividend FAQ Dataset (Romanized Nepali Questions)
File: nepse_dividend_faq_romanized.jsonl
Total records: 2,000
Format: JSON Lines (.jsonl) — one JSON object per line
Language: Nepali (ne / ISO 639-3 npi), answers in Devanagari script (Deva); questions in romanized Nepali (Latin letters)
Domain: Financial services — NEPSE (Nepal Stock Exchange) historical dividend records
Task type: Instruction-following (instruction-following)
Generation type: Real (real) — this is… See the full description on the dataset page: https://huggingface.co/datasets/sabin1234/NEPSE_Dividend_FAQ_Dataset_Romanized_Nepali_Questions.