datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
spider
Spider Unified dataset
Documentation comming soon
spiqa
SPIQA Dataset Card
Dataset Details
Dataset Name: SPIQA (Scientific Paper Image Question Answering)
Paper: SPIQA: A Dataset for Multimodal Question Answering on Scientific Papers
Github: SPIQA eval and metrics code repo
Dataset Summary: SPIQA is a large-scale and challenging QA dataset focused on figures, tables, and text paragraphs from scientific research papers in various computer science domains. The figures cover a wide variety of plots… See the full description on the dataset page: https://huggingface.co/datasets/google/spiqa.SpinBench
SpinBench: Perspective and Rotation as a Lens on Spatial Reasoning in VLMs
🌐 Project page |
🤗 Dataset |
📑 Paper |
💻 Code
SpinBench is a cognitively grounded diagnostic benchmark for evaluating spatial reasoning in vision-language models (VLMs).
SpinBench is designed around the core challenge of spatial reasoning: perspective taking, the ability to reason about how scenes and object relations change under viewpoint transformation. Since perspective taking requires… See the full description on the dataset page: https://huggingface.co/datasets/YuyouZhang/SpinBench.mascarade-spice-dataset
Ailiance — SPICE & Analog Simulation Q&A
🇫🇷 Ailiance — curated by Ailiance for production deployment ; co-published with the upstream electron-rare/mascarade-spice-dataset. 🇪🇺 Compatible EU AI Act (Template AI Office, July 2025).
Q&A bilingue (FR/EN) sur la simulation SPICE et l'analyse de circuits analogiques : ngspice, LTspice, modèles MOSFET/BJT, topologies analogiques, ampli-op, filtres actifs, sources de courant, polarisation.
Statistics
Métrique… See the full description on the dataset page: https://huggingface.co/datasets/Ailiance-fr/mascarade-spice-dataset.MedPriv-Bench_dataset
MedPriv-Bench
MedPriv-Bench evaluates the privacy--utility trade-off of language models in
medical open-ended question answering. Each example contains a synthetic
patient context, injected privacy-sensitive facts, a question, and a
ground-truth answer. The benchmark supports evaluating whether a model remains
clinically useful while avoiding disclosure of the injected facts.
Split sizes
Split
Rows
Composition
train
2,200
1,315 benchmark-construction… See the full description on the dataset page: https://huggingface.co/datasets/Spiderman01/MedPriv-Bench_dataset.SPIQA_50K_Re
SPIQA 50K Re-annotated
QA annotations on scientific paper figures from the SPIQA dataset.
Dataset Structure
Each sample contains:
Field
Description
image
Relative path to the figure image
question
Question about the figure
thinking
Chain-of-thought reasoning
answer
Final answer
Splits
train (spiqa_50k.json): 50,000 samples
train_reannotate (spiqa_50k_reannotate.json): 49,975 samples with more detailed chain-of-thought reasoning… See the full description on the dataset page: https://huggingface.co/datasets/scimdr/SPIQA_50K_Re.mascarade-spice-dataset
Mascarade — SPICE & Analog Simulation Q&A
Description
Q&A bilingue (FR/EN) sur la simulation SPICE et l'analyse de circuits analogiques : ngspice, LTspice, modèles MOSFET/BJT, topologies analogiques, ampli-op, filtres actifs, sources de courant, polarisation.
Ce dataset fait partie de la famille Mascarade, un corpus thématique destiné au fine-tuning LoRA de modèles compacts (cible : Gemma-3n-E4B et équivalents) pour des assistants spécialisés en électronique embarquée.… See the full description on the dataset page: https://huggingface.co/datasets/electron-rare/mascarade-spice-dataset.spider-text2sql-bench
Dataset Card for spider-text2sql-bench
spider-text2sql-bench 是 Spider 1.0 官方訓練集之 OpenAI Messages 格式版本，共 7,000 筆，將原始之 question / schema / sql 重新組裝為 system / user / assistant 三 role 之對話結構。除原生之 messages 欄位外，另拆解出獨立之 system / user / assistant 字串欄位，可作為 Text-to-SQL 模型之 SFT 訓練語料，亦可直接用於 benchmark evaluation pipeline（以 user 作為 prompt，比對模型輸出與 assistant 之標準答案 SQL）。
Dataset Details
Dataset Description
Spider 1.0 為 Yale LILY Group 於 EMNLP 2018 發表之大規模跨領域 Text-to-SQL… See the full description on the dataset page: https://huggingface.co/datasets/lianghsun/spider-text2sql-bench.qwen_qa_pairs_cli_training.jsonl
Data sources
Multiple datasets from Hugging Face related to natural language to CLI pairs were gathered.
Human reviewed synthetic data from Claude Opus4.6 and ChatGPT4.5 were added.
A handful of grounding rows related to the organisation "Spicy Lemonade" were added (see details below)
Data processing
As part of the processing, data was converted to the Alpaca format with instruction (natural language), input (typically blank) and output (the CLI command) columns.
The… See the full description on the dataset page: https://huggingface.co/datasets/spicy-lemonade/qwen_qa_pairs_cli_training.jsonl.spinoza-treatise-emendation-intellect-100-qa
Description
This dataset contains 100 synthetic question-answer pairs based on Spinoza's posthumously
published Tractatus de Intellectus Emendatione (1677) as translated by R. H. L. Elwes'
On the Improvement of the Understanding (Treatise on the Emendation of the Intellect (1883).
The questions and answers represent a comprehensive overview of the ideas and principles set
forth by Spinoza in the treatise.
The question-answer pairs were generated, reviewed, and refined in iterative… See the full description on the dataset page: https://huggingface.co/datasets/joshause/spinoza-treatise-emendation-intellect-100-qa.cyberstrike-sft-120k
CyberStrike SFT 120K
The largest open-source offensive cybersecurity SFT dataset
121,422 expert-level red team instruction-response pairs across 15 security generators
Quick Start •
Why CyberStrike •
Domains •
Data Format •
Training Guide •
Benchmarks •
Contributing •
License
Why CyberStrike?
Most LLMs refuse or give surface-level answers to offensive security questions. Security professionals —… See the full description on the dataset page: https://huggingface.co/datasets/spinochenza/cyberstrike-sft-120k.spider_th
Spider Thai Dataset
Thai translation of the official Spider benchmark (A Large-Scale Human-Labeled Dataset for Complex and Cross-Domain Semantic Parsing and Text-to-SQL Task).
Dataset Description
This dataset contains Thai translations of the Spider text-to-SQL benchmark, translated from the official Spider data source.
Source
Original Dataset: Spider Benchmark
Paper: Spider: A Large-Scale Human-Labeled Dataset for Complex and Cross-Domain Semantic Parsing and… See the full description on the dataset page: https://huggingface.co/datasets/Porameht/spider_th.gemma_qa_pairs_cli_training.jsonl
Data sources
Multiple datasets from Hugging Face related to natural language to CLI pairs were gathered.
Human reviewed synthetic data from Claude Opus4.6 and ChatGPT4.5 were added.
A handful of grounding rows related to the organisation "Spicy Lemonade" were added (see details below)
Data processing
As part of the processing, data was converted to the Alpaca format with instruction (natural language), input (typically blank) and output (the CLI command) columns.
The… See the full description on the dataset page: https://huggingface.co/datasets/spicy-lemonade/gemma_qa_pairs_cli_training.jsonl.spiral-bench
SPIRAL-Bench v0
A small benchmark for testing whether the wording of a retrieval query changes the
balance of the evidence a retriever returns.
Built for the Ouroboros project, which studies self-confirming retrieval loops in agentic RAG.
The question this dataset exists to answer
In agentic RAG, the system writes its own follow-up search queries, and it writes them using
what it currently believes. A retriever ranks documents by how closely they match the wording… See the full description on the dataset page: https://huggingface.co/datasets/vigneshwar234/spiral-bench.spiqa-chat-format
SPIQA (Chat-Format Preparation)
This dataset is a chat-format preparation of SPIQA for multimodal supervised fine-tuning (SFT).
Format
This format is commonly referred to as:
chat-format SFT data
instruction-tuning conversations
OpenAI-style messages format
Included files
train.jsonl
validation.jsonl
stats.json
prepare_spiqa_unsloth.py
Source
Base dataset: google/spiqa
Original Dataset Highlights
Original dataset: google/spiqa… See the full description on the dataset page: https://huggingface.co/datasets/ali5341/spiqa-chat-format.GreenBond-Spillover-Instruct
GreenBond-Spillover-Instruct
Specialized instruction-tuning dataset for sovereign green bond analysis, spillover-effect detection, and narrative risk assessment.
Dataset Info
Property
Value
Total examples
3,000
Format
JSONL
Fields
instruction, input, output
Language
English
Category Distribution
Category
Count
Sentiment classification
752
Greenwashing detection
593
Narrative tagging
467
Spillover Q&A
331
Event extraction… See the full description on the dataset page: https://huggingface.co/datasets/Nurlykhan/GreenBond-Spillover-Instruct.