datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
wiki_dprThis is the wikipedia split used to evaluate the Dense Passage Retrieval (DPR) model.
It contains 21M passages from wikipedia along with their DPR embeddings.
The wikipedia articles were split into multiple, disjoint text blocks of 100 words as passages.flores
Dataset Card for Flores 200
Dataset Summary
⚠️ This repository is no longer being updated ⚠️
A newer version of the FLORES dataset managed by the Open Language Data Initiative
is available at https://huggingface.co/datasets/openlanguagedata/flores_plus.
FLORES is a benchmark dataset for machine translation between English and low-resource languages.
The creation of FLORES-200 doubles the existing language coverage of FLORES-101.
Given the nature of the new… See the full description on the dataset page: https://huggingface.co/datasets/facebook/flores.kilt_tasks
Dataset Card for KILT
Dataset Summary
KILT has been built from 11 datasets representing 5 types of tasks:
Fact-checking
Entity linking
Slot filling
Open domain QA
Dialog generation
All these datasets have been grounded in a single pre-processed Wikipedia dump, allowing for fairer and more consistent evaluation as well as enabling new task setups such as multitask and transfer learning with minimal effort. KILT also provides tools to analyze and understand the… See the full description on the dataset page: https://huggingface.co/datasets/facebook/kilt_tasks.natural_reasoningNaturalReasoning is a large-scale dataset for general reasoning tasks. It consists of high-quality challenging reasoning questions backtranslated from pretraining corpora DCLM and FineMath. The questions have been deduplicated and decontaminated from popular reasoning benchmarks including MATH, GPQA, MMLU-Pro, MMLU-STEM. For each question, we extract the reference final answer from the original document from the pretraining corpora if possible. We also provide a model-generated response from… See the full description on the dataset page: https://huggingface.co/datasets/facebook/natural_reasoning.cyberseceval3-visual-prompt-injection
Dataset Card for CyberSecEval 3 - Visual Prompt Injection Benchmark
Dataset Details
Dataset Description
This dataset provides a multimodal benchmark for visual prompt injection, with text/image inputs. It is part of CyberSecEval 3, the third edition of Meta's flagship suite of security benchmarks for LLMs to measure cybersecurity risks and capabilities across multiple domains.
Language(s): English
License: MIT
Dataset Sources
Repository: Link… See the full description on the dataset page: https://huggingface.co/datasets/facebook/cyberseceval3-visual-prompt-injection.FACET-Terminal-Tasks-6k
FACET-Terminal-Tasks-6k
6,020 execution-grounded tasks for terminal agents, coding agents, and executable workflow research
🌐 FACET Project Website
📄 FACET Paper
💻 FACET-Terminal GitHub Repository
🤗 FACET-Terminal Models & Data
Dataset Overview
FACET-Terminal-Tasks-6k contains 6,020 public-release-ready Harbor tasks produced by FACET. Each task is an executable environment rather than a standalone prompt: it includes a natural-language instruction… See the full description on the dataset page: https://huggingface.co/datasets/FACET-Terminal/FACET-Terminal-Tasks-6k.multiloko
MultiLoKo: a multilingual local knowledge benchmark for LLMs
MultiLoKo is a multilingual knowledge benchmark, covering 30 languages plus English.
The questions are separately sourced for each language, with an annotation protocol designed to target locally relevant topics for the respective language.
MultiLoKo contains the original data for each language, as well as both human and machine-authored translations of each non-English subset into English and vice versa, facilitating… See the full description on the dataset page: https://huggingface.co/datasets/facebook/multiloko.BigOBench
👋 Overview
🚀 Introduction
📋 Getting Started with the data
🔥 problem_and_human_solutions_list.jsonl
🔥 complexity_labels_light.jsonl
🔥 complexity_labels_full.jsonl
🔥 time_complexity_test_set.jsonl
🔥 space_complexity_test_set.jsonl
License
📝 Citation
🚀 Introduction
BigO(Bench) is a benchmark of ~300 code problems to be solved in Python, along with 3,105 coding problems… See the full description on the dataset page: https://huggingface.co/datasets/facebook/BigOBench.mddThe Movie Dialog dataset (MDD) is designed to measure how well
models can perform at goal and non-goal orientated dialog
centered around the topic of movies (question answering,
recommendation and discussion).curiosity_dialogsThis dataset contains 14K dialogs (181K utterances) where users and assistants converse about geographic topics like
geopolitical entities and locations. This dataset is annotated with pre-existing user knowledge, message-level dialog
acts, grounding to Wikipedia, and user reactions to messages.FACTORY
Overview
FACTORY is a large-scale, human-verified, and challenging prompt set. We employ a model-in-the-loop approach to ensure quality and address the complexities of evaluating long-form generation. Starting with seed topics from Wikipedia, we expand each topic into a diverse set of prompts using large language models (LLMs). We then apply the model-in-the-loop method to filter out simpler prompts, maintaining a high level of difficulty. Human annotators further refine the prompts… See the full description on the dataset page: https://huggingface.co/datasets/facebook/FACTORY.linguiniDataset of the paper Linguini 🍝: A benchmark for language-agnostic linguistic reasoning
Code: https://github.com/facebookresearch/linguini
FaceInstruct1M
FaceLLaVA: Facial Expression and Attribute Understanding through Instruction Tuning
WACV 2026
This is the official dataset for the WACV 2026 Round 1 Early Accept paper (6.4% acceptance rate) - Face-LLaVA: Facial Expression and Attribute Understanding through Instruction Tuning.
Refer to our GitHub repository for more information on the method and codebase.
📦 Data Organization… See the full description on the dataset page: https://huggingface.co/datasets/chaubeyG/FaceInstruct1M.llamafirewall-alignmentcheck-evals
Dataset Card for LlamaFirewall AlignmentCheck Evals
Dataset Details
Dataset Description
This dataset provides a dataset for prompt injection in an agentic environment. It is part of LlamaFirewall, an open-source security focused guardrail framework designed to serve as a final layer of defense against security risks associated with AI Agents. Specifically, this dataset is designed to evaluate the susceptibility of language models, and detect any misalignment… See the full description on the dataset page: https://huggingface.co/datasets/facebook/llamafirewall-alignmentcheck-evals.airs-bench
AIRS-Bench: a Suite of Tasks for Frontier AI Research Science Agents
The AI Research Science Benchmark (AIRS-Bench) quantifies the autonomous research abilities of LLM agents in the area of machine learning. AIRS-Bench comprises 20 tasks from state-of-the-art machine learning papers spanning diverse domains: NLP, Code, Math, biochemical modelling, and time series forecasting.
Each task is specified by a ⟨problem, dataset, metric⟩ triplet and a SOTA value. The agent receives the… See the full description on the dataset page: https://huggingface.co/datasets/facebook/airs-bench.HoneyBee
HoneyBee: Data Recipes for Vision-Language Reasoners
This is the official data release for the paper: https://arxiv.org/abs/2510.12225.
Github Repo: https://github.com/facebookresearch/HoneyBee_VLM.
Abstract
Recent advances in vision-language models (VLMs) have made them highly effective at reasoning tasks. However, the principles underlying the construction of performant VL reasoning training datasets remain poorly understood. In this work, we introduce several data… See the full description on the dataset page: https://huggingface.co/datasets/facebook/HoneyBee.face-eval
Agentic Sycophancy Evaluation Dataset
Evaluation dataset for studying sycophancy as Chain-of-Thought (CoT) unfaithfulness
in LLMs under both tool-channel and user-channel cue delivery, across five
artifact sources.
Overview
Total rows: 5100
Axes: 5 (domain-expertise, egalitarianism, epistemic-posture, ethics, political)
Sources: 5 (browser_history, email, notes, profile, slack)
Designed around a 2×2 grid: role (tool / user) × register (summary / raw artifact).… See the full description on the dataset page: https://huggingface.co/datasets/edinburgh-dawg/face-eval.facebook_darija_dataset
Darija Facebook Posts Dataset
Dataset Details
Dataset Description
This dataset consists of more than 5k public posts from Facebook. Each post contains text content, metadata.
This dataset containt more than 400K darija tokens.
Curated by: @abdeljalilELmajjodi
Language(s) (NLP): Multiple (primarily Moroccon Arabic)
Uses
This dataset could be used for:
Training and testing language models on social media content
Analyzing social media posting… See the full description on the dataset page: https://huggingface.co/datasets/abdeljalilELmajjodi/facebook_darija_dataset.FaceInstruct1M
FaceLLaVA: Facial Expression and Attribute Understanding through Instruction Tuning
WACV 2026
This is the official dataset for the WACV 2026 Round 1 Early Accept paper (6.4% acceptance rate) - Face-LLaVA: Facial Expression and Attribute Understanding through Instruction Tuning.
Refer to our GitHub repository for more information on the method and codebase.
📦 Data Organization… See the full description on the dataset page: https://huggingface.co/datasets/zyq888/FaceInstruct1M.seshat-perspectiveSeshat-perspective is the first historical databank synthetically annotated with a perspectivist approach by means of multiple Large Language Models: Deepseek (dr1), Llama (l31l) and Mistral (m3m)
Paper with field description and validation procedure: https://github.com/facells/fabio-celli-publications/blob/main/docs/2026_perspective_seshat_clicit26.pdf
Code for replication: https://colab.research.google.com/drive/1_4aUNGjl7_uhLZZKE7mAYHPhWYUZ9jvr?usp=sharing
Amazigh-Facebook-Data-Export
Dataset Card for Amazigh Facebook Dataset (Filtered)
This dataset contains a curated, filtered collection of personal Facebook posts and comments written by an Amazigh speaker. It captures real-world communication usage of Amazigh in the Amazigh script (Tifinagh). The speaker utilized Southern Moroccan Amazigh (Tashelhit) and leaned to the extent possible toward Standard Moroccan Amazigh. Some posts and comments uniquely contain parallel multi-script and multi-language segments… See the full description on the dataset page: https://huggingface.co/datasets/abdelhaqueidali/Amazigh-Facebook-Data-Export.SCRuB-dataset
SCRuB — Social Concept Reasoning under Rubric-Based Evaluation
SCRuB is a dataset suite for studying how large language models handle socially sensitive, open-ended essay prompts. It comprises three components:
Component
Description
Rows
SCRuBSample
30 curated study prompts used as stimuli in a human annotation study
30
SCRuBAnnotations
Expert essays, model responses, and quality judgments from a two-task annotation study
300 + 78 + 20 + 900 + 900
SCRuBEval4,711… See the full description on the dataset page: https://huggingface.co/datasets/facebook/SCRuB-dataset.