beaver
Datasets
All datasets matching “beaver”BeaverTails
Dataset Card for BeaverTails
BeaverTails is an AI safety-focused collection comprising a series of datasets.
This repository includes human-labeled data consisting of question-answer (QA) pairs, each identified with their corresponding harm categories.
It should be noted that a single QA pair can be associated with more than one category.
The 14 harm categories are defined as follows:
Animal Abuse: This involves any form of cruelty or harm inflicted on animals, including physical… See the full description on the dataset page: https://huggingface.co/datasets/PKU-Alignment/BeaverTails.PragmaSafe-Beavertails
PragmaVL
This dataset is an augmented version of PKU-Alignment/BeaverTails-V.
Project page: https://sii-fleeecermw.github.io/PragmaVL-iclr26/
Paper: https://arxiv.org/abs/2603.13292
Code: https://github.com/SII-FLEEECERmw/Pragma-VL
Data Organization
The release is organized into 19 categories, and each category contains train and validation splits:
animal_abuse
dangerous_behavior
deception_in_personal_relationships
discriminatory_depictions
environmental_damage… See the full description on the dataset page: https://huggingface.co/datasets/SII-fleeeecer/PragmaSafe-Beavertails.beaver-query
Dataset Card for beaver-query
Homepage and leaderboard |
Github repository |
Paper
Beaver is a holistic framework for evaluating performance on complex, private‑enterprise text‑to‑SQL tasks.
This repository includes questions and corresponding annotations. We reserve a portion of the full question set as a private, hidden test set.
Each sample contains:
id: ID of the question
category: one of real, complex query, domain-specific query, domain-specific complex query.
real indicates… See the full description on the dataset page: https://huggingface.co/datasets/beaverbench/beaver-query.BeaverTails-Evaluation
Dataset Card for BeaverTails-Evaluation
BeaverTails is an AI safety-focused collection comprising a series of datasets.
This repository contains test prompts specifically designed for evaluating language model safety.
It is important to note that although each prompt can be connected to multiple categories, only one category is labeled for each prompt.
The 14 harm categories are defined as follows:
Animal Abuse: This involves any form of cruelty or harm inflicted on animals… See the full description on the dataset page: https://huggingface.co/datasets/PKU-Alignment/BeaverTails-Evaluation.BeaverTails-VWarning: This dataset may contain sensitive or harmful content. Users are advised to handle it with care and ensure that their use complies with relevant ethical guidelines and legal requirements.
1. Usage
If you want to use load_dataset(), you can directly use as follows:
from datasets import load_dataset
train_dataset = load_dataset('PKU-Alignment/BeaverTails-V', name='animal_abuse')['train']
eval_dataset = load_dataset('PKU-Alignment/BeaverTails-V'… See the full description on the dataset page: https://huggingface.co/datasets/PKU-Alignment/BeaverTails-V.beaver-table
Dataset Card for beaver-table
Homepage and leaderboard |
Github repository |
Paper
Beaver is a holistic framework for evaluating performance on complex, private‑enterprise text‑to‑SQL tasks.
This repository includes the full collection of tables. Each table contains:
db: ID of the database the table belongs to
table_name: name of the table in the database
column_names: names of the columns in the table
column_types: data types of the columns in the table
example_rows: example rows… See the full description on the dataset page: https://huggingface.co/datasets/beaverbench/beaver-table.