MetaCLIP
MetaCLIP (Metadata-Curated Language-Image Pre-training) is a data curation recipe and a family of vision-language models from Meta AI, introduced in the 2023 paper "Demystifying CLIP Data" by Hu Xu, Saining…
Data & Datasets
Explore Data & Datasets through related topics and the articles other pages reference most.
Explore articles
Browse subtopics (38)
Articles that also belong to these categories. Counts cover all of Data & Datasets.
All articles (116)Machine Learning (65)Natural Language Processing (15)Open Source AI (15)Computer Vision (13)AI Benchmarks (11)Large Language Models (11)Statistics (10)AI Ethics (6)Meta AI (6)Robotics (5)Embodied AI (4)AI Companies (3)Model Evaluation (3)NVIDIA (3)AI Research (2)Data Science (2)Deep Learning (2)MLOps (2)Multimodal AI (2)Training & Optimization (2)AI Code Generation (1)AI History (1)AI Infrastructure (1)AI Models (1)AI Policy & Regulation (1)AI Safety (1)AI Tools & Products (1)AI for Science (1)Algorithms (1)Artificial Intelligence (1)Chinese AI (1)Developer Tools (1)Enterprise AI (1)Generative AI (1)Google (1)Google DeepMind (1)Mathematics (1)Robotics Companies (1)
Showing 61-116 of 116 articles
MimicGen
MimicGen is a data generation system developed by researchers at NVIDIA's Seattle Robotics Lab and Learning and Perception Research group that automatically produces large-scale robot manipulation datasets…
Nemotron-CC
Nemotron-CC is a large-scale, open English-language pretraining dataset for large language models released by NVIDIA in December 2024.
Noise
Noise in machine learning is any unwanted, irrelevant, or random variation in data that obscures the true underlying patterns a model is trying to learn.
Non-Response Bias
Non-response bias is the error that arises when the people or units that do not respond to a survey, study, or data collection process differ systematically from those that do
Normalization
Normalization is the process of scaling numerical data to a standard range or distribution so that features and activations are comparable and downstream computation behaves predictably.
Numerical Data
Numerical data (also called quantitative data) is information expressed as numbers on a continuous or discrete scale that supports arithmetic operations such as addition, subtraction, multiplication, and…
NuminaMath
NuminaMath is a family of openly licensed competition-mathematics resources developed by the non-profit Project Numina, spanning the largest public dataset of competition math problems and solutions, a set of…
One-Hot Encoding
One-hot encoding is a data preprocessing technique that converts a categorical variable with $$k$$ distinct categories into $$k$$ binary columns, where each category is represented by a vector that contains…
Open Catalyst Project
The Open Catalyst Project (OCP) is a research collaboration between Meta AI's Fundamental AI Research group (FAIR) and Carnegie Mellon University's Department of Chemical Engineering.
Open Images
Open Images is a large annotated image dataset published by Google for computer vision research.
Open X-Embodiment
Open X-Embodiment (OXE) is a large-scale collaborative robotics research initiative, led by Google DeepMind and announced in October 2023, that produced the largest open-source real robot dataset and a family…
OpenOrca
OpenOrca is a large open-source instruction-tuning dataset that augments the FLAN Collection with chain-of-thought responses generated by OpenAI's GPT-3.5 and GPT-4 APIs.
OpenThoughts
OpenThoughts is an open-source initiative and a series of datasets of verified reasoning traces created to train open reasoning models.
Outlier Detection
Outlier detection is the process of identifying data points, observations, or patterns that deviate so markedly from the rest of a dataset that they are likely to have been generated by a different process.
Oversampling
Oversampling is a data preprocessing technique in machine learning that fixes class imbalance by increasing the number of minority class examples in the training set, either by duplicating existing samples…
PASCAL VOC
PASCAL VOC (Pattern Analysis, Statistical Modelling and Computational Learning Visual Object Classes) is a long-running benchmark dataset and annual challenge for object recognition, object detection…
Participation Bias
Participation bias is a systematic error that arises when the individuals who choose to take part in a study, survey, or data collection effort differ in meaningful ways from those who do not, so the resulting…
Preprocessing
Preprocessing is the stage of a machine learning workflow that transforms raw data into a clean, consistent, numerical format that learning algorithms can use.
Project Aria
Project Aria is an egocentric data-collection research program run by Meta's Reality Labs Research. It was announced on September 16, 2020 .
Proxy labels
A proxy label (also called a surrogate label) is an observable, easy-to-collect stand-in for the true target a model should predict, used to train that model when the real target is too expensive, too slow, or…
Quantile bucketing
Quantile bucketing, also called quantile binning, equal-frequency binning, or quantile discretization
Rater
A rater is a person (or, increasingly, a model) who assigns labels, scores, or judgments to data items so those items can be used to train, evaluate, or align a machine learning system.
RedPajama
RedPajama is a family of large-scale, openly licensed datasets for training large language models (LLMs), created by Together AI with academic and open-source partners to reproduce, in fully open form
RefinedWeb
RefinedWeb is a large-scale English pretraining dataset for large language models, built from filtered and deduplicated Common Crawl web data alone and released in June 2023 by the Technology Innovation…
Reporting Bias
Reporting bias is a type of data bias in machine learning that occurs when the frequency of events, properties, or outcomes captured in a dataset does not reflect their real-world frequency, because people…
RoboCasa
RoboCasa is a large-scale simulation framework for training generalist robots to perform everyday household tasks.
Sampling Bias
Sampling bias is a systematic error in statistics and machine learning that occurs when a sample is collected so that some members of the intended population have a higher or lower probability of being…
Scaling
In machine learning and artificial intelligence, scaling is an umbrella term for several distinct practices: (1) feature scaling, the data-preprocessing step that rescales input features to a common range so…
Segment Anything Model and Dataset (SAM and SA-1B)
Segment Anything Model (SAM) is a promptable image segmentation foundation model released by Meta AI on April 5, 2023 that lets users "cut out" any object in an image with a single click, box, or mask prompt…
Selection Bias
Selection bias is a systematic error that occurs when the data used for analysis, training, or evaluation does not accurately represent the population or domain it is intended to describe
Self-Instruct
Self-Instruct is a semi-automated framework for aligning a pretrained large language model with instruction-following behavior by bootstrapping its training data from the model itself, introduced in December…
SlimPajama
SlimPajama is a 627-billion-token English-language pre-training corpus for large language models, produced by Cerebras Systems in collaboration with the Opentensor Foundation by extensively cleaning and…
Snorkel
Snorkel is an open-source software framework and methodology for programmatic data labeling that started at Stanford University in 2015 and is licensed under Apache 2.0 .
Sparse Feature
A sparse feature is a feature in machine learning whose values are predominantly zero or empty across a dataset.
Sparse Representation
Sparse representation is a data encoding strategy in which most values in a vector, matrix, or tensor are zero (or near zero), with only a small fraction of elements carrying non-zero values.
Specific Labs
Specific Labs is a San Francisco company that buys and licenses operational data and source code from businesses, and packages it as training and evaluation material for AI labs.
SuperGLUE
SuperGLUE is a public benchmark for evaluating general-purpose English language understanding in machine learning systems, consisting of eight challenging reasoning and reading-comprehension tasks, two…
Surge AI
Surge AI (legal entity Surge Labs Inc., often stylized SurgeHQ) is an American data annotation and human evaluation company headquartered in San Francisco, California that supplies frontier artificial…
Synthetic Feature
A synthetic feature (also called a constructed feature or derived feature) is a variable created by transforming, combining, or otherwise manipulating one or more existing features in a dataset, rather than…
Synthetic data
Synthetic data is data that is artificially generated by algorithms, simulations, or generative models rather than collected from real-world events, used as a substitute for or supplement to real data when…
Tf.Example
tf.train.Example (commonly written as tf.Example) is a Protocol Buffers message type that TensorFlow uses as its standard on-disk representation for a single training or inference example.
The Pile (dataset)
The Pile is an 825.18 GiB (approximately 886 GB) English text corpus designed for training large language models, assembled from 22 diverse, high-quality subsets spanning academic, professional, internet…
The Stack (BigCode dataset)
The Stack is a family of large, permissively-licensed source-code datasets built by the BigCode project, an open scientific collaboration jointly led by Hugging Face and ServiceNow Research, to train and…
The Stack v2
The Stack v2 is a large open dataset of source code released by BigCode in February 2024 as the training dataset behind the StarCoder2 family of code models.
Training Set
A training set is the portion of data used to fit a statistical or machine learning model.
Training-Serving Skew
Training-serving skew is a difference between a machine learning model's performance during training and its performance during serving (production inference).
TxT360
TxT360 is an open large-scale pretraining corpus for large language models, released in October 2024 by the LLM360 project, a collaboration led by Petuum and the Mohamed bin Zayed University of Artificial…
UltraChat
UltraChat is a large-scale synthetic multi-turn instructional conversation dataset released in May 2023 by the OpenBMB group at Tsinghua University, comprising approximately 1.5 million dialogues generated by…
Visual question answering
Visual question answering (VQA) is the task of producing a natural-language answer to a natural-language question about an image.
WRAP (Web Rephrase Augmented Pre-training)
WRAP (Web Rephrase Augmented Pre-training) is a synthetic-data pre-training method introduced in the paper "Rephrasing the Web: A Recipe for Compute and Data-Efficient Language Modeling," posted to arXiv on 29…
Weak supervision
Weak supervision is a machine learning paradigm in which models are trained from noisy, limited, imprecise, or programmatically generated labels rather than from large, expensively hand annotated datasets.
Web scraping
Web scraping is the automated extraction of data from websites, performed by programs that request pages the way a browser does and then parse the returned HTML, or the APIs behind it, into structured records .
WildChat
WildChat is a large public corpus of real conversations between human users and ChatGPT, released by researchers at the Allen Institute for AI (AI2) and Cornell University.
WordNet
WordNet is a large lexical database of English developed at Princeton University beginning in 1985.
Z-Score Normalization
Z-score normalization, also called standardization, standard score normalization, or z-score scaling, is a data preprocessing technique that transforms a numerical feature so that it has a mean of 0 and a…