Perplexity Correlations

community

https://github.com/TristanThrush/perplexity-correlations

AI & ML interests

Simple Statistical Tools for Pretraining Data Selection

perplexity-correlations's activity

Tristan

updated 8 models 4 months ago

perplexity-correlations/fasttext-lambada-es-target

Updated Sep 23, 2024

perplexity-correlations/fasttext-lambada-de-target

Updated Sep 23, 2024

perplexity-correlations/fasttext-lambada-fr-target

Updated Sep 23, 2024

perplexity-correlations/fasttext-lambada-it-target

Updated Sep 23, 2024

perplexity-correlations/fasttext-lambada-target

Updated Sep 23, 2024

perplexity-correlations/fasttext-sciq-target

Updated Sep 23, 2024

perplexity-correlations/fasttext-piqa-target

Updated Sep 23, 2024 • 1

perplexity-correlations/fasttext-arc-easy-target

Updated Sep 23, 2024

Tristan

authored 2 papers about 1 year ago

I am a Strange Dataset: Metalinguistic Tests for Language Models

Paper • 2401.05300 • Published Jan 10, 2024 • 4

Winoground: Probing Vision and Language Models for Visio-Linguistic Compositionality

Paper • 2204.03162 • Published Apr 7, 2022 • 1

Tristan

authored 3 papers over 1 year ago

Towards Language Models That Can See: Computer Vision Through the LENS of Natural Language

Paper • 2306.16410 • Published Jun 28, 2023 • 28

Models in the Loop: Aiding Crowdworkers with Generative Annotation Assistants

Paper • 2112.09062 • Published Dec 16, 2021

Dynatask: A Framework for Creating Dynamic AI Benchmark Tasks

Paper • 2204.01906 • Published Apr 5, 2022

Tristan

authored 3 papers almost 2 years ago

Measuring Data

Paper • 2212.05129 • Published Dec 9, 2022 • 1

The BigScience ROOTS Corpus: A 1.6TB Composite Multilingual Dataset

Paper • 2303.03915 • Published Mar 7, 2023 • 6

BLOOM: A 176B-Parameter Open-Access Multilingual Language Model

Paper • 2211.05100 • Published Nov 9, 2022 • 28