Saved in:
Bibliographic Details
Main Authors: Gulati, Aryan, Dong, Xingjian, Hurtado, Carlos, Shekkizhar, Sarath, Swayamdipta, Swabha, Ortega, Antonio
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2407.13141
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910533109678080
author Gulati, Aryan
Dong, Xingjian
Hurtado, Carlos
Shekkizhar, Sarath
Swayamdipta, Swabha
Ortega, Antonio
author_facet Gulati, Aryan
Dong, Xingjian
Hurtado, Carlos
Shekkizhar, Sarath
Swayamdipta, Swabha
Ortega, Antonio
contents As language models become more general purpose, increased attention needs to be paid to detecting out-of-distribution (OOD) instances, i.e., those not belonging to any of the distributions seen during training. Existing methods for detecting OOD data are computationally complex and storage-intensive. We propose a novel soft clustering approach for OOD detection based on non-negative kernel regression. Our approach greatly reduces computational and space complexities (up to 11x improvement in inference time and 87% reduction in storage requirements) and outperforms existing approaches by up to 4 AUROC points on four different benchmarks. We also introduce an entropy-constrained version of our algorithm, which leads to further reductions in storage requirements (up to 97% lower than comparable approaches) while retaining competitive performance. Our soft clustering approach for OOD detection highlights its potential for detecting tail-end phenomena in extreme-scale data settings.
format Preprint
id arxiv_https___arxiv_org_abs_2407_13141
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Out-of-Distribution Detection through Soft Clustering with Non-Negative Kernel Regression
Gulati, Aryan
Dong, Xingjian
Hurtado, Carlos
Shekkizhar, Sarath
Swayamdipta, Swabha
Ortega, Antonio
Machine Learning
As language models become more general purpose, increased attention needs to be paid to detecting out-of-distribution (OOD) instances, i.e., those not belonging to any of the distributions seen during training. Existing methods for detecting OOD data are computationally complex and storage-intensive. We propose a novel soft clustering approach for OOD detection based on non-negative kernel regression. Our approach greatly reduces computational and space complexities (up to 11x improvement in inference time and 87% reduction in storage requirements) and outperforms existing approaches by up to 4 AUROC points on four different benchmarks. We also introduce an entropy-constrained version of our algorithm, which leads to further reductions in storage requirements (up to 97% lower than comparable approaches) while retaining competitive performance. Our soft clustering approach for OOD detection highlights its potential for detecting tail-end phenomena in extreme-scale data settings.
title Out-of-Distribution Detection through Soft Clustering with Non-Negative Kernel Regression
topic Machine Learning
url https://arxiv.org/abs/2407.13141