Out-of-Distribution Detection using Synthetic Data Generation

Momin Abbas, Muneeza Azmat, Raya Horesh, Mikhail Yurochkin

Keywords: out-of-distribution detection, out-of-distribution generalization, synthetic data

TL;DR: This work presents an effective OOD detection method using LLM-generated synthetic proxies, eliminating the need for external OOD data. Experiments show it reduces false positives and outperforms baseline methods in text classification tasks.

Abstract: Distinguishing in- and out-of-distribution (OOD) inputs is crucial for reliable deployment of classification systems. However, OOD data is typically unavailable or difficult to collect, posing a significant challenge for accurate OOD detection. In this work, we present a method that harnesses the generative capabilities of Large Language Models (LLMs) to create high-quality synthetic OOD proxies, eliminating the dependency on any external OOD data source. We study the efficacy of our method on classical text classification tasks such as toxicity detection and sentiment classification as well as classification tasks arising in LLM development and deployment, such as training a reward model for RLHF and detecting misaligned generations. Extensive experiments on nine InD-OOD dataset pairs and various model sizes show that our approach dramatically lowers false positive rates (achieving a perfect zero in some cases) while maintaining high accuracy on in-distribution tasks, outperforming baseline methods by a significant margin.

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the COLM Code of Ethics on https://colmweb.org/CoE.html

Author Guide: I certify that this submission complies with the submission instructions as described on https://colmweb.org/AuthorGuide.html

Flagged For Ethics Review: true

Ethics Comments: I'm not sure if requires ethic revision due to the content in the used datasets.

Submission Number: 270