# Contrastive Out-of-Distribution Detection for Pretrained Transformers

arXiv: https://arxiv.org/abs/2104.08812 · HTML: https://mlsanity.com/paper/2104.08812/

## Key facts

- Cited by 71 papers in the map (data through 14 Oct 2026).
- Builds on 36 of its 54 references that are in the map.
- First posted on arXiv 18 Apr 2021.

## Builds on (most cited first, 30 of 36)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://mlsanity.com/paper/1907.11692/) (2019)
- [A Simple Framework for Contrastive Learning of Visual Representations](https://mlsanity.com/paper/2002.05709/) (2020)
- [Representation Learning with Contrastive Predictive Coding](https://mlsanity.com/paper/1807.03748/) (2018)
- [Momentum Contrast for Unsupervised Visual Representation Learning](https://mlsanity.com/paper/1911.05722/) (2019)
- [GLUE: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding](https://mlsanity.com/paper/1804.07461/) (2018)
- [SQuAD: 100,000+ Questions for Machine Comprehension of Text](https://mlsanity.com/paper/1606.05250/) (2016)
- [Supervised Contrastive Learning](https://mlsanity.com/paper/2004.11362/) (2020)
- [A Broad-Coverage Challenge Corpus for Sentence Understanding through Inference](https://mlsanity.com/paper/1704.05426/) (2017)
- [A Baseline for Detecting Misclassified and Out-of-Distribution Examples in Neural Networks](https://mlsanity.com/paper/1610.02136/) (2016)
- [Concrete Problems in AI Safety](https://mlsanity.com/paper/1606.06565/) (2016)
- [A Simple Unified Framework for Detecting Out-of-Distribution Samples and Adversarial Attacks](https://mlsanity.com/paper/1807.03888/) (2018)
- [Understanding Contrastive Representation Learning through Alignment and Uniformity on the Hypersphere](https://mlsanity.com/paper/2005.10242/) (2020)
- [Enhancing The Reliability of Out-of-distribution Image Detection in Neural Networks](https://mlsanity.com/paper/1706.02690/) (2017)
- [Energy-based Out-of-distribution Detection](https://mlsanity.com/paper/2010.03759/) (2020)
- [Neural Network Acceptability Judgments](https://mlsanity.com/paper/1805.12471/) (2018)
- [Deep Anomaly Detection with Outlier Exposure](https://mlsanity.com/paper/1812.04606/) (2018)
- [Self-Supervised Learning of Pretext-Invariant Representations](https://mlsanity.com/paper/1912.01991/) (2019)
- [Deep Visual Domain Adaptation: A Survey](https://mlsanity.com/paper/1802.03601/) (2018)
- [Towards Open Set Deep Networks](https://mlsanity.com/paper/1511.06233/) (2015)
- [CSI: Novelty Detection via Contrastive Learning on Distributionally Shifted Instances](https://mlsanity.com/paper/2007.08176/) (2020)
- [Generalized ODIN: Detecting Out-of-distribution Image without Learning from Out-of-distribution Data](https://mlsanity.com/paper/2002.11297/) (2020)
- [Training Confidence-calibrated Classifiers for Detecting Out-of-Distribution Samples](https://mlsanity.com/paper/1711.09325/) (2017)
- [An Evaluation Dataset for Intent Classification and Out-of-Scope Prediction](https://mlsanity.com/paper/1909.02027/) (2019)
- [Pretrained Transformers Improve Out-of-Distribution Robustness](https://mlsanity.com/paper/2004.06100/) (2020)
- [Supervised Contrastive Learning for Pre-trained Language Model Fine-tuning](https://mlsanity.com/paper/2011.01403/) (2020)
- [DeCLUTR: Deep Contrastive Learning for Unsupervised Textual Representations](https://mlsanity.com/paper/2006.03659/) (2020)
- [Multi30K: Multilingual English-German Image Descriptions](https://mlsanity.com/paper/1605.00459/) (2016)
- [Classification-Based Anomaly Detection for General Data](https://mlsanity.com/paper/2005.02359/) (2020)

## Built on by (most cited first, 30 of 71)

- [Trustworthy LLMs: a Survey and Guideline for Evaluating Large Language Models' Alignment](https://mlsanity.com/paper/2308.05374/) (2023)
- [Delving into Out-of-Distribution Detection with Vision-Language Representations](https://mlsanity.com/paper/2211.13445/) (2022)
- [Out-of-Distribution Detection and Selective Generation for Conditional Language Models](https://mlsanity.com/paper/2209.15558/) (2022)
- [POEM: Out-of-Distribution Detection with Posterior Sampling](https://mlsanity.com/paper/2206.13687/) (2022)
- [A Survey of Safety and Trustworthiness of Large Language Models through the Lens of Verification and Validation](https://mlsanity.com/paper/2305.11391/) (2023)
- [Types of Out-of-Distribution Texts and How to Detect Them](https://mlsanity.com/paper/2109.06827/) (2021)
- [Context-faithful Prompting for Large Language Models](https://mlsanity.com/paper/2303.11315/) (2023)
- [GOOD-D: On Unsupervised Graph Out-Of-Distribution Detection](https://mlsanity.com/paper/2211.04208/) (2022)
- [Partial and Asymmetric Contrastive Learning for Out-of-Distribution Detection in Long-Tailed Recognition](https://mlsanity.com/paper/2207.01160/) (2022)
- [Uncertainty in Natural Language Processing: Sources, Quantification, and Applications](https://mlsanity.com/paper/2306.04459/) (2023)
- [A Survey on Out-of-Distribution Detection in NLP](https://mlsanity.com/paper/2305.03236/) (2023)
- [Is Fine-tuning Needed? Pre-trained Language Models Are Near Perfect for Out-of-Domain Detection](https://mlsanity.com/paper/2305.13282/) (2023)
- [Rainproof: An Umbrella To Shield Text Generators From Out-Of-Distribution Data](https://mlsanity.com/paper/2212.09171/) (2022)
- [How Good Are LLMs at Out-of-Distribution Detection?](https://mlsanity.com/paper/2308.10261/) (2023)
- [Does Your Model Classify Entities Reasonably? Diagnosing and Mitigating Spurious Correlations in Entity Typing](https://mlsanity.com/paper/2205.12640/) (2022)
- [EAT: Towards Long-Tailed Out-of-Distribution Detection](https://mlsanity.com/paper/2312.08939/) (2023)
- [Unsupervised Layer-wise Score Aggregation for Textual OOD Detection](https://mlsanity.com/paper/2302.09852/) (2023)
- [Exploring Large Language Models for Multi-Modal Out-of-Distribution Detection](https://mlsanity.com/paper/2310.08027/) (2023)
- [Fine-Tuning Deteriorates General Textual Out-of-Distribution Detection by Distorting Task-Agnostic Features](https://mlsanity.com/paper/2301.12715/) (2023)
- [Towards Computationally Feasible Deep Active Learning](https://mlsanity.com/paper/2205.03598/) (2022)
- [Estimating Soft Labels for Out-of-Domain Intent Detection](https://mlsanity.com/paper/2211.05561/) (2022)
- [Multi-Level Knowledge Distillation for Out-of-Distribution Detection in Text](https://mlsanity.com/paper/2211.11300/) (2022)
- [Does DetectGPT Fully Utilize Perturbation? Bridging Selective Perturbation to Fine-tuned Contrastive Learning Detector would be Better](https://mlsanity.com/paper/2402.00263/) (2024)
- [Your Finetuned Large Language Model is Already a Powerful Out-of-distribution Detector](https://mlsanity.com/paper/2404.08679/) (2024)
- [SeTAR: Out-of-Distribution Detection with Selective Low-Rank Approximation](https://mlsanity.com/paper/2406.12629/) (2024)
- [Enhancing Out-of-Distribution Detection in Natural Language Understanding via Implicit Layer Ensemble](https://mlsanity.com/paper/2210.11034/) (2022)
- [The future is different: Large pre-trained language models fail in prediction tasks](https://mlsanity.com/paper/2211.00384/) (2022)
- [Bag of Tricks for In-Distribution Calibration of Pretrained Transformers](https://mlsanity.com/paper/2302.06690/) (2023)
- [Multitask Active Learning for Graph Anomaly Detection](https://mlsanity.com/paper/2401.13210/) (2024)
- [Shh, don't say that! Domain Certification in LLMs](https://mlsanity.com/paper/2502.19320/) (2025)

## Abstract (arXiv)

Pretrained Transformers achieve remarkable performance when training and test data are from the same distribution. However, in real-world scenarios, the model often faces out-of-distribution (OOD) instances that can cause severe semantic shift problems at inference time. Therefore, in practice, a reliable model should identify such instances, and then either reject them during inference or pass them over to models that handle another distribution. In this paper, we develop an unsupervised OOD detection method, in which only the in-distribution (ID) data are used in training. We propose to fine-tune the Transformers with a contrastive loss, which improves the compactness of representations, such that OOD instances can be better differentiated from ID ones. These OOD instances can then be accurately detected using the Mahalanobis distance in the model's penultimate layer. We experiment with comprehensive settings and achieve near-perfect OOD detection performance, outperforming baselines drastically. We further investigate the rationales behind the improvement, finding that more compact representations through margin-based contrastive learning bring the improvement. We release our code to the community for future research.