# COME: Test-time adaption by Conservatively Minimizing Entropy

arXiv: https://arxiv.org/abs/2410.10894 · HTML: https://mlsanity.com/paper/2410.10894/

## Key facts

- Cited by 31 papers in the map (data through 14 Oct 2026).
- Builds on 43 of its 58 references that are in the map.
- First posted on arXiv 12 Oct 2024.

## Builds on (most cited first, 30 of 43)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [On Calibration of Modern Neural Networks](https://mlsanity.com/paper/1706.04599/) (2017)
- [Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learning](https://mlsanity.com/paper/1506.02142/) (2015)
- [Simple and Scalable Predictive Uncertainty Estimation using Deep Ensembles](https://mlsanity.com/paper/1612.01474/) (2016)
- [A Baseline for Detecting Misclassified and Out-of-Distribution Examples in Neural Networks](https://mlsanity.com/paper/1610.02136/) (2016)
- [Describing Textures in the Wild](https://mlsanity.com/paper/1311.3618/) (2013)
- [Energy-based Out-of-distribution Detection](https://mlsanity.com/paper/2010.03759/) (2020)
- [Deep Anomaly Detection with Outlier Exposure](https://mlsanity.com/paper/1812.04606/) (2018)
- [AugMix: A Simple Data Processing Method to Improve Robustness and Uncertainty](https://mlsanity.com/paper/1912.02781/) (2019)
- [Do We Really Need to Access the Source Data? Source Hypothesis Transfer for Unsupervised Domain Adaptation](https://mlsanity.com/paper/2002.08546/) (2020)
- [Deep Neural Networks are Easily Fooled: High Confidence Predictions for Unrecognizable Images](https://mlsanity.com/paper/1412.1897/) (2014)
- [Evidential Deep Learning to Quantify Classification Uncertainty](https://mlsanity.com/paper/1806.01768/) (2018)
- [Predictive Uncertainty Estimation via Prior Networks](https://mlsanity.com/paper/1802.10501/) (2018)
- [Continual Test-Time Domain Adaptation](https://mlsanity.com/paper/2203.13591/) (2022)
- [Efficient Test-Time Model Adaptation without Forgetting](https://mlsanity.com/paper/2204.02610/) (2022)
- [Towards Stable Test-Time Adaptation in Dynamic Wild World](https://mlsanity.com/paper/2302.12400/) (2023)
- [ViM: Out-Of-Distribution with Virtual-logit Matching](https://mlsanity.com/paper/2203.10807/) (2022)
- [MEMO: Test Time Robustness via Adaptation and Augmentation](https://mlsanity.com/paper/2110.09506/) (2021)
- [Open-Set Recognition: a Good Closed-Set Classifier is All You Need?](https://mlsanity.com/paper/2110.06207/) (2021)
- [OpenOOD: Benchmarking Generalized Out-of-Distribution Detection](https://mlsanity.com/paper/2210.07242/) (2022)
- [Contrastive Test-Time Adaptation](https://mlsanity.com/paper/2204.10377/) (2022)
- [Mitigating Neural Network Overconfidence with Logit Normalization](https://mlsanity.com/paper/2205.09310/) (2022)
- [Parameter-free Online Test-time Adaptation](https://mlsanity.com/paper/2201.05718/) (2022)
- [OpenOOD v1.5: Enhanced Benchmark for Out-of-Distribution Detection](https://mlsanity.com/paper/2306.09301/) (2023)
- [Trusted Multi-View Classification with Dynamic Evidential Fusion](https://mlsanity.com/paper/2204.11423/) (2022)
- [Uncertainty Quantification and Deep Ensembles](https://mlsanity.com/paper/2007.08792/) (2020)
- [In or Out? Fixing ImageNet Out-of-Distribution Detection Evaluation](https://mlsanity.com/paper/2306.00826/) (2023)
- [Confidence-Aware Learning for Deep Neural Networks](https://mlsanity.com/paper/2007.01458/) (2020)
- [Benchmarking Neural Network Robustness to Common Corruptions and Surface Variations](https://mlsanity.com/paper/1807.01697/) (2018)

## Built on by (most cited first, 30 of 31)

- [The Unreasonable Effectiveness of Entropy Minimization in LLM Reasoning](https://mlsanity.com/paper/2505.15134/) (2025)
- [Right Question is Already Half the Answer: Fully Unsupervised LLM Reasoning Incentivization](https://mlsanity.com/paper/2504.05812/) (2025)
- [Test-Time Learning for Large Language Models](https://mlsanity.com/paper/2505.20633/) (2025)
- [Ranked Entropy Minimization for Continual Test-Time Adaptation](https://mlsanity.com/paper/2505.16441/) (2025)
- [Monitoring Risks in Test-Time Adaptation](https://mlsanity.com/paper/2507.08721/) (2025)
- [You only need 4 extra tokens: Synergistic Test-time Adaptation for LLMs](https://mlsanity.com/paper/2510.10223/) (2025)
- [When and Where to Reset Matters for Long-Term Test-Time Adaptation](https://mlsanity.com/paper/2603.03796/) (2026)
- [ZeroSiam: An Efficient Asymmetry for Test-Time Entropy Optimization without Collapse](https://mlsanity.com/paper/2509.23183/) (2025)
- [D-TPT: Dimensional Entropy Maximization for Calibrating Test-Time Prompt Tuning in Vision-Language Models](https://mlsanity.com/paper/2510.09473/) (2025)
- [Class-aware Domain Knowledge Fusion and Fission for Continual Test-Time Adaptation](https://mlsanity.com/paper/2510.12150/) (2025)
- [Think Twice before Adaptation: Improving Adaptability of DeepFake Detection via Online Test-Time Adaptation](https://mlsanity.com/paper/2505.18787/) (2025)
- [Feature-Based Instance Neighbor Discovery: Advanced Stable Test-Time Adaptation in Dynamic World](https://mlsanity.com/paper/2506.06782/) (2025)
- [When Test-Time Adaptation Meets Self-Supervised Models](https://mlsanity.com/paper/2506.23529/) (2025)
- [Neural Collapse in Test-Time Adaptation](https://mlsanity.com/paper/2512.10421/) (2025)
- [Nip Rumors in the Bud: Retrieval-Guided Topic-Level Adaptation for Test-Time Fake News Video Detection](https://mlsanity.com/paper/2601.11981/) (2026)
- [Tempora: Characterising the Time-Contingent Utility of Online Test-Time Adaptation](https://mlsanity.com/paper/2602.06136/) (2026)
- [ProtoDCS: Towards Robust and Efficient Open-Set Test-Time Adaptation for Vision-Language Models](https://mlsanity.com/paper/2602.23653/) (2026)
- [Unsupervised Layer-Wise Dynamic Test Time Adaptation for LLMs](https://mlsanity.com/paper/2602.09719/) (2026)
- [EviATTA: Evidential Active Test-Time Adaptation for Medical Segment Anything Models](https://mlsanity.com/paper/2603.14666/) (2026)
- [Training-Free Test-Time Contrastive Learning for Large Language Models](https://mlsanity.com/paper/2604.13552/) (2026)
- [VISTA: Variance-Gated Inter-Sequence Test-Time Adaptation for Multi-Sequence MRI Segmentation](https://mlsanity.com/paper/2605.17433/) (2026)
- [Self-Improving In-Context Learning](https://mlsanity.com/paper/2605.23180/) (2026)
- [Sample-wise Targeted Adversarial Attacks on Test-time Adaptation](https://mlsanity.com/paper/2605.23411/) (2026)
- [Private and Stable Test-Time Adaptation with Differential Privacy](https://mlsanity.com/paper/2606.01908/) (2026)
- [Entropy Minimization without Model Collapse: Mitigating Prediction Bias in Medical Imaging](https://mlsanity.com/paper/2606.02339/) (2026)
- [Closing the Loop on Latent Reasoning via Test-Time Reconstruction](https://mlsanity.com/paper/2606.06252/) (2026)
- [EASE-TTT: Evidence-Aligned Selective Test-Time Training for Long-Context Question Answering](https://mlsanity.com/paper/2606.06906/) (2026)
- [MixTTA: Low-Rank Cross-Channel Mixing for Reliable Test-Time Adaptation](https://mlsanity.com/paper/2606.28142/) (2026)
- [Explanation Stability of Test-Time Adaptation in Computational Pathology: A Large-Scale Benchmark](https://mlsanity.com/paper/2608.07062/) (2026)
- [Test-Time Adaptation for ECG Classification via SQI-Gated Self-Training and Beat-Rhythm Consistency](https://mlsanity.com/paper/2608.23347/) (2026)

## Abstract (arXiv)

Machine learning models must continuously self-adjust themselves for novel data distribution in the open world. As the predominant principle, entropy minimization (EM) has been proven to be a simple yet effective cornerstone in existing test-time adaption (TTA) methods. While unfortunately its fatal limitation (i.e., overconfidence) tends to result in model collapse. For this issue, we propose to Conservatively Minimize the Entropy (COME), which is a simple drop-in replacement of traditional EM to elegantly address the limitation. In essence, COME explicitly models the uncertainty by characterizing a Dirichlet prior distribution over model predictions during TTA. By doing so, COME naturally regularizes the model to favor conservative confidence on unreliable samples. Theoretically, we provide a preliminary analysis to reveal the ability of COME in enhancing the optimization stability by introducing a data-adaptive lower bound on the entropy. Empirically, our method achieves state-of-the-art performance on commonly used benchmarks, showing significant improvements in terms of classification accuracy and uncertainty estimation under various settings including standard, life-long and open-world TTA, i.e., up to 34.5% improvement on accuracy and 15.1% on false positive rate.