Simplification Is All You Need against Out-of-Distribution Overconfidence

Keke Tang, Chao Hou, Weilong Peng, Xiang Fang, Zhize Wu, Yongwei Nie, Wenping Wang, Zhihong Tian

Abstract

Deep Neural Networks (DNNs) are notoriously overconfident when encountering data they were not trained on, a phenomenon known as Out-of-Distribution (OOD) overconfidence. This paper argues that the root cause of this flaw is model over-complexity—specifically, excessive parameter capacity and high nonlinearity. We propose a model simplification framework that utilizes knowledge distillation to reduce capacity and selective ReLU removal to decrease nonlinearity. Our results demonstrate that these simplified models significantly mitigate OOD overconfidence and improve detection performance across standard benchmarks without sacrificing accuracy on the original task.

Introduction

In controlled laboratory settings, DNNs achieve state-of-the-art performance. However, in real-world applications such as autonomous driving or medical diagnosis, they face the dangerous challenge of OOD overconfidence. When presented with unfamiliar inputs, these models often produce incorrect predictions with near-total confidence.

Existing solutions typically rely on post-hoc calibration or training with auxiliary OOD datasets. We argue these are merely symptomatic treatments. Instead, we identify over-parameterization and high nonlinearity as the structural drivers that allow models to "hallucinate" patterns in noise. By simplifying the model architecture, we can create smoother decision boundaries that naturally reflect uncertainty for OOD data.

Methodology

We propose a two-pronged strategy for model simplification:

  1. Capacity Reduction via Knowledge Distillation (KD): Rather than deploying massive "teacher" networks, we use KD to train smaller "student" networks. The student captures essential features from the teacher while lacking the excess parameters that fuel overconfident OOD predictions.
  2. Nonlinearity Reduction via Selective ReLU Removal: ReLU activations are the primary source of nonlinearity in DNNs. We introduce the Negative Response Ratio (NRO) to identify ReLU layers with minimal contribution to task performance. These layers are selectively replaced with identity mappings during the distillation process.
OOD detection performance (FPR95 and AUROC) for a ResNet-152 teacher and various distilled student models. The ResNet-50 student achieves the 'Goldilocks' capacity, outperforming the teacher.

Key Insights and Results

The "Goldilocks" Capacity

Our experiments reveal a sweet spot for model size. As shown in the comparison between a ResNet-152 teacher and its distilled students, reducing capacity improves OOD metrics like FPR95 and AUROC up to a point. However, if the model becomes too small (e.g., ResNet-18 in some contexts), it may begin to lose the ability to handle the original In-Distribution (ID) task effectively.

Shifting the Confidence Distribution

Simplification directly impacts the model's confidence on OOD samples. On the CIFAR-100 benchmark, a standard ResNet-50 exhibited a Mean Maximum Confidence (MMC) of 72.12% on OOD data, whereas our simplified ResNet-18 student reduced this to 48.35%.

Confidence distributions of the teacher, student, and student with ReLU removal across four OOD benchmarks. Simplification shifts the distribution toward lower confidence for OOD samples.

Synergy and Safety

Unlike many state-of-the-art methods, our approach is data-independent and does not require exposure to outlier data during training. Furthermore, simplification acts as a foundation that enhances existing OOD detection tools (e.g., Energy scores or ReAct). When combined, simplified models outperform specialized mitigation methods like ACET and CEDA, particularly in challenging large-scale scenarios like ImageNet-1k.

Conclusion

Model simplification offers a robust, principled approach to the OOD overconfidence problem. By smoothing the model's response through reduced capacity and nonlinearity, we achieve safer, more reliable predictions. Future work will explore the application of these principles to modern architectures like Vision Transformers (ViTs) and further investigate the theoretical underpinnings of why simplification specifically targets OOD overconfidence.

Summary updated

Abstract

Deep neural networks (DNNs) often exhibit out-of-distribution (OOD) overconfidence, producing overly confident predictions on OOD samples. We attribute this issue to the inherent over-complexity of DNNs and investigate two key aspects: capacity and nonlinearity. First, we demonstrate that reducing model capacity through knowledge distillation can effectively mitigate OOD overconfidence. Second, we show that selectively reducing nonlinearity by removing ReLU operations further alleviates the issue. Building on these findings, we present a practical guide to model simplification, combining both strategies to significantly reduce OOD overconfidence. Extensive experiments validate the effectiveness of this approach in mitigating OOD overconfidence and demonstrate its superiority over state-of-the-art methods. Additionally, our simplification strategies can be combined with existing OOD detection techniques to further enhance OOD detection performance.
BibTeX
@inproceedings{tang2025simplification,
  author = {Keke Tang and Chao Hou and Weilong Peng and Xiang Fang and Zhize Wu and Yongwei Nie and Wenping Wang and Zhihong Tian},
  title = {{Simplification Is All You Need against Out-of-Distribution Overconfidence}},
  booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition},
  year = {2025},
  doi = {10.1109/cvpr52734.2025.00474},
}
Figure