IC-1003Pretrained ResNet-50 and ViT-B/16 exhibit neuron activation patterns that are separable between in-distribution and out-of-distribution inputs, enabling post-hoc OOD detection without model modification

Yibing Liu, Chris XING TIAN, Haoliang Li, Lei Ma, Shiqi Wang

SourceNeuron Activation Coverage: Rethinking Out-of-distribution Detection and Generalization

The paper formulates neuron activation states as the product of raw neuron outputs and KL-divergence gradients, then measures the coverage of these states under in-distribution training data. On pretrained ResNet-50 and ViT-B/16 (both trained on ImageNet-1k), out-of-distribution inputs trigger neuron states that fall outside the expected coverage area, producing lower coverage scores than in-distribution inputs. This separability in the models' internal neuron behavior allows a post-hoc uncertainty score (NAC-UE) to distinguish IN from OOD samples without modifying the model weights, achieving state-of-the-art AUROC across three OOD datasets.

Evidence
correlational
Key metric
AUROC: ResNet-50 96.52 (iNaturalist), 91.45 (OpenImage-O), 97.9 (Textures); ViT-B/16 93.72 (iNaturalist), 91.58 (OpenImage-O), 94.17 (Textures)
Caveat
The separability is measured using the paper's specific neuron state formulation (z ⊙ ∂dkl/∂z); raw neuron outputs alone show less separation (FPR95 35.72% vs 16.58% on layer4 of ResNet-50). The method requires a small subset of training data (1,000–50,000 images) to build the coverage model.
Model
ResNet / ResNet-152 / ResNet-101 / ResNet-50-BN ResNet50, ViT ViT-B/16
Datasets
ImageNet-1k / ImageNet / ImageNet-1k-val / ImageNet-Val [source], iNaturalist [eval], OpenImages-O / OpenImage-O [eval], Textures [eval]
Methods
VIM [compared-to], k-nearest neighbours classifier / Nearest-neighbor baseline [compared-to], ReAct [compared-to], ASH [compared-to]
Related work
VIM [compared-to], k-nearest neighbours classifier / Nearest-neighbor baseline [compared-to], ReAct [compared-to], ASH [compared-to]
Extraction
automatic-extraction