Sample-wise neural collapse reveals that feature-classifier misalignment drives TTA degradation under shifts, which NCTTA corrects via hybrid geometric-predictive targets.
arXiv preprint arXiv:2303.15361 , year=
6 Pith papers cite this work. Polarity classification is still indexing.
6
Pith papers citing it
representative citing papers
A multi-level diversification wrapper for test-time adaptation that treats entropy minimization as multi-hypothesis inference to reduce underspecification and improve robustness by 1-4%.
IDEA is a TTA framework for VLN that builds a dynamic asset library from Fisher-weighted soft prompts and domain coordinates, then uses convex-hull projection for cross-domain bridging and training-free adaptation.
A technique for parametric stylistic control in latent diffusion models learns disentangled directions from synthetic datasets and applies them via guidance composition while preserving semantics.
MoASE++ combines activation sparsity experts with domain-adaptive on-policy distillation to achieve state-of-the-art continual test-time adaptation on image classification and segmentation benchmarks.
citing papers explorer
-
Neural Collapse in Test-Time Adaptation
Sample-wise neural collapse reveals that feature-classifier misalignment drives TTA degradation under shifts, which NCTTA corrects via hybrid geometric-predictive targets.
-
Multi-Hypothesis Test-Time Adaptation to Mitigate Underspecification
A multi-level diversification wrapper for test-time adaptation that treats entropy minimization as multi-hypothesis inference to reduce underspecification and improve robustness by 1-4%.
-
Turning Adaptation into Assets: Cross-Domain Bridging for Online Vision-Language Navigation
IDEA is a TTA framework for VLN that builds a dynamic asset library from Fisher-weighted soft prompts and domain coordinates, then uses convex-hull projection for cross-domain bridging and training-free adaptation.
-
Stylistic Attribute Control in Latent Diffusion Models
A technique for parametric stylistic control in latent diffusion models learns disentangled directions from synthetic datasets and applies them via guidance composition while preserving semantics.
-
MoASE++: Mixture of Activation Sparsity Experts with Domain-Adaptive On-policy Distillation for Continual Test Time Adaptation
MoASE++ combines activation sparsity experts with domain-adaptive on-policy distillation to achieve state-of-the-art continual test-time adaptation on image classification and segmentation benchmarks.