Copertina

Hessian Surgery: Class-Targeted Post-Hoc Rebalancing via Hessian Spike Perturbation

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Vigna, Hugo, Bontemps, Samuel
Natura:	Preprint
Pubblicazione:	2026
Soggetti:	Machine Learning Computer Vision and Pattern Recognition
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

EPTQ: Enhanced Post-Training Quantization via Hessian-guided Network-wise Optimization
di: Gordon, Ofir, et al.
Pubblicazione: (2023)

The effects of Hessian eigenvalue spectral density type on the applicability of Hessian analysis to generalization capability assessment of neural networks
di: Gabdullin, Nikita
Pubblicazione: (2025)

Taming the Long Tail: Rebalancing Adversarial Training via Adaptive Perturbation
di: Zhang, Lilin, et al.
Pubblicazione: (2026)

Visualizing high-dimensional loss landscapes with Hessian directions
di: Böttcher, Lucas, et al.
Pubblicazione: (2022)

Deferred Poisoning: Making the Model More Vulnerable via Hessian Singularization
di: He, Yuhao, et al.
Pubblicazione: (2024)

Investigating generalization capabilities of neural networks by means of loss landscapes and Hessian analysis
di: Gabdullin, Nikita
Pubblicazione: (2024)

MGVQ: Synergizing Multi-dimensional Sensitivity-Aware and Gradient-Hessian Fusion for Vector Quantization
di: Wang, Zhong, et al.
Pubblicazione: (2026)

Fast Post-Hoc Confidence Fusion for 3-Class Open-Set Aerial Object Detection
di: Loukovitis, Spyridon, et al.
Pubblicazione: (2025)

Technical Report: Activation Residual Hessian Quantization (ARHQ) for Low-Bit LLM Quantization
di: Wang, YiFeng, et al.
Pubblicazione: (2026)

PostHoc FREE Calibrating on Kolmogorov Arnold Networks
di: Liang, Wenhao, et al.
Pubblicazione: (2025)

APHQ-ViT: Post-Training Quantization with Average Perturbation Hessian Based Reconstruction for Vision Transformers
di: Wu, Zhuguanyu, et al.
Pubblicazione: (2025)

CAMP-HiVe: Cyclic Pair Merging based Efficient DNN Pruning with Hessian-Vector Approximation for Resource-Constrained Systems
di: Uddin, Mohammad Helal, et al.
Pubblicazione: (2025)

Post-Hoc Guidance for Consistency Models by Joint Flow Distribution Learning
di: Hsu, Chia-Hong, et al.
Pubblicazione: (2026)

Discriminative Feature Attributions: Bridging Post Hoc Explainability and Inherent Interpretability
di: Bhalla, Usha, et al.
Pubblicazione: (2023)

Not Blind but Silenced: Rebalancing Vision and Language via Adversarial Counter-Commonsense Equilibrium
di: Xiao, Qingxin, et al.
Pubblicazione: (2026)

Long-Tail Learning with Rebalanced Contrastive Loss
di: De Alvis, Charika, et al.
Pubblicazione: (2023)

Post-Hoc MOTS: Exploring the Capabilities of Time-Symmetric Multi-Object Tracking
di: Szabó, Gergely, et al.
Pubblicazione: (2024)

Train Once, Forget Precisely: Anchored Optimization for Efficient Post-Hoc Unlearning
di: Sanga, Prabhav, et al.
Pubblicazione: (2025)

Hessian Geometry of Latent Space in Generative Models
di: Lobashev, Alexander, et al.
Pubblicazione: (2025)

Feature Attribution Stability Suite: How Stable Are Post-Hoc Attributions?
di: Subramaniakuppusamy, Kamalasankari, et al.
Pubblicazione: (2026)

Post-Hoc Concept Disentanglement: From Correlated to Isolated Concept Representations
di: Erogullari, Eren, et al.
Pubblicazione: (2025)

On Background Bias of Post-Hoc Concept Embeddings in Computer Vision DNNs
di: Schwalbe, Gesina, et al.
Pubblicazione: (2025)

Removing Geometric Bias in One-Class Anomaly Detection with Adaptive Feature Perturbation
di: Hermary, Romain, et al.
Pubblicazione: (2025)

SEM: Sparse Embedding Modulation for Post-Hoc Debiasing of Vision-Language Models
di: Guimard, Quentin, et al.
Pubblicazione: (2026)

Long-Tailed Object Detection Pre-training: Dynamic Rebalancing Contrastive Learning with Dual Reconstruction
di: Duan, Chen-Long, et al.
Pubblicazione: (2024)

Ranked Activation Shift for Post-Hoc Out-of-Distribution Detection
di: Guglielmo, Gianluca, et al.
Pubblicazione: (2026)

MOS: Model Surgery for Pre-Trained Model-Based Class-Incremental Learning
di: Sun, Hai-Long, et al.
Pubblicazione: (2024)

Clinical Domain Knowledge-Derived Template Improves Post Hoc AI Explanations in Pneumothorax Classification
di: Yuan, Han, et al.
Pubblicazione: (2024)

DeepSeek on a Trip: Inducing Targeted Visual Hallucinations via Representation Vulnerabilities
di: Islam, Chashi Mahiul, et al.
Pubblicazione: (2025)

Merge-Friendly Post-Training Quantization for Multi-Target Domain Adaptation
di: Shin, Juncheol, et al.
Pubblicazione: (2025)

Adversarial Example Defense via Perturbation Grading Strategy
di: Zhu, Shaowei, et al.
Pubblicazione: (2022)

Breaking Class Barriers: Efficient Dataset Distillation via Inter-Class Feature Compensator
di: Zhang, Xin, et al.
Pubblicazione: (2024)

ReSpike: Residual Frames-based Hybrid Spiking Neural Networks for Efficient Action Recognition
di: Xiao, Shiting, et al.
Pubblicazione: (2024)

Improving Generalization of Universal Adversarial Perturbation via Dynamic Maximin Optimization
di: Zhang, Yechao, et al.
Pubblicazione: (2025)

Rebalancing Multi-Label Class-Incremental Learning
di: Du, Kaile, et al.
Pubblicazione: (2024)

Towards Robust Spiking Neural Networks:Mitigating Heterogeneous Training Vulnerability via Dominant Eigencomponent Projection
di: Zhang, Desong, et al.
Pubblicazione: (2025)

BSO: Binary Spiking Online Optimization Algorithm
di: Liang, Yu, et al.
Pubblicazione: (2025)

Defending Against Gradient Inversion Attacks for Biomedical Images via Learnable Data Perturbation
di: Jiang, Shiyi, et al.
Pubblicazione: (2025)

FairDiffusion: Enhancing Equity in Latent Diffusion Models via Fair Bayesian Perturbation
di: Luo, Yan, et al.
Pubblicazione: (2024)

Reducing Class-Wise Performance Disparity via Margin Regularization
di: Zhu, Beier, et al.
Pubblicazione: (2026)