Copertina

SAFE-KD: Risk-Controlled Early-Exit Distillation for Vision Backbones

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autore principale:	Khazem, Salim
Natura:	Preprint
Pubblicazione:	2026
Soggetti:	Machine Learning Artificial Intelligence Computer Vision and Pattern Recognition
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

Margin and Consistency Supervision for Calibrated and Robust Vision Models
di: Khazem, Salim
Pubblicazione: (2026)

AdapterTune: Zero-Initialized Low-Rank Adapters for Frozen Vision Transformers
di: Khazem, Salim
Pubblicazione: (2026)

TopoLoRA-SAM: Topology-Aware Parameter-Efficient Adaptation of Foundation Segmenters for Thin-Structure and Cross-Domain Binary Semantic Segmentation
di: Khazem, Salim
Pubblicazione: (2026)

PolygoNet: Leveraging Simplified Polygonal Representation for Effective Image Classification
di: Khazem, Salim, et al.
Pubblicazione: (2025)

MC-RFM: Geometry-Aware Few-Shot Adaptation via Mixed-Curvature Riemannian Flow Matching
di: Khazem, Salim, et al.
Pubblicazione: (2026)

Vision-LSTM: xLSTM as Generic Vision Backbone
di: Alkin, Benedikt, et al.
Pubblicazione: (2024)

Cyclical Temporal Encoding and Hybrid Deep Ensembles for Multistep Energy Forecasting
di: Khazem, Salim, et al.
Pubblicazione: (2025)

DeepKD: A Deeply Decoupled and Denoised Knowledge Distillation Trainer
di: Huang, Haiduo, et al.
Pubblicazione: (2025)

DeeAD: Dynamic Early Exit of Vision-Language Action for Efficient Autonomous Driving
di: HU, Haibo, et al.
Pubblicazione: (2025)

KD-OCT: Efficient Knowledge Distillation for Clinical-Grade Retinal OCT Classification
di: Nourbakhsh, Erfan, et al.
Pubblicazione: (2025)

ScaleKD: Strong Vision Transformers Could Be Excellent Teachers
di: Fan, Jiawei, et al.
Pubblicazione: (2024)

ViR: Towards Efficient Vision Retention Backbones
di: Hatamizadeh, Ali, et al.
Pubblicazione: (2023)

Adv-KD: Adversarial Knowledge Distillation for Faster Diffusion Sampling
di: Mekonnen, Kidist Amde, et al.
Pubblicazione: (2024)

Vision Backbone Efficient Selection for Image Classification in Low-Data Regimes
di: Guerin, Joris, et al.
Pubblicazione: (2024)

DisCoM-KD: Cross-Modal Knowledge Distillation via Disentanglement Representation and Adversarial Learning
di: Ienco, Dino, et al.
Pubblicazione: (2024)

PTEENet: Post-Trained Early-Exit Neural Networks Augmentation for Inference Cost Optimization
di: Lahiany, Assaf, et al.
Pubblicazione: (2025)

Personalized Federated Learning via Backbone Self-Distillation
di: Wang, Pengju, et al.
Pubblicazione: (2024)

Fast yet Safe: Early-Exiting with Risk Control
di: Jazbec, Metod, et al.
Pubblicazione: (2024)

FerKD: Surgical Label Adaptation for Efficient Distillation
di: Shen, Zhiqiang
Pubblicazione: (2023)

BYO-Eval: Build Your Own Dataset for Fine-Grained Visual Assessment of Multimodal Language Models
di: Arnould, Ludovic, et al.
Pubblicazione: (2025)

EPSD: Early Pruning with Self-Distillation for Efficient Model Compression
di: Chen, Dong, et al.
Pubblicazione: (2024)

Group Distributionally Robust Dataset Distillation with Risk Minimization
di: Vahidian, Saeed, et al.
Pubblicazione: (2024)

Unlocking Pre-trained Image Backbones for Semantic Image Synthesis
di: Berrada, Tariq, et al.
Pubblicazione: (2023)

CAPEEN: Image Captioning with Early Exits and Knowledge Distillation
di: Bajpai, Divya Jyoti, et al.
Pubblicazione: (2024)

Amortized-Precision Quantization for Early-Exit Vision Transformers
di: Fang, Rui, et al.
Pubblicazione: (2026)

Early Exit and Multi Stage Knowledge Distillation in VLMs for Video Summarization
di: Khan, Anas Anwarul Haq, et al.
Pubblicazione: (2025)

Towards Long-window Anchoring in Vision-Language Model Distillation
di: Zhou, Haoyi, et al.
Pubblicazione: (2025)

Masking Teacher and Reinforcing Student for Distilling Vision-Language Models
di: Lee, Byung-Kwan, et al.
Pubblicazione: (2025)

Dataset Distillation for Pre-Trained Self-Supervised Vision Models
di: Cazenavette, George, et al.
Pubblicazione: (2025)

EfficientTrain++: Generalized Curriculum Learning for Efficient Visual Backbone Training
di: Wang, Yulin, et al.
Pubblicazione: (2024)

Synergy and Diversity in CLIP: Enhancing Performance Through Adaptive Backbone Ensembling
di: Rodriguez-Opazo, Cristian, et al.
Pubblicazione: (2024)

DELT: A Simple Diversity-driven EarlyLate Training for Dataset Distillation
di: Shen, Zhiqiang, et al.
Pubblicazione: (2024)

Distilling Out-of-Distribution Robustness from Vision-Language Foundation Models
di: Zhou, Andy, et al.
Pubblicazione: (2023)

Closed-Form Linear-Probe Dataset Distillation for Pre-trained Vision Models
di: Peng, Bincheng, et al.
Pubblicazione: (2026)

Weight Copy and Low-Rank Adaptation for Few-Shot Distillation of Vision Transformers
di: Grigore, Diana-Nicoleta, et al.
Pubblicazione: (2024)

Adaptive Knowledge Distillation for Classification of Hand Images using Explainable Vision Transformers
di: Nguyen, Thanh Thi, et al.
Pubblicazione: (2024)

COMODO: Cross-Modal Video-to-IMU Distillation for Efficient Egocentric Human Activity Recognition
di: Chen, Baiyu, et al.
Pubblicazione: (2025)

DeiT-LT Distillation Strikes Back for Vision Transformer Training on Long-Tailed Datasets
di: Rangwani, Harsh, et al.
Pubblicazione: (2024)

Multi-Scale Visual Prompting for Lightweight Small-Image Classification
di: Khazem, Salim
Pubblicazione: (2025)

DARK: Diagonal-Anchored Repulsive Knowledge Distillation for Vision-Language Models under Extreme Compression
di: Saeed, Numan, et al.
Pubblicazione: (2026)