Imagen de Portada

Towards Certification of Uncertainty Calibration under Adversarial Attacks

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Emde, Cornelius, Pinto, Francesco, Lukasiewicz, Thomas, Torr, Philip H. S., Bibi, Adel
Formato:	Preprint
Publicado:	2024
Materias:	Machine Learning
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

Shh, don't say that! Domain Certification in LLMs
por: Emde, Cornelius, et al.
Publicado: (2025)

When Do Prompting and Prefix-Tuning Work? A Theory of Capabilities and Limitations
por: Petrov, Aleksandar, et al.
Publicado: (2023)

Prompting a Pretrained Transformer Can Be a Universal Approximator
por: Petrov, Aleksandar, et al.
Publicado: (2024)

Efficient Error Certification for Physics-Informed Neural Networks
por: Eiras, Francisco, et al.
Publicado: (2023)

Bi-Factorial Preference Optimization: Balancing Safety-Helpfulness in Language Models
por: Zhang, Wenxuan, et al.
Publicado: (2024)

FORCE: Transferable Visual Jailbreaking Attacks via Feature Over-Reliance CorrEction
por: Lin, Runqi, et al.
Publicado: (2025)

Beyond Linear Probes: Dynamic Safety Monitoring for Language Models
por: Oldfield, James, et al.
Publicado: (2025)

Detecting LLM Hallucination Through Layer-wise Information Deficiency: Analysis of Ambiguous Prompts and Unanswerable Questions
por: Kim, Hazel, et al.
Publicado: (2024)

Do as I do (Safely): Mitigating Task-Specific Fine-tuning Risks in Large Language Models
por: Eiras, Francisco, et al.
Publicado: (2024)

Towards Interpretable Deep Local Learning with Successive Gradient Reconciliation
por: Yang, Yibo, et al.
Publicado: (2024)

Universal In-Context Approximation By Prompting Fully Recurrent Models
por: Petrov, Aleksandar, et al.
Publicado: (2024)

Segment, Select, Correct: A Framework for Weakly-Supervised Referring Segmentation
por: Eiras, Francisco, et al.
Publicado: (2023)

MIP against Agent: Malicious Image Patches Hijacking Multimodal OS Agents
por: Aichberger, Lukas, et al.
Publicado: (2025)

Continual Learning on a Diet: Learning from Sparsely Labeled Streams Under Constrained Computation
por: Zhang, Wenxuan, et al.
Publicado: (2024)

A Stable, Fast, and Fully Automatic Learning Algorithm for Predictive Coding Networks
por: Salvatori, Tommaso, et al.
Publicado: (2022)

Efficient Lifelong Model Evaluation in an Era of Rapid Progress
por: Prabhu, Ameya, et al.
Publicado: (2024)

From Categories to Classifiers: Name-Only Continual Learning by Exploring the Web
por: Prabhu, Ameya, et al.
Publicado: (2023)

PROSAC: Provably Safe Certification for Machine Learning Models under Adversarial Attacks
por: Feng, Chen, et al.
Publicado: (2024)

Revisiting Uncertainty Estimation and Calibration of Large Language Models
por: Tao, Linwei, et al.
Publicado: (2025)

Robustness Certificates for Neural Networks against Adversarial Attacks
por: Taheri, Sara, et al.
Publicado: (2025)

Hard Regularization to Prevent Deep Online Clustering Collapse without Data Augmentation
por: Mahon, Louis, et al.
Publicado: (2023)

Towards the Training of Deeper Predictive Coding Neural Networks
por: Qi, Chang, et al.
Publicado: (2025)

Mixture of Experts Made Intrinsically Interpretable
por: Yang, Xingyi, et al.
Publicado: (2025)

Rethinking Safety in LLM Fine-tuning: An Optimization Perspective
por: Kim, Minseon, et al.
Publicado: (2025)

On the Robustness of Adversarial Training Against Uncertainty Attacks
por: Ledda, Emanuele, et al.
Publicado: (2024)

On Pretraining Data Diversity for Self-Supervised Learning
por: Hammoud, Hasan Abed Al Kader, et al.
Publicado: (2024)

SynthCLIP: Are We Ready for a Fully Synthetic CLIP Training?
por: Hammoud, Hasan Abed Al Kader, et al.
Publicado: (2024)

Safer by Diffusion, Broken by Context: Diffusion LLM's Safety Blessing and Its Failure Mode
por: He, Zeyuan, et al.
Publicado: (2026)

No "Zero-Shot" Without Exponential Data: Pretraining Concept Frequency Determines Multimodal Model Performance
por: Udandarao, Vishaal, et al.
Publicado: (2024)

Model Merging and Safety Alignment: One Bad Model Spoils the Bunch
por: Hammoud, Hasan Abed Al Kader, et al.
Publicado: (2024)

Calibration Attacks: A Comprehensive Study of Adversarial Attacks on Model Confidence
por: Obadinma, Stephen, et al.
Publicado: (2024)

Measuring Uncertainty Calibration
por: Ciosek, Kamil, et al.
Publicado: (2025)

Exact Certification of Data-Poisoning Attacks Using Mixed-Integer Programming
por: Sosnin, Philip, et al.
Publicado: (2026)

Ensuring Calibration Robustness in Split Conformal Prediction Under Adversarial Attacks
por: Qian, Xunlei, et al.
Publicado: (2025)

Focus On This, Not That! Steering LLMs with Adaptive Feature Specification
por: Lamb, Tom A., et al.
Publicado: (2024)

Label Delay in Online Continual Learning
por: Csaba, Botos, et al.
Publicado: (2023)

Benchmarking Predictive Coding Networks -- Made Simple
por: Pinchetti, Luca, et al.
Publicado: (2024)

A Survey on Deep Learning Approaches for Tabular Data Generation: Utility, Alignment, Fidelity, Privacy, Diversity, and Beyond
por: Stoian, Mihaela Cătălina, et al.
Publicado: (2025)

Improving Semantic Uncertainty Quantification in Language Model Question-Answering via Token-Level Temperature Scaling
por: Lamb, Tom A., et al.
Publicado: (2026)

Relationship between Uncertainty in DNNs and Adversarial Attacks
por: Ogonna, Mabel, et al.
Publicado: (2024)