analiza aiqbench
This research introduces Conflict-aware Evidential Deep Learning (C-EDL), a lightweight post-hoc method designed to mitigate the vulnerability of standard uncertainty quantification models to adversarial attacks and out-of-distribution inputs. By leveraging representational disagreement, C-EDL significantly improves the detection of malicious perturbations while maintaining high accuracy for legitimate data without requiring model retraining. The approach offers a practical pathway toward deploying more reliable deep learning systems in high-stakes environments where overconfidence on corrupted inputs poses critical risks.
Robust Adversarial Quantification via Conflict-Aware Evidential Deep Learning
arXiv:2506.05937v3 Announce Type: replace-cross Abstract: Reliability of deep learning models is critical for deployment in high-stakes applications, where out-of-distribution or adversarial inputs may lead to detrimental outcomes. Evidential Deep Learning, an efficient paradigm for uncertainty quantification, models predictions as Dirichlet distributions of a single forward pass. However, EDL is particularly vulnerable to adversarially perturbed inputs, making overconfident errors. Conflict-aware Evidential Deep Learning~\mbox{(C-EDL)} is a lightweight post-hoc uncertainty quantification approach that mitigates these issues, enhancing adversarial and OOD robustness without retraining. C-EDL generates diverse, task-preserving transformations per input and quantifies representational disagreement to calibrate uncertainty estimates when needed. C-EDL's conflict-aware prediction adjustment improves detection of OOD and adversarial inputs, maintaining high in-distribution accuracy and low computational overhead. Our experimental evaluation shows that C-EDL significantly outperforms state-of-the-art EDL variants and competitive baselines, achieving substantial reductions in coverage for OOD data (up to $\approx55\%$) and adversarial data (up to $\approx90\%$), across a range of datasets, attack types, and uncertainty metrics.
Abstract:Reliability of deep learning models is critical for deployment in high-stakes applications, where out-of-distribution or adversarial inputs may lead to detrimental outcomes. Evidential Deep Learning, an efficient paradigm for uncertainty quantification, models predictions as Dirichlet distributions of a single forward pass. However, EDL is particularly vulnerable to adversarially perturbed inputs, making overconfident errors. Conflict-aware Evidential Deep Learning~\mbox{(C-EDL)} is a lightweight post-hoc uncertainty quantification approach that mitigates these issues, enhancing adversarial and OOD robustness without retraining. C-EDL generates diverse, task-preserving transformations per input and quantifies representational disagreement to calibrate uncertainty estimates when needed. C-EDL's conflict-aware prediction adjustment improves detection of OOD and adversarial inputs, maintaining high in-distribution accuracy and low computational overhead. Our experimental evaluation shows that C-EDL significantly outperforms state-of-the-art EDL variants and competitive baselines, achieving substantial reductions in coverage for OOD data (up to $\approx55\%$) and adversarial data (up to $\approx90\%$), across a range of datasets, attack types, and uncertainty metrics.
Updated to the published ICLR 2026 version, including revised title. Published version: this https URL
Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
arXiv:2506.05937 [cs.LG]
(or arXiv:2506.05937v3 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2506.05937 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 262 cuvinte
Subiecte conexe
Robust Evidential Learning Through Latent Consistency
arXiv · 2 oct. 2026
Synthetic document finetuning for instilling positive traits
AI Alignment Forum · 16 iun. 2026
The Three Filters: Why Almost Every Plan to Survive ASI Fails Miserably
LessWrong · 10 iun. 2026
Latent reasoning architectures would likely undermine CoT, our strongest oversight tool
AI Alignment Forum · 23 sept. 2026