AI Paper Atlas

Research map / Adversarial robustness and security / Certified and empirical robustness

Adversarial Robustness of Implicit Neural Representation-Based Classifiers

Jayoung Kim, Kookjin Lee, Noseong Park, Sanghyun Hong

ICML 2026 · Adversarial robustness and security › Certified and empirical robustness · approach: Robustness-accuracy trade-off analysis

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use FD INR surrogate instead of Naive INR surrogate (ablation)PGD attack on DWS · MNISTperturbed accuracy: 10.12% -> 0.01%Improved
use FD INR surrogate instead of Naive INR surrogate (ablation)PGD attack on NFT · MNISTperturbed accuracy: 20.57% -> 0.00%Improved
use FD INR surrogate instead of Naive INR surrogate (ablation)PGD attack on Inr2Vec · F-MNISTperturbed accuracy: 17.91% -> 13.89%Improved
use FD INR surrogate instead of Naive INR surrogate (ablation)PGD attack on Inr2Vec · CIFAR-10perturbed accuracy: 0.46% -> 5.22%Hurt
use FD embedding surrogate instead of Naive embedding surrogate (ablation)PGD attack on Inr2Vec · CIFAR-10perturbed accuracy: 4.58% -> 13.59%Hurt
perturb INR weights (function space) instead of input space via surrogate (ablation)attack on NFT · CIFAR-10perturbed accuracy: 0.15% -> 39.82%Hurt
perturb INR weights (function space) instead of input space via surrogate (ablation)attack on DWS · MNISTperturbed accuracy: 0.01% -> 47.74%Hurt
add diffusion denoising defenseFD INR attack on NFN · F-MNISTperturbed accuracy: 0.27% -> 14.22%Improved
add diffusion denoising defenseFD INR attack on NFT · MNISTperturbed accuracy: 0.00% -> 0.00%No clear effect
add diffusion denoising defenseFD INR attack on ScaleGMN · CIFAR-10perturbed accuracy: 0.25% -> 4.84%Improved
add function-space adversarial trainingNFN · MNISTperturbed accuracy: 25.8% -> 81.1%Improved
add function-space adversarial trainingNFT · CIFAR-10perturbed accuracy: 39.8% -> 24.1%Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy