Diverge to Converge: Mutual Heterogeneous Learning for Robust Pruning

Learning with Bilevel-Minimax Optimization for Efficient and Reliable Transfer Attacks

This work proposes BMAT (Bilevel-Minimax Adversarial Transfer), an integrated bottom-up solver that combines a Soft Weight Modulator and an Implicit Gradient Approximator to enable ternary coupling among initialization, surrogate adaptation, and perturbation optimization.

Adversarial LassoNet: Robust Feature Selection via Stability-Driven Sparse Learning

Adversarial LassoNet is proposed, a stability-driven sparse feature selection framework that integrates input-space adversarial perturbations with the hierarchical sparsity mechanism of LassoNet and an NTK-inspired spectral analysis to characterize how perturbation-driven training can reduce gradient concentration.

Robust CurveMoE: Multi-Norm Adversarial Defense for Mixture-of-Experts Models via Mode Connectivity

Multi-norm adversarial defense aims to protect neural networks against perturbations defined by different norm constraints, but existing methods typically optimize competing robustness objectives within a single parameter configuration, leading to substantial training cost and unfavorable robustness trade-offs. We propose Robust CurveMoE, an efficient mixture-of-experts framework that connects models specialized for different perturbation norms through a low-loss path and exploits the complementary robustness profiles of models along this path. Robust CurveMoE derives clean and norm-specialized experts from robustness-constrained curve locations and selectively expertizes only influential layers, while sharing the remaining parameters across routing paths. To further reduce curve-construction cost, we introduce contribution-guided partial updating, which selects influential curve parameters using initialization-based gradient scores. We also theoretically bound the objective gap between partial and full curve optimization. Experiments on CIFAR-100 and ImageNet-100 with WideResNet and Vision Transformer architectures show that Robust CurveMoE consistently improves clean, norm-specific, and Union accuracy over MSD and ERMC. In particular, it improves Union accuracy by 2.37 and 2.13 percentage points over the strongest baseline on CIFAR-100 and ImageNet-100, respectively. Extensive ablations further validate the effectiveness of partial updating, selective expertization, and robustness-constrained expert selection.

Robustness Meets Uncertainty: Evidential Adversarial Training for Robust Selective Classification

Evidential Adversarial Training (EV-AT), which models uncertainty through a Dirichlet distribution and combines an evidence-based loss promoting clean accuracy and reliable uncertainty with a robust evidence-alignment loss matching clean and adversarial predictions in log Dirichlet-parameter space, is proposed.

Balancing clean accuracy and gradient robustness via mask-guided mixup and adaptive label refinement

CONDITION NUMBER-AWARE PRUNING: PRESERVING MATHEMATICAL STABILITY IN SPARSE NEURAL NETWORKS

This paper proposes a condition number-aware pruning framework that explicitly preserves mathematical stability during the pruning process, and significantly improves both standard accuracy and adversarial robustness compared to conventional pruning methods.