arXiv Machine Learning June 16, 2026 By Sumin Cho, Dongwon Kim, Kwangsu Kim

One-Step Generalization Ratio Guided Optimization for Domain Generalization

Read the original on arXiv Machine Learning →

arXiv:2606. 16301v1 Announce Type: new Abstract: Domain Generalization (DG) aims to train models that generalize to unseen target domains but often overfit to domain-specific features, known as undesired correlations.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.

safety

arXiv AI

Aug 12

BPG: Balancing Plasticity and Generalization for Domain Incremental Learning

arXiv:2608. 10804v1 Announce Type: cross Abstract: Deep neural networks excel in various tasks but struggle to generalize across evolving data distributions, leading to significant performance degradation under domain shifts.

By Qiang Wang, Songlin Dong, Shaokun Wang, Jizhou Han, Xiang Song, Chenhao Ding, Yuhang He, Yihong Gong

benchmarks

arXiv Machine Learning

Jul 8

Leveraging Extragradient for Effective Sharpness-Aware Minimization in Deep Learning

arXiv:2607. 06151v1 Announce Type: new Abstract: Generalization remains a pivotal challenge in deep learning, where traditional optimizers like Stochastic Gradient Descent (SGD) often converge to sharp minima, leading to overfitting and reduced performance on unseen data.

By Yao Fu, Chunxia Zhang, Junmin Liu, Yihang Jin, Haishan Ye, Yuanao Yang

benchmarks

arXiv AI

Sep 10

SAEs Can Improve Unlearning: Dynamic Sparse Autoencoder Guardrails for Precision Unlearning in LLMs

The paper introduces Dynamic DAE Guardrails (DSG), a method that uses Dynamic Sparse Autoencoders to perform precision unlearning in large language models. DSG leverages principled feature selection and a dynamic classifier to target activation-based unlearning, outperforming existing gradient‑based methods in terms of computational efficiency, stability, sequential unlearning, resistance to relearning attacks, data efficiency, and interpretability.

By Aashiq Muhamed, Jacopo Bonato, Mona Diab, Virginia Smith

llms safety

arXiv Machine Learning

Aug 11

Domain-Aware Pruning: Sparsity and Domain Generalization via Regularized Probabilistic Masking

arXiv:2608. 08624v1 Announce Type: new Abstract: Domain generalization (DG) and neural network pruning are conventionally treated as distinct objectives, targeting out-of-distribution (OOD) robustness and model efficiency, respectively.

By Parham Sazdar, Mostafa Tavassolipour, Reshad Hosseini

fine-tuning efficiency benchmarks safety

arXiv Machine Learning

Jul 1

A Theory of How Pretraining Shapes Inductive Bias in Fine-Tuning

arXiv:2602. 20062v2 Announce Type: replace Abstract: Pretraining and fine-tuning are central stages in modern machine learning systems.

By Nicolas Anguita, Francesco Locatello, Andrew M. Saxe, Marco Mondelli, Flavia Mancini, Samuel Lippl, Clementine Domine

llms fine-tuning safety

arXiv AI

Jun 2

SORA: Free Second-Order Attacks in Fast Adversarial Training

arXiv:2606. 00738v1 Announce Type: cross Abstract: Adversarial Training (AT) is a leading defense against adversarial examples but often suffers from Catastrophic Overfitting (CO) in efficient single-step variants, where robustness to multi-step attacks collapses despite high single-step performance.

By Mazdak Teymourian, Ramtin Moslemi, Farzan Rahmani, Mohammad Hossein Rohban

benchmarks safety

One email a morning, machine-written

Related stories

BPG: Balancing Plasticity and Generalization for Domain Incremental Learning

Leveraging Extragradient for Effective Sharpness-Aware Minimization in Deep Learning

SAEs Can Improve Unlearning: Dynamic Sparse Autoencoder Guardrails for Precision Unlearning in LLMs

Domain-Aware Pruning: Sparsity and Domain Generalization via Regularized Probabilistic Masking

A Theory of How Pretraining Shapes Inductive Bias in Fine-Tuning

SORA: Free Second-Order Attacks in Fast Adversarial Training