Native Robustness

Native robustness in machine learning focuses on developing models inherently resistant to various forms of input perturbations, including adversarial attacks and noisy data, without relying solely on post-hoc defenses. Current research emphasizes techniques like ensemble methods, reprogramming existing models, and modifying training procedures (e.g., using different learning rates for specific model layers or incorporating regularization methods) to improve robustness across diverse model architectures, including convolutional neural networks, vision transformers, and large language models. This field is crucial for deploying reliable AI systems in safety-critical applications, such as healthcare and autonomous driving, where model resilience to unexpected inputs is paramount.

Papers

August 8, 2024

Enhancing Robustness of Retrieval-Augmented Language Models with In-Context Learning
Seong-Il Park, Seung-Woo Choi, Na-Hyun Kim, Jay-Yoon Lee
Native Robustness Context Learning Open Domain Question Answering Machine Reading Comprehension Retrieval Augmented Language Model Open Domain Question Answering Datasets Retrieval Head

August 6, 2024

RHiOTS: A Framework for Evaluating Hierarchical Time Series Forecasting Algorithms
Luis Roque, Carlos Soares, Luís Torgo
Native Robustness New Framework Hierarchical Time Series Forecasting Algorithm Hierarchical Time Series Forecasting

August 5, 2024

August 4, 2024

A Novel Metric for Measuring the Robustness of Large Language Models in Non-adversarial Scenarios
Samuel Ackerman, Ella Rabinovich, Eitan Farchi, Ateret Anaby-Tavor
Native Robustness Benchmark Datasets Novel Metric Adversarial Information

August 2, 2024

July 29, 2024

Revisiting the robustness of post-hoc interpretability methods
Jiawen Wei, Hugues Turbé, Gianmarco Mengaldo
Native Robustness Post Hoc Interpretability Hoc Interpretability

July 26, 2024

The Cross-environment Hyperparameter Setting Benchmark for Reinforcement Learning
Andrew Patterson, Samuel Neumann, Raksha Kumaraswamy, Martha White, Adam White
Reinforcement Learning Native Robustness New Benchmark Continuous Environment Hyperparameter Configuration

July 25, 2024

July 24, 2024

Revising the Problem of Partial Labels from the Perspective of CNNs' Robustness
Xin Zhang, Yuqi Song, Wyatt McCurdy, Xiaofeng Wang, Fei Zuo
Convolutional Neural Network Native Robustness Multi Label Visual Perspective Pseudo Labeling Right Problem Partial Label

July 23, 2024

On ADMM in Heterogeneous Federated Learning: Personalization, Robustness, and Fairness
Shengkun Zhu, Jinshan Zeng, Sheng Wang, Yuan Sun, Xiaodong Li, Yuan Yao, Zhiyong Peng
Native Robustness Procedural Fairness Heterogeneous Data Heterogeneous Federated Learning Alternating Direction Method of Multiplier Statistical Heterogeneity Superlinear Convergence Personalized Fl

July 22, 2024

July 20, 2024

Native Robustness

Papers

Enhancing Robustness of Retrieval-Augmented Language Models with In-Context Learning

RHiOTS: A Framework for Evaluating Hierarchical Time Series Forecasting Algorithms

RCDM: Enabling Robustness for Conditional Diffusion Model

On the Robustness of Malware Detectors to Adversarial Samples

A Novel Metric for Measuring the Robustness of Large Language Models in Non-adversarial Scenarios

Coalitions of Large Language Models Increase the Robustness of AI Agents

Assessing Robustness of Machine Learning Models using Covariate Perturbations

Certifiably Robust Encoding Schemes

Revisiting the robustness of post-hoc interpretability methods

The Cross-environment Hyperparameter Setting Benchmark for Reinforcement Learning

Adversarial Robust Decision Transformer: Enhancing Robustness of RvS via Minimax Returns-to-go

Sparse vs Contiguous Adversarial Pixel Perturbations in Multimodal Models: An Empirical Analysis

Revising the Problem of Partial Labels from the Perspective of CNNs' Robustness

On ADMM in Heterogeneous Federated Learning: Personalization, Robustness, and Fairness

Robustness of Speech Separation Models for Similar-pitch Speakers

Targeted Latent Adversarial Training Improves Robustness to Persistent Harmful Behaviors in LLMs

Increasing the Robustness of Model Predictions to Missing Sensors in Earth Observation

Craft: Cross-modal Aligned Features Improve Robustness of Prompt Tuning

Out of spuriousity: Improving robustness to spurious correlations without group annotations

ARoFace: Alignment Robustness to Improve Low-Quality Face Recognition