Robustness Evaluation Framework

Robustness evaluation frameworks aim to assess the reliability and stability of machine learning models, particularly deep learning models, under various perturbations or adversarial attacks. Current research focuses on developing standardized evaluation methods across different modalities (image, text, etc.), encompassing diverse attack types and incorporating metrics that capture both accuracy and the stability of model outputs or explanations. These frameworks are crucial for building trustworthy AI systems, improving model generalization, and ensuring reliable performance in real-world applications where noisy or manipulated inputs are common.

Papers

October 10, 2023

Revisit Input Perturbation Problems for LLMs: A Unified Robustness Evaluation Framework for Noisy Slot Filling Task
Guanting Dong, Jinxu Zhao, Tingfeng Hui, Daichi Guo, Wenlong Wan, Boqi Feng, Yueyan Qiu, Zhuoma Gongque, Keqing He, Zechen Wang, Weiran Xu
Large Language Model Natural Language Processing Slot Filling Dialog Model Input Perturbation Robustness Evaluation Framework

August 24, 2023

Don't Look into the Sun: Adversarial Solarization Attacks on Image Classifiers
Paul Gavrikov, Janis Keuper
Native Robustness Image Classifier Robustness Evaluation Framework Multi Class Out of Distribution

June 15, 2023

Robustness Analysis on Foundational Segmentation Models
Madeline Chantry Schiappa, Shehreen Azad, Sachidanand VS, Yunhao Ge, Ondrej Miksik, Yogesh S. Rawat, Vibhav Vineet
Native Robustness Multimodal Model Visual Foundation Model Segmentation Foundation Model Enhanced Robustness Robustness Analysis Robustness Evaluation Framework

May 29, 2023

From Adversarial Arms Race to Model-centric Evaluation: Motivating a Unified Automatic Robustness Evaluation Framework
Yangyi Chen, Hongcheng Gao, Ganqu Cui, Lifan Yuan, Dehan Kong, Hanlu Wu, Ning Shi, Bo Yuan, Longtao Huang, Hui Xue, Zhiyuan Liu, Maosong Sun, Heng Ji
Adversarial Attack Robustness Evaluation Robustness Evaluation Framework Model Centric Evaluation

December 19, 2022

TextGrad: Advancing Robustness Evaluation in NLP by Gradient-Driven Optimization
Bairu Hou, Jinghan Jia, Yihua Zhang, Guanhua Zhang, Yang Zhang, Sijia Liu, Shiyu Chang
Adversarial Example Adversarial Robustness NLP Field Adversarial DEfense Robustness Evaluation Gradient Based Optimization Robustness Evaluation Framework

September 5, 2022

"Is your explanation stable?": A Robustness Evaluation Framework for Feature Attribution
Yuyou Gan, Yuhao Mao, Xuhong Zhang, Shouling Ji, Yuwen Pu, Meng Han, Jianwei Yin, Ting Wang
Adversarial Attack Line by Line Explanation Feature Attribution Explanation Guided Robustness Evaluation Framework

Robustness Evaluation Framework

Papers

Revisit Input Perturbation Problems for LLMs: A Unified Robustness Evaluation Framework for Noisy Slot Filling Task

Don't Look into the Sun: Adversarial Solarization Attacks on Image Classifiers

Robustness Analysis on Foundational Segmentation Models

From Adversarial Arms Race to Model-centric Evaluation: Motivating a Unified Automatic Robustness Evaluation Framework

TextGrad: Advancing Robustness Evaluation in NLP by Gradient-Driven Optimization

"Is your explanation stable?": A Robustness Evaluation Framework for Feature Attribution