Unified Multilingual Robustness Evaluation Toolkit for Natural Language Processing

CrossNorm and SelfNorm for Generalization under Distribution Shifts, ICCV 2021

[NeurIPS'21] "AugMax: Adversarial Composition of Random Augmentations for Robust Training" by Haotao Wang, Chaowei Xiao, Jean Kossaifi, Zhiding Yu, Animashree Anandkumar, and Zhangyang Wang.

The Official Implementation of the ICCV-2021 Paper: Semantically Coherent Out-of-Distribution Detection.

Official PyTorch Repository of "Tailoring Self-Supervision for Supervised Learning" (ECCV 2022 Paper)

[NeurIPS 2021] "Drawing Robust Scratch Tickets: Subnetworks with Inborn Robustness Are Found within Randomly Initialized Networks" by Yonggan Fu, Qixuan Yu, Yang Zhang, Shang Wu, Xu Ouyang, David Cox, Yingyan Lin

Noise Injection Techniques provides a comprehensive exploration of methods to make machine learning models more robust to real-world bad data. This repository explains and demonstrates Gaussian noise, dropout, mixup, masking, adversarial noise, and label smoothing, with intuitive explanations, theory, and practical code examples.

This repository is an implementation of https://link.springer.com/chapter/10.1007/978-3-030-72699-7_35 article. it uses evolutionary strategy (NSGA-II algorithm specificially) to configure image filters parameters in order to attack adversarially to a neural network.

Augmentation for CV using frequency shortcuts

Code repository for the paper "Doubly-Trained Adversarial Data Augmentation for Neural Machine Translation" with instructions to reproduce the results. Paper accepted by the conference of the Association for Machine Translation in the Americas (AMTA 2022)

Benchmark LLM jailbreak resilience across providers with standardized tests, adversarial mode, rich analytics, and a clean Web UI.

Quilt: Robust Data Segment Selection against Concept Drifts (AAAI 2024)

🔍 Benchmark jailbreak resilience in LLMs with JailBench for clear insights and improved model defenses against jailbreak attempts.

Production-grade demonstration of AI safety mechanisms: Out-of-Distribution detection and adversarial robustness testing for NLP models using DistilBERT, PyTorch, and TextAttack.

This repository contains the solution for Assignment 1 of the Deep Learning course at the University of Tehran, focusing on image classification, adversarial attacks, and defensive techniques..

Robust Tickets Transfer Better: Drawing More Transferable Subnetworks in Transfer Learning

Rust-based CNN for garbage classification into 6 categories with high accuracy on GPU.

Adversarial testing and robustness evaluation for the Crucible framework

Regime-based evaluation framework for financial NLP stability. Implements chronological cross-validation, semantic drift quantification via Jensen-Shannon divergence, and multi-faceted robustness profiling. Replicates Sun et al.'s (2025) methodology with modular, auditable Python codebase.

Modular and extensible Python framework for applying synthetic inference and controlled perturbations to AI model inputs, labels, and hyperparameters. Evaluate robustness, sensitivity, and stability of algorithms under realistic variations and adverse scenarios.