Trusted-System-Lab

Trusted-System-Lab's repositories

WraLU

WraLU is an artifact for the paper "ReLU Hull Approximation" (POPL'24), which provides a sound but incomplete neural network verifier by over-approximating ReLU function hull.

Language:PythonApache-2.0400

WraAct is an artifact for the paper "Convex Hull Approximation for Activation Functions" (OOPSLA'25), which provides a sound but incomplete neural network verifiers by over-approximating the function hulls of various activation functions (including leaky ReLU, ReLU, sigmoid, tanh, and maxpool).

Language:PythonMIT100

PdD

PdD is an artifact for the paper "Formalizing Robustness Against Character-Level Perturbations for Neural Network Language Models" (ICFEM'23), which provides a tool for adversarial training by generating character-level perturbation

Language:PythonMIT200

UQ-Trust-Lab.github.io

Language:HTML100

GPTs-ThirdSpy

An automated framework designed to extract GPTs’ pri- vacy settings

000

PFCon

This is the code repo of PFCon

Language:Python000

AIM

Language:PythonApache-2.0300

GPTs-Filtor

GPTs-Filtor leverage the unique characteristics of GPTs’ deployment, to conduct in-depth analysis and detection of file knowledge leakage at both user interaction and network transmission levels

MIT000