Public Model

Public models, pre-trained and publicly available machine learning models, are increasingly central to AI research and applications, serving as both tools and subjects of study. Current research focuses on improving their fairness and equity across diverse downstream tasks, mitigating security vulnerabilities like data exfiltration and transfer attacks through novel defense mechanisms, and optimizing their efficiency for faster and higher-quality generation. This work is crucial for advancing responsible AI development, ensuring the security and reliability of deployed models, and promoting broader accessibility and usability of powerful machine learning tools.

Papers

October 17, 2024

Representing Model Weights with Language using Tree Experts
Eliahu Horwitz, Bar Cavia, Jonathan Kahana, Yedid Hoshen
Human Language Tree Specie Model Weight Public Model Model Tree

June 4, 2024

Building Socially-Equitable Public Models
Yejia Liu, Jianyi Yang, Pengfei Li, Tongxin Li, Shaolei Ren
Fair Model Policy Gradient Algorithm Public Model

November 13, 2023

Transpose Attack: Stealing Datasets with Bidirectional Training
Guy Amit, Mosh Levy, Yisroel Mirsky
Deep Neural Network Data Set Adversary Agent Bidirectional Training Public Model Data Exfiltration

October 26, 2023

PubDef: Defending Against Transfer Attacks From Public Models
Chawin Sitawarin, Jaewon Chang, David Huang, Wesson Altoyan, David Wagner
Adversarial Attack White Box Transfer Attack White Box Adversarial Attack Public Model

October 19, 2023

SecurityNet: Assessing Machine Learning Vulnerabilities on Public Models
Boyang Zhang, Zheng Li, Ziqing Yang, Xinlei He, Michael Backes, Mario Fritz, Yang Zhang
Machine Learning Machine Learning Model Membership Inference Attack Advanced Machine Learning Target Model Public Model

June 15, 2023

OMS-DPM: Optimizing the Model Schedule for Diffusion Probabilistic Models
Enshu Liu, Xuefei Ning, Zinan Lin, Huazhong Yang, Yu Wang
Generative Model Probabilistic Model Diffusion Probabilistic Model Controllable Generation Stable Diffusion Model High Quality Generation Public Model

June 2, 2023

A Modular Test Bed for Reinforcement Learning Incorporation into Industrial Applications
Reuf Kozlica, Georg Schäfer, Simon Hirländer, Stefan Wegenkittl
Reinforcement Learning Simulation Study Use Case Test Bed Industrial Application Real Time Customization Public Model

May 2, 2023

Great Models Think Alike: Improving Model Reliability via Inter-Model Latent Agreement
Ailin Deng, Miao Xiong, Bryan Hooi
Latent Space Prediction Confidence Cross Model Model Reliability Public Model

November 3, 2022

Making Machine Learning Datasets and Models FAIR for HPC: A Methodology and Case Study
Pei-Hung Lin, Chunhua Liao, Winson Chen, Tristan Vanderbruggen, Murali Emani, Hailu Xu
Machine Learning Case Study NCD Method Fair Machine Learning High Performance Computing Fairness Assessment Public Model

September 28, 2022

MLink: Linking Black-Box Models from Multiple Domains for Collaborative Inference
Mu Yuan, Lan Zhang, Zimu Zheng, Yi-Nan Zhang, Xiang-Yang Li
Machine Learning Model Black Box Model Multi Domain Collaborative Inference Model Inference Multi Model Public Model

August 12, 2022

Is Your Model Sensitive? SPeDaC: A New Benchmark for Detecting and Classifying Sensitive Personal Data
Gaia Gambarelli, Aldo Gangemi, Rocco Tripodi
New Benchmark Data Detection Sensitive Data Private Data Personal Information Public Model Information Detection