Model Provenance and Attribution
Research on identifying model origins, verifying model lineage, and ensuring proper attribution of foundation models and their derivatives
Model Provenance and Attribution
Research on Large Language Models in Model Provenance and Attribution
Tracking Model DNA: Securing LLM Supply Chains
Novel framework for verifying AI model origins and derivatives
Securing the ML Supply Chain
Understanding the hidden dependencies and risks in AI ecosystems
Digital Fingerprints in AI Outputs
Revealing the Unique Signatures of Large Language Models
Detecting Copyright Infringement in AI Models
A Novel Approach to Verify if Vision-Language Models Used Copyrighted Content
Reading LLM Fingerprints Through Timing
Novel identification technique uses token timing patterns instead of content analysis
Tracing the Origins of Black-Box LLMs
A novel approach to identify unauthorized LLM usage
Detecting LLM Plagiarism
A novel mathematical approach to identify copied language models
