Text-to-Image Models Leave Identifiable Signatures: Implications for Leaderboard Security
机构 * University of Massachusetts Amherst(马萨诸塞大学阿默斯特分校) ; Northeastern University(东北大学)
Comments Accepted at Lock-LLM Workshop, NeurIPS 2025
高校专区
机构 * University of Massachusetts Amherst(马萨诸塞大学阿默斯特分校) ; Northeastern University(东北大学)
Comments Accepted at Lock-LLM Workshop, NeurIPS 2025
机构 * University of Massachusetts Amherst(马萨诸塞大学阿默斯特分校) ; Adobe Research(Adobe研究院)
Comments Webpage: https://oindrilasaha.github.io/SIGMA-Gen/