1 paper
Dmitri Iourovitski, Sanat Sharma, Rakshak Talwar
As content generated by Large Language Model (LLM) has grown exponentially, the ability to accurately identify and fingerprint such text has become increasingly crucial. In this wo…