June 11, 2024, 4:13 a.m. | Abe Bohan Hou, Jingyu Zhang, Yichen Wang, Daniel Khashabi, Tianxing He

cs.CR updates on arXiv.org arxiv.org

arXiv:2402.11399v2 Announce Type: replace-cross
Abstract: Recent watermarked generation algorithms inject detectable signatures during language generation to facilitate post-hoc detection. While token-level watermarks are vulnerable to paraphrase attacks, SemStamp (Hou et al., 2023) applies watermark on the semantic representation of sentences and demonstrates promising robustness. SemStamp employs locality-sensitive hashing (LSH) to partition the semantic space with arbitrary hyperplanes, which results in a suboptimal tradeoff between robustness and speed. We propose k-SemStamp, a simple yet effective enhancement of SemStamp, utilizing k-means clustering …

algorithms arxiv attacks clustering cs.cl cs.cr cs.cy cs.lg detection generated hashing hou inject language machine representation robustness semantic sensitive signatures text token vulnerable watermarks

Information Technology Specialist I, LACERA: Information Security Engineer

@ Los Angeles County Employees Retirement Association (LACERA) | Pasadena, CA

Manager Pentest H/F

@ Hifield | Sèvres, France

Information System Security Officer

@ Parsons Corporation | USA VA Chantilly (Client Site)

Vulnerability Analyst, Mid

@ Booz Allen Hamilton | USA, VA, McLean (8283 Greensboro Dr, Hamilton)

SAP Security and Compliance Auditor

@ Bosch Group | Warszawa, Poland

Head of Product Security (Business team)

@ Zalando | Berlin