6 papers
Anchor-ECC: Local Integrity Checking for Watermarked LLM Outputs via Error-Correcting Codes
Zewei Deng, Muhammad Siddeek, Liyan Xie +4
LLM watermarking has become an effective approach to distinguishing AI-generated text from human-written text by embedding detectable patterns during generation. However, a small p…
Score-Based Change-Point Detection and Region Localization for Spatio-Temporal Point Processes
Wenbin Zhou, Liyan Xie, Shixiang Zhu
We study sequential change-point detection for spatio-temporal point processes, where actionable detection requires not only identifying when a distributional change occurs but als…
Detecting Post-generation Edits to Watermarked LLM Outputs via Combinatorial Watermarking
Liyan Xie, Muhammad Siddeek, Mohamed Seif +2
Watermarking has become a key technique for proprietary language models, enabling the distinction between AI-generated and human-written text. However, in many real-world scenarios…
Sequential Change Detection in Correlation Structures with Window-Limited Statistics
Jie Gao, Liyan Xie, Zhaoyuan Li
We consider detecting change points in the correlation structure of streaming data with minimum assumptions posed on the underlying data distribution. Detection statistics are cons…
Sequential Change Point Detection via Denoising Score Matching
Wenbin Zhou, Liyan Xie, Zhigang Peng +1
Sequential change-point detection plays a critical role in numerous real-world applications, where timely identification of distributional shifts can greatly mitigate adverse outco…
Recurrent Neural Goodness-of-Fit Test for Time Series
Aoran Zhang, Wenbin Zhou, Liyan Xie +1
Time series data are crucial across diverse domains such as finance and healthcare, where accurate forecasting and decision-making rely on advanced modeling techniques. While gener…