Author

Jia-Xun Li

1 paper indexed here

We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.

Not the right person? Other researchers publish under this name.

#machine learning Preprint Sep 2026

Robust Detection of LLM-Generated Text under Contamination

Clipping improves robustness in both studies, with gains varying across detectors and contamination settings, and at a target false-positive rate, clipping improves the log-likelihood--log-rank ratio detector's true-positive rate.

Jia-Xun Li, Saptarshi Chakraborty, Ambuj Tewari · 0 citations