AI Paper Atlas

Research map / Large language models / Language, safety and interpretability

Black-Box Detection of LLM-Generated Text Using Generalized Jensen Shannon Divergence

Shuangyi Chen, Ashish J Khisti

ICML 2026 · Large language models › Language, safety and interpretability · approach: Zero-shot machine-text detection · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
summarize surprisal dynamics with a GJS gap against human and machine references (SurpMark)proxy-LM black-box detection · closed-source generators (GPT-5-Chat, GPT-4.1-mini, Gemini-1.5-Flash)AUROC (avg): 69.55 (best baseline R-Detect) -> 78.32Improved
SurpMark vs best baseline on open-source generatorsproxy-LM black-box detection · 9 open-source generatorsAUROC (avg): 90.04 (Lastde++) -> 91.23Improved
use k-means instead of equal-width binning (ablation)SurpMark discretization · WritingPrompts@GPT5-chat (k=7)AUROC: 62.90 -> 82.05Improved
use k-means instead of equal-mass binning (ablation)SurpMark discretization · XSum@GPT-4.1-mini (k=7)AUROC: 71.40 -> 80.42Improved
use k-means instead of equal-mass binning (ablation)SurpMark discretization · WritingPrompts@GPT5-chat (k=7)AUROC: 84.42 (equal-mass) vs 82.05 (k-means)No clear effect
increase number of bins k beyond 6-7 (ablation)SurpMark · GPT-J-6B and Llama-3.2-3BAUROCNo clear effect
use higher-order Markov chain instead of first-order (ablation)SurpMark · SQuAD and appendix datasetsAUROCHurt
second-order instead of first-order surprisal-state model (ablation)discretized surprisal states · GPT-5-chat textperplexity reduction: +0.528% relative perplexity gainNo clear effect
increase number of reference samples up to 100 (ablation)SurpMark · reference size sweepAUROCImproved
increase test text length from 50 to about 150-200 tokens (ablation)SurpMark · test length sweepAUROCImproved
use references from a different domain (ablation)SurpMark · DetectRL YelpAUROC: 90.18 -> 85.48 (arXiv-ref)Hurt
use a stronger proxy LM (ablation)SurpMark · proxy model sweepAUROCNo clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy