AI Paper Atlas

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

On the Salience of Low-Probability Tokens for AI-Generated Text Detection: A Multiscale Uncertainty Perspective

Yikai Guo, Bin Wang, Xilai Fan, Wenjun Ke, Haoran Luo

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
restrict statistics to low-probability token positions (ablation)Uncertainty++ · Reddit black-box, 12 source modelsAUROC: 88.86 -> 95.60Improved
include the local tail log-probability component (ablation)Uncertainty++ · Reddit black-boxAUROC: 63.80 -> 95.60Improved
add Renyi entropy global uncertainty (ablation)Uncertainty++ · Reddit black-boxAUROC: 94.71 -> 95.60Improved
compute Likelihood, LogRank, Fast-DetectGPT only over low-probability tokensLikelihood / LogRank / Fast-DetectGPT · XSum/WritingPrompts/Reddit, Neo-2.7 sourceAUROC: +24.53, +18.97, +4.20 averageImproved
vary Renyi order alpha and weight betaUncertainty++ · Reddit black-boxAUROC: within 1%No clear effect
vary low-probability percentile rhoUncertainty++ · Reddit black-boxAUROCMixed
increase number of conditional samples mUncertainty++ · Reddit black-boxAUROCNo clear effect
use Uncertainty instead of Lastdedetector · Black-box 12 source modelsAUROC: 86.49 -> 88.74Improved
use Uncertainty instead of Lastde on Llama3-8 sourcedetector · Black-box Llama3-8AUROC: 99.60 -> 98.43Hurt
use Uncertainty++ instead of Lastde++detector · White-box 12 source modelsAUROC: 98.93 -> 99.07Improved
use Uncertainty++ against strong GPT-5 paraphrasingdetector · arXiv paraphrased by GPT-5AUROC: 94.08 -> 77.57Hurt
use Uncertainty methods under frequency penalty decodingUncertainty / Uncertainty++ · arXiv GPT-5 decoding variantsAUROC: +7.47 and +11.58 over Lastde and Lastde++Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 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