AI Paper Atlas

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

Telescope: Improving Zero Shot Detection of LLM Generated Content By Measuring Token Repetition Probability

Christopher Nassif, Joshua F. Cooper

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
use only bigram context instead of full context (ablation)Telescope Perplexity · HC3AUC: 0.995 -> 0.897Hurt
use only bigram context instead of full context (ablation)Perplexity · HC3AUC: 0.991 -> 0.761Hurt
use trigram context instead of full context (ablation)Telescope Perplexity · HC3AUC: 0.995 -> 0.921Hurt
score the current token given itself (Telescope) instead of next-token perplexitySmolLM 360M detector · Detect LLM TextAUROC: 0.89307 -> 0.99219Improved
score the current token (Telescope) instead of next-token perplexitySmolLM 360M detector · ESL GPT4o MiniAUROC: 0.82523 -> 0.99983Improved
Telescope instead of Binocularszero-shot detector · GB News ChatGPTAUROC: 0.98773 -> 0.90480Hurt
Telescope instead of Perplexityzero-shot detector · GB Essay ChatGPTAUROC: 0.99810 -> 0.98628Hurt
Telescope instead of Binocularszero-shot detector · HC3 PlusAUROC: 0.88510 -> 0.98451Improved
raise target sampling temperature from 0 to 1.2 (ablation)Telescope Perplexity · GB Essay GPT-4oAUROC: 0.999992 -> 0.9988Hurt
adversarial prompt asking target model to repeat key words (ablation)Telescope Perplexity · GB Essay GPT-4oAUROC: 0.99993 -> 0.996Hurt
evaluate Telescope on its own training data split instead of PerplexitySmolLM 360M · FineWeb vs CosmopediaAUC: 0.9523 -> 0.9956Improved
Telescope instead of Binoculars on Russian textGemma2-2B · M4 RussianAUC: 0.89772 -> 0.66543Hurt

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