AI Paper Atlas

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

DMAP: A Distribution Map for Text

Tom Kempton, Julia Rozanova, Parameswaran Kamalaruban, Maeve Madigan, Karolina Wresilo, Yoann Launay, David Sutton, Stuart Burrell

ICLR 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
switch generated text from top-k=50 to pure samplingFast-DetectGPT (Llama-3.1-8B) · XSum black-box detectionAUROC: 0.702 -> 0.200Hurt
switch from top-k=50 to pure samplingDetectGPT (Llama-3.1-8B) · XSum black-box detectionAUROC: 0.606 -> 0.408Hurt
switch from top-k=50 to pure samplingBinoculars (Llama-3.1-8B) · XSum black-box detectionAUROC: 0.825 -> 0.325Hurt
switch from top-k=50 to pure samplingFast-DetectGPT (Llama-3.1-8B) · WritingPromptsAUROC: 0.915 -> 0.289Hurt
fine-tune Pythia on temperature-sampled synthetic responses (vs human data)Pythia 1B · DMAP head biasDMAP shapeHurt

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