AI Paper Atlas

Research map / Time series / Anomaly detection

GAPS: Gradient-Aware Adaptation-Gap Scoring for Time-Series Anomaly Detection with Foundation Models

Jongwon Kim, Byunghun Song, Young Myoung Ko

NeurIPS 2026 · Time series › Anomaly detection · approach: Time series anomaly detection evaluation

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes GAPS, a gradient-aware adaptation-gap score for unsupervised time-series anomaly detection with foundation models, aiming to avoid reconstruction error's sensitivity to input noise.

Problem

Reconstruction error, the standard unsupervised time-series anomaly detection signal used by time-series foundation models, tracks input noise level rather than true anomalousness, especially where the noise envelope itself is informative.

Approach

Scores anomalies using a gradient-aware adaptation gap of the foundation model rather than raw reconstruction error, so the score is less driven by noise level. Details beyond the abstract are not available.

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