AI Paper Atlas

Research map / Time series / Foundation models for time series

Sonar-TS: Search-Then-Verify Natural Language Querying for Time Series Databases

Zhao Tan, Yiji Zhao, Shiyu Wang, Chang Xu, Yuxuan Liang, Xiping Liu, Shirui Pan, Ming Jin

ICML 2026 · Time series › Foundation models for time series · approach: Time-series language reasoning · 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
add self-correction loop (ablation)Sonar-TS · NLQTSBench Sliding Windowscore: 0.7250 -> 0.7827Improved
add self-correction loop (ablation)Sonar-TS · NLQTSBench Averagescore: 0.5930 -> 0.6144Improved
add feature tables (SAX index) (ablation)Sonar-TS · NLQTSBench Shape Identificationscore: 0.1561 -> 0.3336Improved
add feature tables (SAX index) (ablation)Sonar-TS · NLQTSBench Composite Trendscore: 0.0422 -> 0.2988Improved
add feature tables (SAX index) (ablation)Sonar-TS · NLQTSBench Periodicity Detectionscore: 0.8804 -> 0.8625Hurt
add experiences (prompt cold start) (ablation)Sonar-TS · NLQTSBench Contextual Anomalyscore: 0.2501 -> 0.4767Improved
add experiences (prompt cold start) (ablation)Sonar-TS · NLQTSBench Insight Synthesisscore: 0.3415 -> 0.7395Improved
add experiences (prompt cold start) (ablation)Sonar-TS · NLQTSBench Averagescore: 0.4686 -> 0.6144Improved
add Python verification (ablation)Sonar-TS · NLQTSBench Periodicity Detectionscore: 0.0154 -> 0.8625Improved
add Python verification (ablation)Sonar-TS · NLQTSBench Subsequence Matchingscore: 0.0396 -> 0.9439Improved
add Python verification (ablation)Sonar-TS · NLQTSBench Averagescore: 0.2721 -> 0.6144Improved
add Python verification (ablation)Sonar-TS · NLQTSBench Atomic Retrievalscore: 0.8294 -> 0.8609No 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