Sonar-TS: New Framework for Natural Language Querying of Time Series Databases
Researchers have developed Sonar-TS, a neuro-symbolic framework that enables non-expert users to query large time series databases using natural language, addressing limitations of existing Text-to-SQL methods. The system uses a Search-Then-Verify pipeline that combines SQL queries with Python program generation to identify and verify temporal patterns like shapes and anomalies. The work includes NLQTSBench, the first large-scale benchmark for evaluating natural language queries over time series data at database scale.
Sonar-TS addresses the challenge of Natural Language Querying for Time Series Databases (NLQ4TSDB), which aims to help non-expert users retrieve meaningful events, intervals, and summaries from massive temporal records. The framework employs a neuro-symbolic approach inspired by active sonar: it uses a feature index to identify candidate windows via SQL queries, then generates Python programs to verify candidates against raw signals. This two-stage approach overcomes limitations of traditional Text-to-SQL methods, which were not designed for continuous morphological intents such as shapes or anomalies, and time series models that struggle with ultra-long histories. The researchers also introduced NLQTSBench, the first large-scale benchmark designed specifically for evaluating natural language queries over time series databases at scale. Experiments demonstrate that Sonar-TS effectively handles complex temporal queries where conventional methods fail, establishing both a general framework and evaluation standard for future research in this domain.
What different sources said
- arXiv cs.AICenter
Sonar-TS: Search-Then-Verify Natural Language Querying for Time Series Databases
Related
Genetic Drift, Not Selection, Drives Rapid Feather Color Evolution in Island Bird Radiation
A new study of an island bird radiation found that rapid evolution of feather coloration is driven primarily by genetic drift in small populations rather than sexual or ecological selection. The research integrated whole-genome data with detailed plumage measurements across complete species sampling to test whether signaling trait evolution correlates with speciation rates. The findings suggest that neutral demographic processes play a central role in generating phenotypic diversity during island radiations, challenging assumptions about the mechanisms driving rapid evolution.
New AI Model Improves Prediction of Therapeutic Peptide Function from Protein Sequences
Researchers developed a lightweight CNN classifier that predicts whether peptide sequences have therapeutic properties, trained on a database of 54,655 peptides across 48 functional categories. The model uses a novel negative sampling strategy to reduce false positive rates from over 60% in previous approaches to 2.1%. This advancement could accelerate drug discovery by enabling faster computational screening of peptide candidates before expensive experimental testing.
Study Shows Different Metabolic Stress Models Produce Distinct Effects on Human Neuronal Networks
Researchers tested three common in vitro metabolic stress models on human-derived neuronal networks and found each produced different patterns of neuronal activity and cell damage. The models tested were hypoxia alone, oxygen-glucose deprivation (OGD), and hypoxia combined with glutamate exposure. The findings suggest that choice of experimental model significantly affects results and that combining electrophysiological and structural analyses is important for accurately assessing metabolic stress in stroke research.