Sonar-TS: New Framework for Natural Language Querying of Time Series Databases
Researchers have proposed Sonar-TS, a neuro-symbolic system that allows non-expert users to query large time series databases using natural language, accepted at ICML 2026. The framework uses a two-stage 'Search-Then-Verify' pipeline combining SQL-based candidate retrieval with Python program verification against raw signals. It addresses a gap left by existing Text-to-SQL and time series models, which individually struggle with continuous morphological queries or ultra-long data histories.
Sonar-TS is a neuro-symbolic framework designed to tackle Natural Language Querying for Time Series Databases (NLQ4TSDB), a problem where users want to retrieve events, intervals, or summaries from large temporal records using plain language. Existing approaches fall short: Text-to-SQL methods are not built for shape- or anomaly-based queries, while dedicated time series models cannot efficiently handle very long historical records. The system works analogously to active sonar — it first uses a feature index to identify candidate time windows via SQL, then applies generated Python programs to verify those candidates against the raw signal data. Alongside the framework, the authors introduce NLQTSBench, described as the first large-scale benchmark specifically designed for evaluating NLQ over database-scale time series histories. Experiments reported in the paper show Sonar-TS outperforms traditional methods on complex temporal queries. The work was accepted at ICML 2026 and represents what the authors claim is the first systematic study of the NLQ4TSDB problem, providing both a general framework and an evaluation standard for future research.
What's missing
The paper does not detail known limitations such as how Sonar-TS performs on highly irregular or multivariate time series, its computational cost at scale, or how the Python program generation step handles errors or adversarial inputs. The generalizability of NLQTSBench to real-world industrial time series domains beyond those tested also remains an open question.
What different sources said
- arXiv cs.AICenter
Sonar-TS: Search-Then-Verify Natural Language Querying for Time Series Databases
Related
Gut Bacteria Enzyme Found to Break Down Heat-Processed Food Compounds, Producing Novel Biogenic Amines
Researchers have discovered that an enzyme in common gut bacteria can degrade N-epsilon-carboxymethyllysine (CML), a compound formed during thermal food processing, producing previously unknown biogenic amines. The enzyme, ornithine decarboxylase SpeC from enterobacteria, acts on CML and related modified lysine derivatives through a low-level 'underground' catalytic activity. This finding suggests a previously unrecognized communication axis between thermally processed dietary compounds and gut microbial physiology, with potential implications for host health.
Full-Length Gene Sequencing Reveals Two Distinct Bacterial Communities in Black-Legged Ticks Expanding Into Canada
Researchers used Oxford Nanopore full-length 16S rRNA gene sequencing to characterize the microbiome of Ixodes scapularis black-legged ticks collected in Nova Scotia, Canada, distinguishing between tick-adapted bacteria and environmentally acquired bacteria. The study comes as I. scapularis — the primary vector of Lyme disease — is rapidly expanding northward into Canada due to climate change. The findings suggest that environmentally derived bacteria in tick microbiomes are not mere contamination, which has implications for how tick microbiome data is collected and interpreted across surveillance studies.
Study Identifies Metabolic Link Between Cell Envelope Stress and Biofilm Formation in Bacteria
Researchers have discovered that the metabolite acetyl-CoA directly inhibits enzymes that degrade the bacterial signaling molecule c-di-GMP, connecting cell envelope biosynthesis stress to biofilm formation in Pseudomonas aeruginosa. The study found that sub-inhibitory concentrations of antibiotics targeting early peptidoglycan biosynthesis — but not other antibiotic classes — elevate c-di-GMP levels by reducing phosphodiesterase activity, with acetyl-CoA competing for the enzyme active site. Because the relevant enzyme domain is broadly conserved across bacterial species, this checkpoint mechanism may be widespread and could have implications for understanding antibiotic-induced biofilm responses.