CacheRAG: New Semantic Caching System Improves LLM-Based Knowledge Graph Question Answering
Researchers have introduced Larch, a learned query optimization framework designed to reduce the computational cost of semantic filters in AI-augmented SQL queries. Semantic operators powered by large language models are increasingly used to query unstructured data like text and images, but their high inference costs make large-scale use prohibitive. Larch addresses this by intelligently ordering filter evaluations, reducing token usage by 3x–19x compared to existing systems.
A team of researchers has published a preprint on arXiv presenting Larch, a framework that optimizes how semantic filters are executed in AI SQL queries—queries that use large language models to analyze unstructured data such as text, images, and video. The core challenge Larch addresses is that semantic operators are computationally expensive and opaque to traditional database query optimizers, which treat them as black boxes. Larch exploits two properties: the high latency of LLM inference leaves room for heavier runtime optimization, and unstructured data is often accompanied by embeddings that enable efficient semantic comparisons. The framework offers two variants: Larch-A2C, which uses an embedding-augmented Gated Graph Neural Network and frames filter ordering as a Markov decision process, and Larch-Sel, which uses supervised learning to predict filter selectivities and applies dynamic programming to find near-optimal evaluation orders. Evaluated on both real-world datasets and synthetic workloads, both variants consistently outperformed existing approaches—Palimpzest and Quest—reducing total token cost overhead by 3x to 19x.
What's missing
The paper is an unreviewed preprint and has not yet undergone peer review. Key open questions include how Larch performs when embeddings are unavailable or of low quality, and whether the token-cost gains translate proportionally to wall-clock latency reductions in production database systems. The paper does not appear to address potential accuracy trade-offs introduced by the optimization—specifically, whether reordering filters could affect query result correctness in edge cases.
What different sources said
- arXiv cs.AICenter
Larch: Learned Query Optimization for Semantic Predicates
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.