STORM: New Framework Improves Lexical Search by Optimizing Query Expansion with Retrieval Rewards
Researchers introduced STORM, a self-supervised framework that improves lexical query expansion for information retrieval by using retrieval metrics to guide token-level optimization during generation. The method trains smaller language models (0.6B-8B parameters) to rewrite search queries more effectively by scoring candidate expansions against a BM25 index and pruning low-reward continuations. STORM achieves competitive performance with larger proprietary models while maintaining the speed and transparency advantages of traditional lexical search, and demonstrates zero-shot transfer across 18 languages.
STORM (Stepwise Token Optimization with Reward-guided beaM search) addresses a key limitation in modern information retrieval: dense neural models require expensive corpus indexing that must be rebuilt when models change, while traditional lexical retrievers like BM25 remain efficient but suffer from vocabulary mismatch problems. The framework uses retrieval metrics to provide token-level supervision during query rewriting, allowing smaller language models to learn which vocabulary expansions are most effective for retrieval. By scoring candidate expansions at each generation step and pruning low-reward continuations, STORM converts delayed sequence-level retrieval signals into immediate token-level guidance. Across standard benchmarks (TREC DL and BEIR), the approach enables smaller models to match or exceed larger proprietary rewriters while maintaining BM25's speed advantages. The method also demonstrates strong zero-shot multilingual transfer on 18 languages (MIRACL benchmark), outperforming dedicated multilingual dense retrievers on average.
What's missing
The paper does not discuss computational costs or training time comparisons with baseline methods, nor does it address potential limitations in handling domain-specific retrieval tasks or queries with rare vocabulary.
What different sources said
- arXiv cs.AICenter
STORM: Stepwise Token Optimization with Reward-Guided Beam Search
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.