FreshRetailNet-50K: New Dataset Addresses Stockout Bias in Retail Demand Forecasting
Researchers have published FreshRetailNet-LT, an open benchmark dataset containing 50,000 store-product time series with hourly sales and stockout annotations drawn from 898 stores across 18 cities. The dataset addresses a longstanding problem in retail AI: when products run out of stock, actual consumer demand goes unrecorded, causing systematic underestimation in forecasting models. By enabling more accurate latent demand recovery, the resource could improve inventory and pricing decisions for perishable goods.
FreshRetailNet-LT (previously FreshRetailNet-50K) is presented as the first large-scale benchmark specifically designed for censored demand estimation in fresh retail, covering 863 perishable SKUs annotated for stockout events at hourly resolution. The core problem it targets is 'censored' sales data: when a product is out of stock, sales records show zero, but true consumer demand remains hidden, introducing systematic bias into any model trained on such data. The dataset spans two years and includes rich contextual covariates such as promotional discounts, precipitation data, and temporal features, enabling more realistic modeling. The authors demonstrate a two-stage approach—first reconstructing latent demand during stockout periods using the precise hourly annotations, then training forecasting models on the recovered demand—achieving a 2.73% improvement in prediction accuracy and reducing systematic demand underestimation from 7.37% to near-zero bias. Both the dataset and accompanying code have been openly released, with the goal of advancing research in demand imputation, perishable inventory optimization, and causal retail analytics.
What's missing
The paper is a preprint and has not yet undergone formal peer review. Key open questions include how well the two-stage modeling approach generalizes to retail contexts outside China (the dataset's geographic scope), and whether the stockout annotations were manually verified or algorithmically generated (and at what error rate). The long-term version (LT) spanning two years is noted only in a comment, with limited methodological detail on how the extended timeframe affects the findings.
What different sources said
- arXiv cs.LGCenter
FreshRetailNet-LT: A Stockout-Annotated Censored Demand Dataset for Latent Demand Recovery and Forecasting in Fresh Retail
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.