CTR-Sink: New Framework Improves Language Models for Click-Through Rate Prediction in Recommendation Systems
Researchers propose CTR-Sink, a framework that addresses how language models struggle with user behavior sequences in click-through rate prediction by introducing attention sink tokens between behaviors. The problem arises because user behavior data lacks the semantic coherence of natural language that language models were trained on, causing attention to scatter ineffectively. The approach could improve recommendation system accuracy by better capturing relationships between user actions.
CTR-Sink introduces behavior-level attention sinks to improve how language models process user behavior sequences for click-through rate prediction in recommendation systems. The core issue is that user behavior sequences consist of discrete actions separated by semantically empty tokens, fundamentally different from the coherent natural language on which language models are pre-trained. This mismatch causes attention mechanisms to scatter across irrelevant tokens rather than focusing on meaningful behavior boundaries and inter-behavior relationships. The proposed solution inserts sink tokens between consecutive behaviors, incorporating recommendation-specific signals like temporal distance to create stable attention anchors. The framework employs a two-stage training strategy to explicitly guide attention toward sink tokens and an attention sink mechanism that amplifies dependencies between sinks to better capture behavioral correlations. Validation on one industrial dataset and two open-source datasets (MovieLens, Kuairec) demonstrates effectiveness across different scenarios.
What different sources said
- arXiv cs.CLCenter
CTR-Sink: Attention Sink for Language Models in Click-Through Rate Prediction
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.