RESEARCHInvestigateNEXT 12 MONTHS
Ranked by Position: Order Sensitivity as an Exploitable Attack Surface in LLM Listwise Recommenders
arXiv cs.CL — Computation and Language
Factual evidence
What the source reports
Research reveals LLM-based recommender systems are vulnerable to position bias, enabling attackers to promote items by reordering candidates.
Open sourceOneBench interpretation
Institutional assessment
So what
Positional bias in LLM rerankers creates an attack surface for manipulating recommendation outputs in financial product surfacing.
Do what
Add `promo@k` metrics to your model validation framework for LLM-based recommender systems.