2 papers
cs.IR2026
When Do Anchor-Based Pointwise LLM Rerankers Help? Retriever Quality, Statistical Scope, and Anchor Design
Utshab Kumar Ghosh, Shubham Chatterjee
Anchor-based pointwise LLM reranking scores each candidate against a shared reference passage to recover cross-document context at pointwise cost. We study when this actually helps…
cs.IT2026
Two-Sided Bounds for Entropic Optimal Transport via a Rate-Distortion Integral
Jingbo Liu
We show that the maximum expected inner product between a random vector and the standard normal vector over all couplings subject to a mutual information constraint or regularizati…