Overview

Call for Papers

Program Committee

Workshop Program

Submissions

This workshop will explore how ranking, performance assessment, and learning to rank can move beyond the assumption that the relevance of documents in a ranking is independent. The focus will be on three key themes: the effect of redundancy on information retrieval utility (e.g. how to measure redundancy, avoid it in evaluation and when training ranking algorithms); the role of diversity (e.g. for mitigating the risk of misinterpreting ambiguous queries); approaches to set-level evaluation and optimization (e.g. how to create reusable set-level judgments that trade off between evaluating relevance relative to the needs of a specific user versus the needs of a distribution of users).