Beyond Single-Score Ranking: Facet-Aware Reranking for Controllable Diversity in Paper Recommendation

TL;DR

SciFACE introduces a facet-aware reranking framework for paper recommendation, enabling controllable diversity by modeling background and method facets separately, with improved accuracy over existing methods.

Contribution

The paper proposes SciFACE, a novel reranking approach that models two independent facets, trained on a small labeled dataset, achieving competitive results with data-efficient learning.

Findings

SciFACE achieves 70.63 NDCG@20 on Background facet.

SciFACE outperforms SPECTER by 5.9 points on Background NDCG@20.

SciFACE improves Method NDCG@20 by 4.1 points over FaBLE.

Abstract

Current paper recommendation systems output a single similarity score that mixes different notions of relatedness, so users cannot specify why papers should be similar. We present SciFACE (Scientific Faceted Cross-Encoder), a reranking framework that models two independent facets: Background (what problem is studied) and Method (how it is solved). SciFACE trains two separate cross-encoders on 5,891 real seed-candidate paper pairs labeled by GPT-4o-mini with facet-specific criteria and validated against human judgments. On CSFCube, SciFACE reaches 70.63 NDCG@20 on Background (5.9 points above SPECTER) and 49.06 NDCG@20 on Method (31.1 points above SPECTER), competitive with state-of-the-art results. Compared with FaBLE without citation pre-training, SciFACE improves Method NDCG@20 by 4.1 points while using 5,891 labeled pairs versus 40K synthetic augmentations. These results show that high-quality grounded facet labels can be more data-efficient than large-scale synthetic augmentation for learning fine-grained scientific similarity.