Saved in:
Bibliographic Details
Main Author: Tao, Duan Ming
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2604.16329
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913041605459968
author Tao, Duan Ming
author_facet Tao, Duan Ming
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2604_16329
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Beyond Single-Score Ranking: Facet-Aware Reranking for Controllable Diversity in Paper Recommendation
Tao, Duan Ming
Information Retrieval
Artificial Intelligence
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.
title Beyond Single-Score Ranking: Facet-Aware Reranking for Controllable Diversity in Paper Recommendation
topic Information Retrieval
Artificial Intelligence
url https://arxiv.org/abs/2604.16329