Aspect-Aware Content-Based Recommendations for Mathematical Research Papers

TL;DR

This paper introduces an aspect-aware recommendation system for mathematical research papers, leveraging a novel GNN model that combines textual, citation, and author data, outperforming existing methods and transferring well to machine learning literature.

Contribution

The paper presents the first aspect-aware content-based recommendation datasets for mathematics and proposes AchGNN, a heterogeneous GNN that improves recommendation accuracy by modeling multiple content aspects.

Findings

AchGNN outperforms prior aspect-based methods across all aspects.

The approach transfers effectively to machine learning papers in the Papers with Code dataset.

Ablation studies highlight the importance of aspect supervision, authorship, and graph structure.

Abstract

Content-based research paper recommendation (CbRPR) has seen advances in computer science and biomedicine, but remains unexplored for mathematics, where paper relatedness is more conceptual than explicit textual or citation-based similarity. Mathematics papers may be connected through shared proof techniques, logical implications, or natural generalizations, yet exhibit minimal textual or citation overlap, rendering existing CbRPR ineffective. To address this gap, we first conduct an expert-driven study characterizing mathematical recommendations, revealing that relevance is inherently \textit{aspect}-driven. Grounded in this insight, we introduce GoldRiM (small, expert-annotated) and SilverRiM (large, automatically derived), the first datasets for \textit{aspect}-aware CbRPR in mathematics. Recognizing that LLM embeddings of mathematical content alone yield suboptimal representation, we propose AchGNN, an \textit{aspect}-conditioned heterogeneous GNN that jointly models textual semantics, citation structure, and author lineage. Across GoldRiM and SilverRiM, AchGNN consistently outperforms prior \textit{aspect}-based CbRPR methods, achieving substantial gains across all evaluated \textit{aspects}. We conduct ablation studies to analyze the contributions of individual \textit{aspect} supervision, authorship lineage, and graph-structural signals to AchGNN's performance. To assess domain generality, we further evaluate AchGNN on the \textit{Papers with Code} dataset of machine learning publications, demonstrating that our \textit{aspect}-aware approach effectively transfers beyond mathematics. We deploy our system on the MaRDI platform to help mathematicians with recommendations and release datasets and code publicly for reproducibility.