Structure-Preserving Graph Contrastive Learning for Mathematical Information Retrieval
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866918379837718528 |
|---|---|
| author | Ku, Chun-Hsi Chen, Hung-Hsuan |
| author_facet | Ku, Chun-Hsi Chen, Hung-Hsuan |
| contents | This paper introduces Variable Substitution as a domain-specific graph augmentation technique for graph contrastive learning (GCL) in the context of searching for mathematical formulas. Standard GCL augmentation techniques often distort the semantic meaning of mathematical formulas, particularly for small and highly structured graphs. Variable Substitution, on the other hand, preserves the core algebraic relationships and formula structure. To demonstrate the effectiveness of our technique, we apply it to a classic GCL-based retrieval model. Experiments show that this straightforward approach significantly improves retrieval performance compared to generic augmentation strategies. We release the code on GitHub.\footnote{https://github.com/lazywulf/formula_ret_aug}. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_08012 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Structure-Preserving Graph Contrastive Learning for Mathematical Information Retrieval Ku, Chun-Hsi Chen, Hung-Hsuan Information Retrieval Digital Libraries This paper introduces Variable Substitution as a domain-specific graph augmentation technique for graph contrastive learning (GCL) in the context of searching for mathematical formulas. Standard GCL augmentation techniques often distort the semantic meaning of mathematical formulas, particularly for small and highly structured graphs. Variable Substitution, on the other hand, preserves the core algebraic relationships and formula structure. To demonstrate the effectiveness of our technique, we apply it to a classic GCL-based retrieval model. Experiments show that this straightforward approach significantly improves retrieval performance compared to generic augmentation strategies. We release the code on GitHub.\footnote{https://github.com/lazywulf/formula_ret_aug}. |
| title | Structure-Preserving Graph Contrastive Learning for Mathematical Information Retrieval |
| topic | Information Retrieval Digital Libraries |
| url | https://arxiv.org/abs/2603.08012 |