arXiv cs.AI· Chuanliu Fan, Junjie Wu, Nan Yu, Guohong Fu·· 12 小时前AI 评分20
Local-Frame Rotary Attention：用折叠残基几何做注意力位置机制，实现抗体特异性表位预测
Local-Frame Rotary Attention for Antibody-Specific Epitope Prediction
AI 导读
研究者提出 Local-Frame Rotary Attention（LFRA），将残基间位移表示在骨架定义的局部坐标系中并直接注入旋转注意力，在保持连续方向几何的同时对独立 SE(3) 变换不变，用于抗体特异性表位预测。现有方法多依赖 PLM 嵌入并借助图、表面或点云编码器注入结构，注意力内部的位置机制仍基本绑定一维序列顺序。代码与数据集已公开。
正文 · 原文
Abstract:Antibody-specific epitope prediction aims to identify which antigen residues are recognized by a given antibody, a task that depends on the three-dimensional complementarity between antibody CDRs and the antigen surface. Existing methods usually leverage PLM embeddings and inject structure through additional graph, surface, or point-cloud encoders, where the positional mechanism inside attention remains largely tied to one-dimensional sequence order. For proteins, the analogue of a token offset is not only sequence separation, but also the three-dimensional displacement between residues after folding. This raises a question, can folded residue geometry serve as the positional mechanism of attention itself? We propose Local-Frame Rotary Attention (LFRA), which expresses inter-residue displacements in backbone-defined local frames and injects them directly into rotary attention. This design preserves continuous directional geometry while ensuring invariance to independent $\mathrm{SE}(3)$ transformations. The code and dataset are available at this https URL.
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2608.01092 [cs.AI] |
| (or arXiv:2608.01092v2 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2608.01092 arXiv-issued DOI via DataCite |
Submission history
From: Chuanliu Fan [view email]
[v1]
Sun, 2 Aug 2026 08:48:07 UTC (4,231 KB)
[v2]
Thu, 5 Nov 2026 05:35:13 UTC (1,115 KB)
来源：arXiv cs.AI · arxiv.org