arXiv cs.CL· Khai Phan Tran, Xue Li·· 11 小时前AI 评分20
Syn2DRE：面向文档级关系抽取的协同双师生框架
Syn2DRE: Synergistic Dual Teacher-Student Framework for Document-level Relation Extraction
AI 导读
Syn2DRE 是一个结合自监督学习与交叉监督的轻量级师生架构半监督框架，用于生成更可靠的伪标签，缓解现有伪标签方法的低效与确认偏差问题。该框架还引入统一的端到端训练策略以进一步提升效率。在多个基准测试上，Syn2DRE 持续优于此前最先进的半监督 DocRE 方法。
正文 · 原文
Abstract:Document-level Relation Extraction (DocRE) aims to identify relations between entities across a document but typically relies on costly large-scale annotations. While pseudo-labelling enables the use of low-quality data, existing methods often suffer from inefficiency and confirmation bias. To address these challenges, we propose Syn2DRE, a semi-supervised framework that combines self-supervised learning with cross-supervision in a lightweight teacher-student architecture to produce more reliable pseudo-labels. Additionally, we introduce a unified end-to-end training strategy to further improve efficiency of the framework. Experiments on multiple benchmarks show that Syn2DRE consistently outperforms prior state-of-the-art semi-supervised DocRE methods.
| Subjects: | Computation and Language (cs.CL); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2611.09342 [cs.CL] |
| (or arXiv:2611.09342v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2611.09342 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Khai Phan Tran [view email]
[v1]
Fri, 6 Nov 2026 11:40:53 UTC (277 KB)
来源：arXiv cs.CL · arxiv.org