TDGI: Translation-Guided Double-Graph Inference for Document-Level Relation Extraction

• Published in: IEEE transactions on pattern analysis and machine intelligence 2025-01, p.1-13
• Main Authors: Zhang, Lingling, Zhong, Yujie, Zheng, Qinghua, Liu, Jun, Wang, Qianying, Wang, Jiaxin, Chang, Xiaojun
• Format: Article
• Language: English
• Subjects: Cognition; Context modeling; Data mining; Document-Level relation extraction; double-graph inference; Electronic mail; relation direction; relation semantics; Robustness; Semantics; Training; Transformers; Translation; translation rule; Vectors
• ISSN: 0162-8828; 2160-9292
• DOI: 10.1109/TPAMI.2025.3528246

Document-level relation extraction (DocRE) aims at predicting relations of all entity pairs in one document, which plays an important role in information extraction. DocRE is more challenging than previous sentence-level relation extraction, as it often requires coreference and logical reasoning across multiple sentences. Graph-based methods are the mainstream solution to this complex reasoning in DocRE. They generally construct the heterogeneous graphs with entities, mentions, and sentences as nodes, co-occurrence and co-reference relations as edges. Their performance is difficult to further break through because the semantics and direction of the relation are not jointly considered in graph inference process. To this end, we propose a novel translation-guided double-graph inference network named TDGI for DocRE. On one hand, TDGI includes two relation semantics-aware and direction-aware reasoning graphs, i.e. mention graph and entity graph, to mine relations among long-distance entities more explicitly. Each graph consists of three elements: vectorized nodes, edges, and direction weights. On the other hand, we devise an interesting translation-based graph updating strategy that guides the embeddings of mention/entity nodes, relation edges, and direction weights following the specific translation algebraic structure, thereby to enhance the reasoning skills of TDGI. In the training procedure of TDGI, we minimize the relation multi-classification loss and triple contrastive loss together to guarantee the model's stability and robustness. Comprehensive experiments on three widely-used datasets show that TDGI achieves outstanding performance comparing with state-of-the-art baselines.