StructCoh: Structured Contrastive Learning for Context-Aware Text Semantic Matching
This work improves text semantic matching for domains like legal document analysis and plagiarism detection, though it is incremental as it builds on existing contrastive learning and graph-based methods.
The paper tackled the problem of text semantic matching by addressing the limitations of pre-trained language models in capturing hierarchical structural patterns and subtle semantic distinctions, resulting in a 6.2% absolute gain in F1-score on legal statute matching benchmarks.
Text semantic matching requires nuanced understanding of both structural relationships and fine-grained semantic distinctions. While pre-trained language models excel at capturing token-level interactions, they often overlook hierarchical structural patterns and struggle with subtle semantic discrimination. In this paper, we proposed StructCoh, a graph-enhanced contrastive learning framework that synergistically combines structural reasoning with representation space optimization. Our approach features two key innovations: (1) A dual-graph encoder constructs semantic graphs via dependency parsing and topic modeling, then employs graph isomorphism networks to propagate structural features across syntactic dependencies and cross-document concept nodes. (2) A hierarchical contrastive objective enforces consistency at multiple granularities: node-level contrastive regularization preserves core semantic units, while graph-aware contrastive learning aligns inter-document structural semantics through both explicit and implicit negative sampling strategies. Experiments on three legal document matching benchmarks and academic plagiarism detection datasets demonstrate significant improvements over state-of-the-art methods. Notably, StructCoh achieves 86.7% F1-score (+6.2% absolute gain) on legal statute matching by effectively identifying argument structure similarities.