Abstract
Relation extraction enables the construction of structured knowledge for many downstream applications. While large language models (LLMs) have shown great promise in this domain, most existing methods concentrate on relation classification, which predicts the semantic relation type between a related entity pair. However, we observe that LLMs often struggle to reliably determine whether a relation exists, especially in cases involving complex sentence structures or intricate semantics, which leads to spurious predictions. Such hallucinations can introduce noisy edges in knowledge graphs, compromising the integrity of structured knowledge and downstream reliability. To address these challenges, we propose DEPTH, a framework that integrates Dependency-aware sEntence simPlification and Two-tiered Hierarchical refinement into the relation extraction pipeline. Given a sentence and its candidate entity pairs, DEPTH operates in two stages: (1) the Grounding module extracts relations for each pair by leveraging their shortest dependency path, distilling the sentence into a minimal yet coherent relational context that reduces syntactic noise while preserving key semantics; (2) the Refinement module aggregates all local predictions and revises them based on a holistic understanding of the sentence, correcting omissions and inconsistencies. We further introduce a causality-driven reward model that mitigates reward hacking by disentangling spurious correlations, enabling robust fine-tuning via reinforcement learning with human feedback. Experiments on six benchmarks demonstrate that DEPTH reduces the average hallucination rate to 7.0\% while achieving a 17.2\% improvement in average F1 score over state-of-the-art baselines.
Abstract (translated)
关系抽取使得为许多下游应用构建结构化知识成为可能。尽管大型语言模型（LLM）在该领域展现出巨大潜力，但大多数现有方法主要集中在关系分类上，即预测相关实体对之间的语义关系类型。然而，我们观察到，在涉及复杂句式或细微语义的情况下，LLMs往往难以可靠地确定是否存在某种关系，这会导致错误预测。这些幻觉会在知识图谱中引入噪音边，从而破坏结构化知识的完整性以及下游应用的可靠性。 为了解决这些问题，我们提出了DEPTH框架，该框架将依赖感知句子简化和两级层次细化整合到关系抽取流程中。给定一个句子及其候选实体对，DEPTH分两步操作：（1）定位模块通过利用最短依存路径来提取每个实体对的关系，并将句子精简为一个最小而连贯的关联语境，从而减少句法噪音同时保留关键语义；（2）优化模块汇总所有局部预测结果并基于整个句子的整体理解进行修订，纠正遗漏和不一致之处。我们进一步提出了一种因果驱动的奖励模型，通过分解虚假相关性来防止奖励黑客行为，并支持通过人类反馈强化学习实现稳健微调。 在六个基准上的实验表明，DEPTH将幻觉率平均降低到7.0%，同时相对于最先进的基线方法实现了17.2%的F1分数提升。
URL
https://arxiv.org/abs/2508.14391
