PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
March 12, 2026Computers3 citationsOpen Access

Grounded Knowledge Graph Extraction via LLMs: An Anchor-Constrained Framework with Provenance Tracking

YYYuzhao YangGCGenlang ChenBHBinhua He

Key Points

  • The research aims to develop a framework for extracting structured triplets from unstructured text while ensuring accuracy and traceability.
  • Proposed the Anchor-Extraction-Verification-Supplement (AEVS) framework.
  • Conducted anchor discovery to identify key elements in the text.
  • Implemented grounded extraction to form triplets linked to the source text.
  • Used restoration-based verification for validating generated triplets.
  • Demonstrated consistent improvements over traditional trained models and LLM-based methods.
  • Ablation studies highlight anchor-based constraints effectively reduce hallucination rates.
  • Quantitative findings include hallucination rates ranging from 0.23% to 20.23% across varied configurations.

Abstract

Knowledge graphs represent real-world facts as structured triplets and underpin a wide range of applications, including question answering, recommendation, and retrieval-augmented generation. Automatically extracting such triplets from unstructured text is essential for scalable knowledge base construction. Traditional extraction methods require task-specific training data and struggle to generalize across domains. Large language models (LLMs) offer an alternative through in-context learning, enabling flexible extraction without fine-tuning. However, LLMs frequently hallucinate—generating plausible triplets unsupported by the source text. The root cause is the lack of provenance: existing methods produce triplets without explicit links to their textual origins, making faithfulness unverifiable. This paper presents Anchor-Extraction-Verification-Supplement (AEVS), a framework that grounds every triplet element to the source text. AEVS operates in three stages: (1) anchor discovery identifies entities, relation phrases, and attribute values with precise positions, forming a constrained extraction vocabulary; (2) grounded extraction generates triplets linked to discovered anchors; and (3) restoration-based verification validates triplets through hierarchical matching, with a coverage-aware supplement ensuring comprehensive extraction. Experiments on WebNLG, REBEL, and Wiki-NRE demonstrate consistent improvements over both trained models and LLM-based baselines. Ablation studies confirm that anchor-based constraints are the primary mechanism for hallucination reduction. Dedicated analyses of anchor discovery quality, computational cost (2.83–4.28 LLM calls per sample), and hallucination rates (0.23–20.23% across model–dataset configurations) provide insights into the framework’s practical applicability and limitations. .

Ask AI
Helpful
Bookmark
Share
View Full Paper

Cite This Study

Yang et al. (2026) studied this question.

synapsesocial.com/papers/69b257fc96eeacc4fcec728fhttps://doi.org/10.3390/computers15030178
Ask AI
Helpful
Bookmark
Share
View Full Paper