AgentMap: Joint Equivalence and Subsumption Discovery for Ontology Matching

Source

• Source name: arXiv
• Source kind: paper
• Authors / organization: Yiping Song, Jiaoyan Chen, Renate Schmidt, Hui Yang, Wen Zhang
• Venue: arXiv: cs.AI
• Published: 2026-07-29
• Collected: 2026-07-30
• URL: http://arxiv.org/abs/2607.27130v1
• PDF: https://arxiv.org/pdf/2607.27130v1
• DOI: UNKNOWN

Plain-Language Brief

Researchers have introduced a new framework for ontology matching called AgentMap. This framework can simultaneously discover both equivalent and subsuming mappings in ontologies, which existing systems cannot do. The framework uses a large language model and a series of interdependent semantic decisions to explore an ontology and identify either the equivalent concept or the most fine-grained subsumer for a given concept.

What the Document Says

• Ontology matching (OM) is traditionally formulated as equivalence discovery or subsumption matching. Existing OM systems can only identify one type of semantic correspondence.
• A new OM task, Hybrid Ontology Matching (HOM), has been introduced that unifies equivalence and subsumption discovery.
• AgentMap is a new OM framework that implements a series of interdependent semantic decisions using a Large Language Model (LLM).
• AgentMap integrates semantic retrieval, hierarchical search, and collaborative multi-agent LLM reasoning to explore an ontology and identify either the equivalent concept or the most fine-grained subsumer.
• Four OM datasets have been extended for a HOM benchmark.
• AgentMap was evaluated under hybrid, equivalence-only, and subsumption-only settings.
• Experimental results show that AgentMap achieves promising performance on the hybrid setting and outperforms both equivalence matching and subsumption matching baselines in their respective settings.

Key Concepts Mentioned

• ontology matching
• equivalence discovery
• subsumption matching
• Hybrid Ontology Matching (HOM)
• AgentMap
• Large Language Model (LLM)
• semantic retrieval
• hierarchical search
• collaborative multi-agent reasoning

Claims Made by the Document

• Existing OM systems can only identify one type of semantic correspondence.
• AgentMap achieves promising performance on the hybrid setting.

Evidence Used

• The text states that existing OM systems can only identify one type of semantic correspondence.
• The text presents experimental results showing AgentMap's performance on the hybrid setting.

Methods / Approach

UNKNOWN

Limitations Stated by the Authors

UNKNOWN

Limitations Noted by the Agent

Agent interpretation — not from the source.

UNKNOWN

Open Questions

UNKNOWN

Notes for Future Q&A

May include interpretation. Labeled where it does.

UNKNOWN

Raw Abstract / Excerpt

Ontology matching (OM) has traditionally been formulated as either equivalence discovery or subsumption matching. The existing OM systems identify only one type of semantic correspondence and cannot simultaneously discover equivalence and subsumption mappings. In this paper, we introduce Hybrid Ontology Matching (HOM), a new OM task that unifies equivalence and subsumption discovery, and accordingly propose a Large Language Model (LLM)-based multi-agent OM framework AgentMap that is implemented by a series of interdependent semantic decisions. Given a concept in the source ontology, AgentMap integrates semantic retrieval, hierarchical search, and collaborative multi-agent LLM reasoning to progressively explore the target ontology, identifying either the equivalent concept, if one exists, or the most fine-grained subsumer. We further extend four OM datasets for a HOM benchmark and evaluate AgentMap under hybrid, equivalence-only, and subsumption-only settings. Experimental results show that AgentMap achieves promising performance on the hybrid setting, and at the same time outperforms equivalence matching and subsumption matching baselines on the equivalence-only and subsumption-only settings, respectively.