Search-on-Graph: Iterative Informed Navigation for Large Language Model Reasoning on Knowledge Graphs
By: Jia Ao Sun , Hao Yu , Fabrizio Gotti and more
Potential Business Impact:
Helps computers answer tricky questions using facts.
Large language models (LLMs) have demonstrated impressive reasoning abilities yet remain unreliable on knowledge-intensive, multi-hop questions -- they miss long-tail facts, hallucinate when uncertain, and their internal knowledge lags behind real-world change. Knowledge graphs (KGs) offer a structured source of relational evidence, but existing KGQA methods face fundamental trade-offs: compiling complete SPARQL queries without knowing available relations proves brittle, retrieving large subgraphs introduces noise, and complex agent frameworks with parallel exploration exponentially expand search spaces. To address these limitations, we propose Search-on-Graph (SoG), a simple yet effective framework that enables LLMs to perform iterative informed graph navigation using a single, carefully designed \textsc{Search} function. Rather than pre-planning paths or retrieving large subgraphs, SoG follows an ``observe-then-navigate'' principle: at each step, the LLM examines actual available relations from the current entity before deciding on the next hop. This approach further adapts seamlessly to different KG schemas and handles high-degree nodes through adaptive filtering. Across six KGQA benchmarks spanning Freebase and Wikidata, SoG achieves state-of-the-art performance without fine-tuning. We demonstrate particularly strong gains on Wikidata benchmarks (+16\% improvement over previous best methods) alongside consistent improvements on Freebase benchmarks.
Similar Papers
Guided Navigation in Knowledge-Dense Environments: Structured Semantic Exploration with Guidance Graphs
Computation and Language
Helps computers find facts faster and smarter.
KG-o1: Enhancing Multi-hop Question Answering in Large Language Models via Knowledge Graph Integration
Computation and Language
Helps computers answer questions by connecting facts.
Synthesize-on-Graph: Knowledgeable Synthetic Data Generation for Continue Pre-training of Large Language Models
Computation and Language
Helps computers learn more from less information.