Think-on-Graph: Deep and Responsible Reasoning of Large Language Model with Knowledge Graph

By Jiashuo Sun, Chengjin Xu, Lumingyuan Tang, Sai Wang, Chen Lin, Yeyun Gong, Harry Shum, Jian Guo. arXiv.org, 2023

485 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/119601712acbe6ee133a1744f0970190c4195519

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

Citations in the corpus

Graph Retrieval-Augmented Generation: A Survey (2024, unknown): “This method capitalizes on the cohesive capabilities of a unified architecture, enabling the model to seamlessly retrieve relevant information and generate coherent responses within a single framework.”
G-Retriever: Retrieval-Augmented Generation for Textual Graph Understanding and Question Answering (2024, unknown): “Most existing work focuses primarily on the knowledge graph [38, 1, 36, 16].”
Large Language Models on Graphs: A Comprehensive Survey (2023, methodology): “These techniques are studied in [42], [124]–[126], [128], [131], [132].”
Unifying Large Language Models and Knowledge Graphs: A Roadmap (2023, background): “…[40] 2021 DRAGON [44] 2022 HKLM [238] 2023 Synergized Reasoning LARK [45] 2023 Siyuan et al. [46] 2023 KSL [239] 2023 StructGPT [237] 2023 Think-on-graph [240] 2023 To better guide LLMs reason through KGs, OreoLM [177] proposes a Knowledge Interaction Layer (KIL) which is inserted amid LLM layers.”
Reasoning on Graphs: Faithful and Interpretable Large Language Model Reasoning (2023, background): “RR (He et al., 2022) and KD-CoT Wang et al. (2023b) aim to retrieve relevant knowledge from KGs to produce faithful reasoning plans for LLMs. Think-on-Graph (Sun et al., 2024) and KG-Agent (Jiang et al., 2024) treat LLMs as agents to interact with KGs via prompting to get the latest knowledge for re”
Retrieval-Augmented Generation for AI-Generated Content: A Survey (2024, unknown): “Concretely, retrieved assets not only impose an additional velocity on particles for distribution initialization, but also help optimize the 2D diffusion model through low-rank adaptation.”
MARS: Multi-hop Adaptive Retrieval and SPARQL Generation for KGQA (2026, unknown): “Our prompting format draws inspiration from earlier retrieval and reasoning approaches [49,61], yet it has been adapted to suit our specific use case.”
ReMindRAG: Low-Cost LLM-Guided Knowledge Graph Traversal for Efficient RAG (2025, unknown): “By contrast, the LLM-guided knowledge graph traversal approaches demonstrate notable advantages [6, 42, 41], but these methods lead to substantially increased LLM invocation costs and a significant rise in irrelevant noise information.”
A Matryoshka Hierarchical RAG for Efficient Multi-Hop Question Answering (2026, unknown): “Additionally, both families incur online costs through iterative LLM-driven planning [13] and expensive KG traversals [32].”
An Empirical Study of Knowledge Graph-Enhanced RAG for Information Security Compliance (2026, unknown): “Integration of knowledge graphs with large language models has emerged as a promising research direction, with recent work demonstrating that explicit structural representations can improve reasoning, reduce hallucination, and enable more interpretable output [27,28].”
MG2-RAG: Multi-Granularity Graph for Multimodal Retrieval-Augmented Generation (2026, unknown): “Iterative graph exploration [35, 63], planning over structured representations [8, 52], memory-inspired architectures [24, 25], and agentic frameworks [15] demonstrate that explicit relational modeling significantly enhances reasoning depth and inter-pretability.”
RAS: Retrieval-And-Structuring for Knowledge-Intensive LLM Generation (2025, unknown): “Prob.”
ArchRAG: Attributed Community-based Hierarchical Retrieval-Augmented Generation (2025, unknown): “These include using LLMs for KG construction [86], and employing KG to enhance the LLM reasoning [27, 29, 41, 45, 58, 67, 69].”