Generative Subgraph Retrieval for Knowledge Graph-Grounded Dialog Generation
Knowledge graph-grounded dialog generation requires retrieving a dialog-relevant subgraph from the given knowledge base graph and integrating it with the dialog history. Previous works typically represent the graph using an external encoder, such as graph neural networks, and retrieve relevant triplets based on the similarity between single-vector representations of triplets and the dialog history. However, these external encoders fail to leverage the rich knowledge of pretrained language models, and the retrieval process is also suboptimal due to the information bottleneck caused by the single-vector abstraction of the dialog history. In this work, we propose Dialog generation with Generative Subgraph Retrieval (DialogGSR), which retrieves relevant knowledge subgraphs by directly generating their token sequences on top of language models. For effective generative subgraph retrieval, we introduce two key methods: (i) structure-aware knowledge graph linearization with self-supervised graph-specific tokens and (ii) graph-constrained decoding utilizing graph structural proximity-based entity informativeness scores for valid and relevant generative retrieval. DialogGSR achieves state-of-the-art performance in knowledge graph-grounded dialog generation, as demonstrated on OpenDialKG and KOMODIS datasets.
Code (1)
Tasks
InformativenessRetrievalvalidMethods 이 논문이 사용한 방법론
Similar Papers 제목 키워드 기반
LLM-based Discriminative Reasoning for Knowledge Graph Question Answering
Large language models (LLMs) based on generative pre-trained Transformer have achieved remarkable performance on knowledge graph question-answering (KGQA) tasks. However, LLMs often produce ungrounded subgraph planning o…
Graph Question AnsweringQuestion AnsweringKnowledge Graph-Augmented Language Models for Knowledge-Grounded Dialogue Generation
Language models have achieved impressive performances on dialogue generation tasks. However, when generating responses for a conversation that requires factual knowledge, they are far from perfect, due to an absence of m…
Contrastive LearningDialogue GenerationKnowledge GraphsRetrieval+2Less is More: Making Smaller Language Models Competent Subgraph Retrievers for Multi-hop KGQA
Retrieval-Augmented Generation (RAG) is widely used to inject external non-parametric knowledge into large language models (LLMs). Recent works suggest that Knowledge Graphs (KGs) contain valuable external knowledge for …
Knowledge GraphsRAGRetrievalRetrieval-augmented Generation+1SRTK: A Toolkit for Semantic-relevant Subgraph Retrieval
Information retrieval based knowledge base question answering (KBQA) first retrieves a subgraph to reduce search space, then reasons on the subgraph to select answer entities. Existing approaches have three issues that i…
Entity LinkingInformation RetrievalKnowledge Base Question AnsweringKnowledge Graphs+2Structure-Grounded Knowledge Retrieval via Code Dependencies for Multi-Step Data Reasoning
Selecting the right knowledge is critical when using large language models (LLMs) to solve domain-specific data analysis tasks. However, most retrieval-augmented approaches rely primarily on lexical or embedding similari…
Code Generation