Graph-enhanced RAG
using knowledge graphs in RAG for grounding LLM results
Paper • 2311.07509 • Published • 2Note Excellent analysis of lift from RAG vs. Graph-enhanced RAG on the correctness of *result sets* of SQL query generation (not the queries themselves), e.g., when there's an intermediary graph query used to enhance the SQL generated.
Increasing the LLM Accuracy for Question Answering: Ontologies to the Rescue!
Paper • 2405.11706 • Published • 2Note "Using the chat with the data benchmark, our primary finding is that our approach increases the overall accuracy to 72% including an additional 8% of "I don't know" unknown results. Thus, the overall error rate is 20%. These results provide further evidence that investing knowledge graphs, namely the ontology, provides higher accuracy for LLM powered question answering systems."
Retrieval-Augmented Generation with Knowledge Graphs for Customer Service Question Answering
Paper • 2404.17723 • Published • 2Note "Empirical assessments on our benchmark datasets, utilizing key retrieval (MRR, Recall@K, NDCG@K) and text generation (BLEU, ROUGE, METEOR) metrics, reveal that our method outperforms the baseline by 77.6% in MRR and by 0.32 in BLEU. Our method has been deployed within LinkedIn’s customer service team for approximately six months and has reduced the median per-issue resolution time by 28.6%."
TRACE the Evidence: Constructing Knowledge-Grounded Reasoning Chains for Retrieval-Augmented Generation
Paper • 2406.11460 • Published • 2Note "Experimental results on three multi-hop QA datasets show that TRACE achieves an average performance improvement of up to 14.03% compared to using all the retrieved documents. Moreover, the results indicate that using reasoning chains as context, rather than the entire documents, is often sufficient to correctly answer questions."
UniOQA: A Unified Framework for Knowledge Graph Question Answering with Large Language Models
Paper • 2406.02110 • Published • 2Note "Experimental findings illustrate that UniOQA notably advances SpCQL Logical Accuracy to 21.2% and Execution Accuracy to 54.9%, achieving the new state-of-the-art results on this benchmark."
From Local to Global: A Graph RAG Approach to Query-Focused Summarization
Paper • 2404.16130 • Published • 4Note see impl: https://github.com/tomasonjo/blogs/blob/master/llm/ms_graphrag.ipynb
DiffKG: Knowledge Graph Diffusion Model for Recommendation
Paper • 2312.16890 • Published • 2Note "graph diffusion" as in generative diffusion (training on ablated sequences), which is super interesting -- not the diffusion in the sense of graph embeddings and Gr
Graph Retrieval-Augmented Generation: A Survey
Paper • 2408.08921 • Published • 4
Mix-of-Granularity: Optimize the Chunking Granularity for Retrieval-Augmented Generation
Paper • 2406.00456 • Published • 2Note "Extensive experiments demonstrate that both MoG and MoGG effectively predict optimal granularity levels, significantly enhancing the performance of the RAG system in downstream tasks"
Don't Forget to Connect! Improving RAG with Graph-based Reranking
Paper • 2405.18414 • Published • 1Note "We introduce G-RAG, a reranker based on graph neural networks (GNNs) between the retriever and reader in RAG. Our method combines both connections between documents and semantic information (via Abstract Meaning Representation graphs) to provide a context-informed ranker for RAG. G-RAG outperforms state-of-the-art approaches while having smaller computational footprint."
GNN-RAG: Graph Neural Retrieval for Large Language Model Reasoning
Paper • 2405.20139 • Published • 3Note "Experimental results show that GNN-RAG achieves state-of-the-art performance in two widely used KGQA benchmarks (WebQSP and CWQ), outperforming or matching GPT-4 performance with a 7B tuned LLM. In addition, GNN-RAG excels on multi-hop and multi-entity questions outperforming competing approaches by 8.9--15.5% points at answer F1."
GRAG: Graph Retrieval-Augmented Generation
Paper • 2405.16506 • Published • 1Note "Extensive experiments on graph multi-hop reasoning benchmarks demonstrate that in scenarios requiring multi-hop reasoning on textual graphs, our GRAG approach significantly outperforms current state-of-the-art RAG methods while effectively mitigating hallucinations."
Augmenting Textual Generation via Topology Aware Retrieval
Paper • 2405.17602 • Published • 1Note "Unlike proximity-based topological similarity which considers nodes residing closely in one network to be similar, role-based topological similarity focuses on identifying nodes with topologically similar neighborhoods"
PlanRAG: A Plan-then-Retrieval Augmented Generation for Generative Large Language Models as Decision Makers
Paper • 2406.12430 • Published • 7G-Retriever: Retrieval-Augmented Generation for Textual Graph Understanding and Question Answering
Paper • 2402.07630 • Published • 1The Geometry of Categorical and Hierarchical Concepts in Large Language Models
Paper • 2406.01506 • Published • 3Leveraging Large Language Models for Semantic Query Processing in a Scholarly Knowledge Graph
Paper • 2405.15374 • Published • 1Empowering Large Language Models to Set up a Knowledge Retrieval Indexer via Self-Learning
Paper • 2405.16933 • Published • 5KG-RAG: Bridging the Gap Between Knowledge and Creativity
Paper • 2405.12035 • Published • 2Enhancing Q&A with Domain-Specific Fine-Tuning and Iterative Reasoning: A Comparative Study
Paper • 2404.11792 • Published • 1HybridRAG: Integrating Knowledge Graphs and Vector Retrieval Augmented Generation for Efficient Information Extraction
Paper • 2408.04948 • Published • 1GraphReader: Building Graph-based Agent to Enhance Long-Context Abilities of Large Language Models
Paper • 2406.14550 • Published • 4Contextual Document Embeddings
Paper • 2410.02525 • Published • 16StructRAG: Boosting Knowledge Intensive Reasoning of LLMs via Inference-time Hybrid Information Structurization
Paper • 2410.08815 • Published • 41