text2cypher

五星大厨:全面Multi-Agent 的客服机器人，基于langraph实现，txt2sql ,txt2cypher, lightrag, 多模态 等

collection of text2cypher datasets, evaluations, and finetuning instructions

Knowledge graph construction and RAG demo using Diffbot and Neo4j

Experiments and benchmarks with Text2Cypher for Graph RAG

Maritime knowledge-graph RAG (Neo4j + FastAPI + React): ground-truth multi-hop retrieval benchmark (vector 0.70 → graph 0.85) + cross-document causal insights from 139 real KMST casualty adjudications.

A few approaches using sequence to sequence (seq2seq) architecture to solve semantice parsing problem

Global aviation network graph platform (Neo4j): dual-source ETL, query performance tuning, and a production-grade Text2Cypher NL query layer (89% evaluated execution match)

基于 GraphRAG 的知识图谱技能（Codex Skill），支持实体抽取与消歧、多跳图检索、Text2Cypher、PII 脱敏、文档增量同步，可作为 AI Agent 的工具被自主调用。

GraphRAG API + chat interface for requirements management — 4-level graph enrichment, Text2Cypher, SSE streaming, LangSmith observability

设计基于知识图谱与检索增强生成（GraphRAG）的倒置LLM查询架构

Agentic Text-to-Cypher over Neo4j: QLoRA fine-tuned model, LangGraph validate/self-correct loop, Redis cache + memory, LangSmith evals.

Cheon Cypher (체온 사이퍼) cookbook: Cypher queries from questions over your graph schema

Experimental research project on cloud compliance analysis using Knowledge Graphs, Neo4j and LLM-based Text-to-Cypher.

PIPE-Cypher is a synthetic data pipeline that creates balanced, executable, privacy-aware NL-to-Cypher benchmarks for enterprise knowledge graphs

Building a GraphRAG pipeline with text2cypher retrieval from scratch in Python to understand how an LLM can answer questions using a knowledge graph as its knowledge base.

Symbolic-Assisted Dense Retriever: explainable retrieval over Neo4j knowledge graphs by synthesizing symbolic queries from dense retrieval results

Requirements traceability knowledge graph: Python graph engine with inference rules, Neo4j, and a Streamlit chat UI that answers questions via a local LLM (read-only Text2Cypher)