LHRLAB/Graph-R1
A reinforcement learning framework that enables LLMs to iteratively reason over knowledge hypergraphs using a structured retrieval-then-reason cycle.
Not currently ranked — collecting fresh signals.
star history
Graph-R1 applies end-to-end reinforcement learning to improve LLMs’ graph reasoning capabilities. It constructs a knowledge hypergraph from n-ary relation extraction and uses an explicit reward mechanism to train the model through an iterative cycle of thought, query generation, subgraph retrieval, and reconsideration. This approach bridges the gap between graph-structured knowledge and language modalities, enabling more effective utilization of structured knowledge for knowledge-intensive tasks.
Frequently asked
- What is LHRLAB/Graph-R1?
- A reinforcement learning framework that enables LLMs to iteratively reason over knowledge hypergraphs using a structured retrieval-then-reason cycle.
- Is Graph-R1 open source?
- Yes — LHRLAB/Graph-R1 is open source, released under the MIT license.
- What language is Graph-R1 written in?
- LHRLAB/Graph-R1 is primarily written in Python.
- How popular is Graph-R1?
- LHRLAB/Graph-R1 has 584 stars on GitHub.
- Where can I find Graph-R1?
- LHRLAB/Graph-R1 is on GitHub at https://github.com/LHRLAB/Graph-R1.