HyperReCo: Retrieving and Connecting Evidence with Hypergraph Neural Networks for LLM Multi-hop Reasoning
This work proposes HyperReCo, a framework for retrieving and connecting evidence with a hypergraph neural network (HyperGNN), and introduces Gradient-Guided Hyper-Path Decoding (GGHD), which uses gradient attribution to interpret the learned interactions and translate them into explicit hyper-paths that help LLMs combi...
Zi-Cheng Zhao, Lin-Hao Luo, Jun-Nan Dong et al.
· 0 citations