作者位次 #2/4ACL 2026Findings· Findings of ACL
GCIG : G raph RAG -based Cross-document Instruction Generation for Boosting LLM Reasoning
Xiaoliang Xu · Huang Yuan · Junmei Wang · Can Xu
摘要
Automatic instruction generation offers a low-cost, high-efficiency pathway for fine-tuning large language models (LLMs). However, existing methods struggle in knowledge-intensive domains and complex reasoning tasks due to their dependence on high-quality seed data, limited coverage of single-doc...