A2RAG: Adaptive Agentic Graph Retrieval for Cost-Aware and Reliable Reasoning

RAIDS Lab Authors

Zhengyi Yang

(corresponding author)

Director

Details

Research Area

Tags

Resources

Awards

Abstract

Graph Retrieval-Augmented Generation (Graph-RAG) enhances multihop question answering by organizing corpora into knowledge graphs and routing evidence through relational structure. However, practical deployments face two persistent bottlenecks: mixed-difficulty workloads where one-size-fits-all retrieval either wastes cost on easy queries or fails on hard multihop cases, and extraction loss, where graph abstraction omits fine-grained qualifiers that remain only in source text. We present A2RAG, an adaptive-and-agentic GraphRAG framework for cost-aware and reliable reasoning. A2RAG couples an adaptive controller that verifies evidence sufficiency and triggers targeted refinement only when necessary, with an agentic retriever that progressively escalates retrieval effort and maps graph signals back to provenance text to remain robust under extraction loss and incomplete graphs. Experiments on HotpotQA and 2WikiMultiHopQA demonstrate that A2RAG achieves +9.9/+11.8 absolute gains in Recall@2, while cutting token consumption and end-to-end latency by about 50% relative to iterative multihop baselines.

Author Affiliations

@inproceedings{liu2026a2rag,
  title = {A2RAG: Adaptive Agentic Graph Retrieval for Cost-Aware and Reliable Reasoning},
  author = {Liu, Jiate and Chen, Zebin and Qiao, Shaobo and Ju, Mingchen and Zhang, Danting and Han, Bocheng and Yu, Shuyue and Shu, Xin and Wu, Jinglin and Wen, Dong and Cao, Xin and Liu, Guanfeng and Yang, Zhengyi},
  url = {http://dx.doi.org/10.1109/icdew71238.2026.00024},
  DOI = {10.1109/icdew71238.2026.00024},
  booktitle = {2026 IEEE 42nd International Conference on Data Engineering Workshops (ICDEW)},
  publisher = {IEEE},
  year = {2026},
  month = May,
  pages = {187-196}
}