PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
May 20, 2026Proceedings of the ACM on Management of Data0 citations

HAMMER: An Automatic RAG Tuning System via Hierarchical Memory-Guided Monte Carlo Tree Search

View Full Paper
YZY K ZhouZWZixuan WangYFYixiang Fang

Key Points

  • The aim is to enhance the tuning of Retrieval-Augmented Generation (RAG) systems by addressing hyperparameter optimization challenges.
  • Proposed HAMMER, a hierarchical memory–guided Monte Carlo Tree Search (MCTS) system.
  • Employed a hierarchical graph memory to organize experimental insights.
  • Developed a query selection technique for cost reduction while maintaining effectiveness.
  • Improved exact match by up to 20.0% and F1-score by 15.2%.
  • Reduced tuning time and token consumption by up to 9×.

Abstract

Retrieval-Augmented Generation (RAG) effectively integrates external knowledge into large language models (LLMs), enhancing accuracy, adaptability, interpretability, and trustworthiness. The modern RAG systems contain multiple modules with many tunable parameters, each of which can greatly influence the overall performance of downstream applications. To identify the optimal configuration for a given RAG pipeline, existing studies have relied on hyperparameter optimization (HPO) methods. However, these methods overlook parameter dependencies, lack experience-based learning, and incur substantial computational costs. To tackle those issues, we propose HAMMER, a hierarchical memory–guided Monte Carlo Tree Search (MCTS) system. Inspired by human learning, HAMMER employs a hierarchical graph memory to organize experimental insights and integrates it with MCTS for more reliable tuning. We further design a theoretically guaranteed query selection technique to reduce cost while preserving effectiveness. Extensive experiments on eight real-world datasets show that HAMMER improves exact match by up to 20.0% and F1-score by 15.2%, while reducing both tuning time and token consumption by up to 9×.

Ask AI
Helpful
Bookmark
Share
View Full Paper

Cite This Study

Zhou et al. (2026) studied this question.

synapsesocial.com/papers/6a0d4f62f03e14405aa9aacchttps://doi.org/10.1145/3802071
Ask AI
Helpful
Bookmark
Share
View Full Paper