Large Language Models (LLMs) often generate incorrect or outdated information, especially in low-resource settings or when dealing with private data. To address this, Retrieval-Augmented Generation (RAG) uses external knowledge bases (KBs), but these can also suffer from inaccuracies. We introduce STACKFEED, a novel Structured Textual Actor-Critic Knowledge base editing with FEEDback approach that iteratively refines the KB based on expert feedback using a multi-actor, centralized critic reinforcement learning framework. STACKFEED defines a ReACT actor agent on each document to perform structured edits based on document specific targeted instructions. Experimental results showcase that STACKFEED significantly improves KB quality and performance of the RAG system. We evaluate STACKFEED on low-resource programming problems, modified python packaged and factual question-answering tasks.
@article{arxiv.2410.10584,
title = {STACKFEED: Structured Textual Actor-Critic Knowledge Base Editing with FeedBack},
author = {Shashank Kirtania and Naman Gupta and Priyanshu Gupta and Krishna Kariya and Sumit Gulwani and Arun Iyer and Suresh Parthasarathy and Arjun Radhakrishna and Sriram K. Rajamani and Gustavo Soares},
journal= {arXiv preprint arXiv:2410.10584},
year = {2025}
}