A Simple and Fast Way to Handle Semantic Errors in Transactions
This work addresses the challenge of integrating LLM-powered agents into computer systems for system researchers and engineers, though it is incremental as it builds on existing transaction management principles.
The paper tackles the problem of handling semantic errors in database transactions generated by large language models (LLMs) by proposing a novel middleware framework based on Invariant Satisfaction (I-Confluence), which ensures database consistency while allowing for human review and removal of incorrect transactions, as evaluated using the TPC-C benchmark.
Many computer systems are now being redesigned to incorporate LLM-powered agents, enabling natural language input and more flexible operations. This paper focuses on handling database transactions created by large language models (LLMs). Transactions generated by LLMs may include semantic errors, requiring systems to treat them as long-lived. This allows for human review and, if the transaction is incorrect, removal from the database history. Any removal action must ensure the database's consistency (the "C" in ACID principles) is maintained throughout the process. We propose a novel middleware framework based on Invariant Satisfaction (I-Confluence), which ensures consistency by identifying and coordinating dependencies between long-lived transactions and new transactions. This middleware buffers suspicious or compensating transactions to manage coordination states. Using the TPC-C benchmark, we evaluate how transaction generation frequency, user reviews, and invariant completeness impact system performance. For system researchers, this study establishes an interactive paradigm between LLMs and database systems, providing an "undoing" mechanism for handling incorrect operations while guaranteeing database consistency. For system engineers, this paper offers a middleware design that integrates removable LLM-generated transactions into existing systems with minimal modifications.