DiLLS: Interactive Diagnosis of LLM-based Multi-agent Systems via Layered Summary of Agent Behaviors

TL;DR

DiLLS is an interactive framework that structures and summarizes multi-agent system behaviors across multiple levels, significantly aiding developers in diagnosing failures efficiently using natural language queries.

Contribution

This paper introduces DiLLS, a novel layered summarization system that organizes complex agent behaviors for easier diagnosis and understanding of failures in LLM-based multi-agent systems.

Findings

DiLLS improves diagnosis efficiency by 30% in user studies.

Developers can identify root causes faster with layered summaries.

The system effectively organizes behaviors into activities, actions, and operations.

Abstract

Large language model (LLM)-based multi-agent systems have demonstrated impressive capabilities in handling complex tasks. However, the complexity of agentic behaviors makes these systems difficult to understand. When failures occur, developers often struggle to identify root causes and to determine actionable paths for improvement. Traditional methods that rely on inspecting raw log records are inefficient, given both the large volume and complexity of data. To address this challenge, we propose a framework and an interactive system, DiLLS, designed to reveal and structure the behaviors of multi-agent systems. The key idea is to organize information across three levels of query completion: activities, actions, and operations. By probing the multi-agent system through natural language, DiLLS derives and organizes information about planning and execution into a structured, multi-layered summary. Through a user study, we show that DiLLS significantly improves developers' effectiveness and efficiency in identifying, diagnosing, and understanding failures in LLM-based multi-agent systems.