DIG to Heal: Scaling General-purpose Agent Collaboration via Explainable Dynamic Decision Paths

2026-02-27

AI Summary

This work addresses the challenges of redundancy, cascading errors, and limited interpretability and controllability that arise when general-purpose large language model agents collaborate without predefined roles or workflows. To this end, the paper introduces the Dynamic Interaction Graph (DIG), a novel framework that models multi-agent collaboration as a temporally evolving causal network. DIG enables, for the first time, real-time observation, interpretation, and intervention in unstructured collaborative processes. By supporting error diagnosis and repair along specific interaction pathways, the method significantly enhances the transparency and controllability of collaborative systems, thereby filling a critical gap in understanding emergent coordination mechanisms among general-purpose agents.

Technology Category

• Multiagent Systems: Coordination and Collaboration
• Cognitive Modeling & Cognitive Systems: Agent Architectures
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG

Abstract

The increasingly popular agentic AI paradigm promises to harness the power of multiple, general-purpose large language model (LLM) agents to collaboratively complete complex tasks. While many agentic AI systems utilize predefined workflows or agent roles in order to reduce complexity, ideally these agents would be truly autonomous, able to achieve emergent collaboration even as the number of collaborating agents increases. Yet in practice, such unstructured interactions can lead to redundant work and cascading failures that are difficult to interpret or correct. In this work, we study multi-agent systems composed of general-purpose LLM agents that operate without predefined roles, control flow, or communication constraints, relying instead on emergent collaboration to solve problems. We introduce the Dynamic Interaction Graph (DIG), which captures emergent collaboration as a time-evolving causal network of agent activations and interactions. DIG makes emergent collaboration observable and explainable for the first time, enabling real-time identification, explanation, and correction of collaboration-induced error patterns directly from agents' collaboration paths. Thus, DIG fills a critical gap in understanding how general LLM agents solve problems together in truly agentic multi-agent systems. The project webpage can be found at: https://happyeureka.github.io/dig.

Problem

• multi-agent systems
• emergent collaboration
• large language models
• explainability
• agent coordination

Innovation

• Dynamic Interaction Graph
• emergent collaboration
• multi-agent systems
• explainable AI
• large language models