ALAS: A Stateful Multi-LLM Agent Framework for Disruption-Aware Planning

TL;DR

ALAS introduces a modular framework for multi-LLM agents that improves real-time planning and disruption recovery, achieving state-of-the-art results in job-shop scheduling benchmarks.

Contribution

The paper presents ALAS, a novel framework that decomposes planning into role-specific agents with persistent state, enhancing disruption handling and scalability in LLM-based planning.

Findings

Sets new best results in static sequential planning benchmarks.

Excels in dynamic scenarios with unexpected disruptions.

Demonstrates scalable and resilient planning with modular multi-agent LLM system.

Abstract

Large language models (LLMs) excel at rapid generation of text and multimodal content, yet they falter on transaction-style planning that demands ACID-like guarantees and real-time disruption recovery. We present Adaptive LLM Agent System (ALAS), a framework that tackles four fundamental LLM deficits: (i) absence of self-verification, (ii) context erosion, (iii) next-token myopia, and (iv) lack of persistent state. ALAS decomposes each plan into role-specialized agents, equips them with automatic state tracking, and coordinates them through a lightweight protocol. When disruptions arise, agents apply history-aware local compensation, avoiding costly global replanning and containing cascade effects. On real-world, large-scale job-shop scheduling benchmarks, ALAS sets new best results for static sequential planning and excels in dynamic reactive scenarios with unexpected disruptions. These gains show that principled modularization plus targeted compensation can unlock scalable and resilient planning with LLMs.