Reflecting with Two Voices: A Co-Adaptive Dual-Strategy Framework for LLM-Based Agent Decision Making

TL;DR

DuSAR introduces a demonstration-free, dual-strategy framework for LLM agents that mimics human metacognition, achieving state-of-the-art results with reduced computational costs in complex environments.

Contribution

The paper presents DuSAR, a novel co-adaptive reasoning framework enabling a frozen LLM to dynamically balance high-level planning and local policies through reflection, without demonstrations or fine-tuning.

Findings

Achieves state-of-the-art performance on ALFWorld and Mind2Web environments.

Reduces token consumption significantly while maintaining high success rates.

Demonstrates the effectiveness of dual-strategy coordination and optional external knowledge integration.

Abstract

Large language model (LLM) agents often rely on external demonstrations or retrieval-augmented planning, leading to brittleness, poor generalization, and high computational overhead. Inspired by human problem-solving, we propose DuSAR (Dual-Strategy Agent with Reflecting) -- a demonstration-free framework that enables a single frozen LLM to perform co-adaptive reasoning via two complementary strategies: a high-level holistic plan and a context-grounded local policy. These strategies interact through a lightweight reflection mechanism, where the agent continuously assesses progress via a Strategy Fitness Score and dynamically revises its global plan when stuck or refines it upon meaningful advancement, mimicking human metacognitive behavior. On both simulated household (ALFWorld) and real-world web (Mind2Web) environments, DuSAR achieves state-of-the-art performance using only open-source LLMs, substantially outperforming all prior methods without any demonstrations or fine-tuning. Remarkably, it also reduces per-step token consumption by a large margin while maintaining strong task success. Ablation studies confirm the necessity of dual-strategy coordination. Moreover, optional integration of expert demonstrations further boosts performance, highlighting DuSAR's flexibility and compatibility with external knowledge.