AgentCollab: A Self-Evaluation-Driven Collaboration Paradigm for Efficient LLM Agents

TL;DR

AgentCollab is a dynamic framework that enables LLM agents to self-evaluate and escalate reasoning capacity only when necessary, balancing efficiency and robustness during complex tasks.

Contribution

It introduces a self-driven collaboration paradigm that uses self-reflection signals for dynamic model escalation, improving agent performance without external routing modules.

Findings

AgentCollab improves accuracy-efficiency trade-offs on multi-step benchmarks.

The framework effectively balances low-cost and high-capacity models during reasoning.

Experimental results show consistent performance gains across diverse tasks.

Abstract

Autonomous agents powered by large language models (LLMs) perform complex tasks through long-horizon reasoning and tool interaction, where a fundamental trade-off arises between execution efficiency and reasoning robustness. Models at different capability-cost levels offer complementary advantages: lower-cost models enable fast execution but may struggle on difficult reasoning segments, while stronger models provide more robust reasoning at higher computational cost. We present AgentCollab, a self-driven collaborative inference framework that dynamically coordinates models with different reasoning capacities during agent execution. Instead of relying on external routing modules, the framework uses the agent's own self-reflection signal to determine whether the current reasoning trajectory is making meaningful progress, and escalates control to a stronger reasoning tier only when necessary. To further stabilize long-horizon execution, we introduce a difficulty-aware cumulative escalation strategy that allocates additional reasoning budget based on recent failure signals. In our experiments, we instantiate this framework using a two-level small-large model setting. Experiments on diverse multi-step agent benchmarks show that AgentCollab consistently improves the accuracy-efficiency Pareto frontier of LLM agents.