Multi-Mission Tool Bench: Assessing the Robustness of LLM based Agents through Related and Dynamic Missions
This addresses the need for more realistic benchmarks in the tool invocation domain, though it is incremental as it builds on existing single-mission evaluation methods.
The paper tackles the problem of evaluating LLM-based agents in complex, multi-mission scenarios by proposing the Multi-Mission Tool Bench, a benchmark with interrelated missions that reveals critical factors affecting agent robustness through experiments on diverse models.
Large language models (LLMs) demonstrate strong potential as agents for tool invocation due to their advanced comprehension and planning capabilities. Users increasingly rely on LLM-based agents to solve complex missions through iterative interactions. However, existing benchmarks predominantly access agents in single-mission scenarios, failing to capture real-world complexity. To bridge this gap, we propose the Multi-Mission Tool Bench. In the benchmark, each test case comprises multiple interrelated missions. This design requires agents to dynamically adapt to evolving demands. Moreover, the proposed benchmark explores all possible mission-switching patterns within a fixed mission number. Specifically, we propose a multi-agent data generation framework to construct the benchmark. We also propose a novel method to evaluate the accuracy and efficiency of agent decisions with dynamic decision trees. Experiments on diverse open-source and closed-source LLMs reveal critical factors influencing agent robustness and provide actionable insights to the tool invocation society.