Are Tools All We Need? Unveiling the Tool-Use Tax in LLM Agents
For researchers and practitioners using LLM agents, this work highlights a critical tradeoff where tool use can degrade performance under semantic noise, challenging the assumption that tools always improve reasoning.
The paper reveals that tool-augmented reasoning can underperform chain-of-thought (CoT) in the presence of semantic distractors, identifying a 'tool-use tax' from the tool-calling protocol. It introduces G-STEP, a lightweight gate that partially recovers performance, but concludes that intrinsic reasoning improvements are needed.
Tool-augmented reasoning has become a popular direction for LLM-based agents, and it is widely assumed to improve reasoning and reliability. However, we demonstrate that this consensus does not always hold: in the presence of semantic distractors, tool-augmented reasoning does not necessarily outperform native CoT. To explain this performance gap, we propose a Factorized Intervention Framework that isolates the cost of prompt formatting, the overhead of the tool-calling protocol, and the actual gain from executing tools. Our analysis reveals a critical tradeoff: under semantic noise, the gains from tools often fail to offset the "tool-use tax", which is the performance degradation introduced by the tool-calling protocol itself. To address this, we introduce G-STEP, a lightweight inference-time gate to mitigate protocol-induced errors. While this yields partial recovery, our findings suggest that more substantial improvements still require strengthening the model's intrinsic reasoning and tool-interaction capabilities.