摘要：循环语言模型在推理基准测试中已展现出令人鼓舞的成果，但其在 Agent 工具使用方面的潜力仍基本未得到探索。我们在组合式工具调用场景中研究这一问题：模型必须协调多个 API 调用、维护中间状态，并在工具交互过程中保留依赖关系。我们在 API-Bank、BFCL 和 NESTful 上评估了原生及后改造的循环语言模型，并在匹配的监督微调方案下比较循环模型与非循环模型，同时改变推理时的循环深度。在受控实验中，循环计算通常有利于组合式和依赖感知型工具使用；而对于孤立的 API 调用，其收益较小，且更依赖具体模型。多步工具使用的准确率通常会随着循环深度增加而提高；不过，自适应推理仅在需要时分配额外计算，因此实现了更有利的计算量—性能权衡。我们的结果表明，对于需要可靠地规划、协调和执行组合式工具使用工作流的 Agent 系统而言，循环语言模型是一种很有前景的架构。

Abstract:Looped language models have shown promising results on reasoning benchmarks, yet their potential for agentic tool use remains largely unexplored. We study this question in compositional tool-calling settings, where models must coordinate multiple API calls, maintain intermediate state, and preserve dependencies across tool interactions. We evaluate native and retrofitted looped language models on API-Bank, BFCL, and NESTful, comparing looped and non-looped models trained under matched supervised fine-tuning recipes and varying recurrent depth at inference time. In controlled experiments, recurrent computation generally benefits compositional and dependency-aware tool use, while providing smaller and more model-dependent gains on isolated API invocation. Accuracy on multi-step tool use generally increases with recurrent depth; adaptive inference, however, achieves a more favorable compute-performance trade-off by allocating additional computation only when needed. Our results suggest that looped language models are a promising architecture for agentic systems that require reliable planning, coordination, and execution of compositional tool use workflows.