Rethinking Inference-Time Scaling in Local Computer-Use Agents: Failure Modes and Compute Tradeoffs· 重新思考本地计算代理的推理时间扩展
Deploying autonomous computer-use agents (CUAs) locally is increasingly important for privacy, cost efficiency, and practical usability, yet improving their performance under strict hardware constraints remains challenging. While recent studies show that inference-time scaling can improve frontier computer-use agents through additional computation during execution, its effectiveness for resource-constrained local models remains poorly understood. We present a systematic empirical study of inference-time scaling in local CUAs across contextual, temporal, structural, and parallel dimensions. We evaluate Qwen3-VL-8B/30B-A3B, UI-TARS-1.5-7B, and OpenCUA-7B on the OSWorld benchmark. Our results show that additional computation often yields diminishing returns while changing failure modes. Conte
探讨本地计算代理在推理时间扩展中的计算权衡与失效模式。
- 核心方法
- 从上下文、时间、结构和并行四个维度对本地计算代理进行系统性的实验研究，评价不同模型在OSWorld基准测试中的表现。
- 适合谁读
- 研究者、工程师
- 要解决的问题
- 在严格硬件限制下，如何有效提升本地计算代理的性能。
- 关键实验
- 使用Qwen3-VL-8B/30B-A3B、UI-TARS-1.5-7B和OpenCUA-7B模型在OSWorld基准测试上进行评估。
- 主要贡献
- 揭示了额外计算资源投入常常导致收益递减及失效模式的变化，为资源受限环境下的模型优化提供了指导。
- 意义与局限
- 研究为理解本地计算代理在资源受限条件下的表现和优化提供了重要见解，有助于提高这类模型的实际应用价值，但也指出了一些局限性。