让边缘小模型与云端大模型动态协作，不改参数也能提升推理能力。
Large-Small Model Collaboration for Enhancing Edge-Deployed Small Models
- 通过树搜索动态决定何时由小模型独立推理或与大模型融合
- 在GSM8K和MATH-500上分别提升8.6和10.7个百分点
- 适合资源受限但需高精度推理的边缘部署场景
边缘设备部署领域专用的小语言模型（SLMs），资源有限；私有云则提供更大的大语言模型（LLMs）。我们提出G-Boost，一种自适应的边云协同框架，可在不更新参数的情况下提升已部署的SLM任务性能。该框架将推理建模为树搜索过程，在每一步选择仅使用SLM推理或融合SLM与LLM的输出——这种融合方式将经过适配的领域知识从SLM传递至云端LLM，且不暴露私有数据。一个过程奖励模型引导蒙特卡洛树搜索，动态选择有益的协作步骤。边缘端运行SLM和搜索控制器，云端托管LLM和奖励模型，仅交换当前上下文信息。在GSM8K和MATH-500上基于Qwen2.5和LLaMA2的评估显示，G-Boost优于仅用SLM、静态融合以及微调基线，在两项基准上分别比MCTS和Proxy-Tuning提升8.6和10.7个百分点。结果证实，基于奖励引导的步骤级动态协作能有效增强推理能力和领域知识利用效率。
原文摘要 · Abstract (English)
Edge devices host domain-specific small language models (SLMs) with limited resources, while private clouds offer larger LLMs. We propose G-Boost, an adaptive edge-cloud framework that improves a deployed SLM's task performance without parameter updates. It formulates reasoning as a tree search, choosing at each step between SLM-only inference and SLM-LLM logit fusion---which transfers domain knowledge from the SLM's adapted version to the cloud LLM without exposing private data. A process reward model guides Monte Carlo tree search to select beneficial collaboration steps dynamically. The edge runs the SLM and search controller; the cloud hosts the LLM and reward model, exchanging only current context. Evaluated on GSM8K and MATH-500 with Qwen2.5 and LLaMA2, G-Boost outperforms the SLM alone, static fusion, and fine-tuned baselines, gaining up to 8.6 and 10.7 percentage points over MCTS and Proxy-Tuning, respectively. Results confirm that step-level, reward-guided dynamic collaboration enhances reasoning and domain utilization for deployed edge SLMs.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。