K8s 运维和平台工程团队终于有了一个轻量级方案来生成 YAML 清单——1.5B 模型就能跑出 91.5% 的通过率，做基础设施自动化的开发者可以直接参考其数据蒸馏思路。
该论文提出一种名为“上下文工具数据蒸馏”的方法，专门用于让小语言模型（SLM，参数最多 4B）生成 Kubernetes YAML 等 DSL 工件。方法通过合成生成和反向指令生成构建语料，并仅将通过外部验证器且匹配领域上下文的样本加入训练。在资源受限条件下，使用 DeepSeek-V4 Flash 作为教师模型，微调 Qwen2.5-Coder-1.5B-Instruct，在 K8s-Distill-Pilot 数据集上达到 91.5% 的完全通过率。关键发现是：输出格式的严格约束比增加训练样本数对结果质量影响更大。
Context-Instrumental Data Distillation for Kubernetes Manifest Generation: Method and Experimental Evaluation
This paper examines the specialization of Small Language Models (SLMs) with up to 4 billion parameters for generating artifacts in domain-specific languages (DSL). Kubernetes manifests are chosen as the target domain. We propose the context-instrumental data distillation method: the source corpus is formed through synthetic generation and, in an extended scheme, through reverse instruction generation from real Kubernetes YAML files, with pairs included in training only upon passing external validators and matching the domain context model. Unlike classical KL-divergence knowledge distillation, the baseline implementation reduces to supervised fine-tuning on instrumentally verified examples. The experimental section presents a pilot implementation under resource-constrained conditions: the DeepSeek-V4 Flash API serves as the teacher for synthetic generation, while Qwen2.5-Coder-1.5B-Instruct is fine-tuned via LoRA on CPU. On the K8s-Distill-Pilot corpus (train_1200, validation_100, test_200), we achieved full-pass@1 = 91.5% (183/200) with a stricter prompt formulation and max_new_tokens=768. The key empirical finding is that for Kubernetes YAML, result quality in the pilot depended more on strict output format requirements than on simply increasing the number of training examples.