AI 中文总结
针对排序模型优化依赖昂贵试错的问题，提出利用模型内部状态的MISO工作流，在广告排序案例中提升归一化熵且减少验证运行，实现手动调优与自动搜索的折中。
AI 中文摘要
排序模型在既定模型家族中不断被优化，但选择扩展、替换或舍弃哪个组件往往依赖于昂贵的试错过程。我们提出模型内部状态优化（Model Internal State Optimization，MISO），这是一种利用模型内部状态（MIS，包括参数、激活值、梯度和归一化统计量）来优先确定此类局部优化决策的系统工作流。MISO从已训练的排序模型中提取MIS，将其聚合为排序、对齐和比较信号，并将这些信号转换为少量可解释的候选编辑。由于MISO会在每次重新训练周期后重新提取，MISO自然支持自适应优化工作流，可随着数据分布和系统需求随时间变化跟踪不断演变的模型行为。在广告排序案例研究中，MISO提高了归一化熵，同时比专家驱动和黑盒扩展工作流需要少得多的验证运行，在手动调优和不透明自动搜索之间提供了实用的折中方案。
英文摘要
Ranking models are repeatedly refined within established model families, yet the choice of which component to scale, replace, or retire is often guided by expensive trial-and-error. We present Model Internal State Optimization (MISO), a systems workflow that uses model internal states (MIS), including parameters, activations, gradients, and normalization statistics, to prioritize such local optimization decisions. MISO extracts MIS from a trained ranking model, aggregates them into ranking, alignment, and comparison signals, and converts those signals into a small set of interpretable candidate edits. Because MIS are re-extracted after each retraining cycle, MISO naturally supports an adaptive optimization workflow that tracks evolving model behavior as data distributions and system requirements shift over time. In an ads ranking case study, MISO improves normalized entropy while requiring substantially fewer validation runs than expert-driven and black-box scaling workflows, offering a practical middle ground between manual tuning and opaque automated search.
CommentsAccepted at the OARS Workshop at ACM RecSys 2026
