论文

结果表明，由 LLM 编排的多 Agent 系统可将传统 AutoML 扩展为可信、自适应且面向生产的 BDaaS 生命周期自动化。The results suggest that LLM-orchestrated multi-agent systems can extend conventional AutoML toward trustworthy, adaptive, and production-oriented BDaaS lifecycle automation.

CoRe（Context Relevance）被提出。该系统在一个大型短视频搜索引擎中每周重新部署，持续运行超过五个月，使用已部署的多模态相关性模型作为源，并以镜像生产融合代数的乘性比值形式来缩小仿真与生产之间的差距。CoRe (Context Relevance) is presented, such a system, redeployed weekly for over five months in a major short-video search engine, using the deployed multimodal relevance model as its source and a multiplicative ratio form mirroring the production fusion algebra to close the simulation-production gap.

本文提出 LongStraw，一个面向目标、感知架构的系统，用于 resident-state 虚拟化、response replay 和分布式梯度执行，它将实时训练图限制在 response 后缀范围内，同时在完整的 GRPO 组内复用代价高昂的 prompt 计算。This work presents LongStraw, an objective-aware, architecture-aware system for resident-state virtualization, response replay, and distributed-gradient execution that bounds the live training graph by the response suffix while reusing the expensive prompt computation across the complete GRPO group.

子空间对齐重连（SAR）表明，从参数几何中提取对推理有效的更新，可作为一种无需训练的机制来提升推理与多领域性能。Subspace-Aligned Rewiring (SAR) shows that extracting reasoning-effective updates from parameter geometry can serve as a training-free mechanism to improve reasoning and multi-domain performance.

本文为两部分组成的专题论文，介绍物理信息神经网络——一类在训练求解监督学习任务时遵循由一般非线性偏微分方程所描述的物理定律的网络；并展示如何利用这些网络推断偏微分方程的解，以及获得对所有输入坐标和自由参数完全可微的物理信息代理模型。This two part treatise introduces physics informed neural networks -- neural networks that are trained to solve supervised learning tasks while respecting any given law of physics described by general nonlinear partial differential equations and demonstrates how these networks can be used to infer solutions topartial differential equations, and obtain physics-informed surrogate models that are fully differentiable with respect to all input coordinates and free parameters.

实验跟踪器能够展示训练的进展，但更改正在运行的训练通常仍需编写特定于训练器的代码。我们提出了 Interactive Training 2，一个通过共享协议引导训练的开源控制平面。训练应用声明其暴露的设置和操作，人类与自动化控制器通过同一接口提交请求，训练循环在安全的控制点对请求进行验证并应用。定制的 Aim 工作区将实时指标与控制面板以及请求和结果的时序记录整合在一起。我们在 a[……]上展示了该系统。Experiment trackers show how training is progressing, but changing a live run still usually requires trainer-specific code. We present Interactive Training 2, an open-source control plane for steering training through a shared protocol. Training applications declare which settings and actions they expose, humans and automated controllers submit requests through the same interface, and the training loop validates and applies them at safe control points. A customized Aim workspace combines live metrics and controls with a chronological record of requests and outcomes. We demonstrate the system a

本文提出问题：与其缩减真实机器人数据占比，不如提高无机器人 UMI 数据的保真度，从而在后训练阶段移除该 anchor；并开源 HiFi-UMI，一套面向轨迹精度、夹爪间相对位姿、同步与视场协同设计的便携式 UMI 数据生产系统。It is asked whether raising the fidelity of robot-free UMI data, rather than shrinking the real-robot fraction, can remove that anchor at post-training, and open-source HiFi-UMI, a portable UMI data-production system co-designed for trajectory accuracy, inter-gripper relative pose, synchronization, and field of view.

设计了一种类型化的领域特定语言（DSL），通过一组可逆算子捕获重复区域、浮点域等常见张量结构，将无损张量压缩建模为程序合成问题。A typed domain-specific language that captures recurring tensor structures, such as repeated regions and floating-point fields, through a set of reversible operators, is designed, which formulates lossless tensor compression as program synthesis.

提出 YOLO-PEFT，一个结构感知的框架，将 adapter 的放置建模为可审计的约束规划问题，以显式、可审查的规划取代手工对目标模块的试错，同时保留已验证的 train-save-merge-export 路径。YOLO-PEFT is proposed, a structure-aware framework that formulates adapter placement as an auditable constraint-planning problem that replaces manual target-module trial and error with explicit, inspectable planning while preserving verified train-save-merge-export paths.

本文是一项关于如何提升蒸馏训练效率的实践研究，围绕两项系统贡献展开，并提出一种融合的 chunked KL loss，使峰值内存随序列长度线性增长。A practitioner's study of how to make distillation training efficient is presented, organised around two systems contributions, and a fused, chunked KL loss is introduced, making peak memory linear in the sequence length.

提出 Decoupling CLI Agent Scaffolding（DCAS），一种后端替换的拦截层，可在不修改 scaffold 的前提下，在任意 CLI scaffold 与任意后端模型之间路由 API 流量，从而支持跨 scaffold 评估与具备规划感知的轨迹采集。Decoupling CLI Agent Scaffolding (DCAS) is introduced, a backend-substitution interception layer that routes API traffic between any CLI scaffold and any backend model without modifying the scaffold, enabling cross-scaffold evaluation and planning-aware trajectory collection.

本文提出 UniMoMo，一种后训练压缩框架，将其形式化为约束图粗化问题，并引入分层自适应保护机制，根据路由暴露度限制对高流量 expert 的合并。UniMoMo, a post-training compression framework formulated as a constrained graph coarsening problem, is introduced, and a layer-adaptive protection mechanism that restricts the merging of high-traffic experts based on their routing exposure is introduced.