【赞助商】
OpenClaw快报
每天五分钟,听听 OpenClaw 快报,带你了解最新动态和业内讨论
传送门 www.xiaoyuzhoufm.com
【目录】
本期的 15 篇论文如下:
[] 🎮 Generative World Renderer at the Speed of Play(以游戏速度运行的生成式世界渲染器)
[] 🔀 DataFlow-Harness: A Grounded Code-Agent Platform for Constructing Editable LLM Data Pipelines(数据流驾驭平台:一种用于构建可编辑LLM数据管道的接地代码代理平台)
[] 🔍 Text Template Tokens Are Implicit Semantic Registers in Diffusion Transformers(文本模板标记是扩散Transformer中的隐式语义寄存器)
[] ⚡ Mage-Flow: An Efficient Native-Resolution Foundation Model for Image Generation and Editing(Mage-Flow:一种用于图像生成与编辑的高效原生分辨率基础模型)
[] 🌍 AlayaWorld: Interactive Long-Horizon World Modeling -- Full Technical Report(AlayaWorld:交互式长时域世界建模——完整技术报告)
[] 🤖 Stale but Stable: Staleness-Adaptive Trust Regions for Stabilizing Asynchronous Reinforcement Learning(陈旧但稳定:面向异步强化学习稳定化的陈旧性自适应信任区域方法)
[] 🌍 ABot-World-0: Infinite Interactive World Rollout on a Single Desktop GPU(ABot-World-0:在单个桌面GPU上实现无限交互式世界展开)
[] 📊 SciForma: Structure-Faithful Generation of Scientific Diagrams(SciForma:科学图表的结构忠实生成)
[] 🔍 AgentDebugX: An Open-Source Toolkit for Failure Observability, Attribution, and Recovery in LLM Agents(AgentDebugX:用于LLM Agent故障可观测性、归因与恢复的开源工具包)
[] ⚡ HPD-Parsing: Hierarchical Parallel Document Parsing(HPD-Parsing:层级并行文档解析)
[] 📊 Two-Level Meta-Rubrics for Evaluating Open-Ended Generation: GAMUT, a Benchmark for Factual Completeness(用于评估开放生成的两级元评价标准:GAMUT,一个面向事实完整性的基准测试)
[] 🧠 ISO: An RLVR-Native Optimization Stack(ISO:一种原生于RLVR的优化栈)
[] 🎙 Transcription Policy as a Latent Variable: Activating Controllable Verbatim ASR with Word-Level Timing(转录策略作为潜在变量:通过词级时序激活可控的逐字语音识别)
[] 🎓 EduPanel: A Three-Agent LLM Judge for Teaching Videos -- Reliability, Complementarity, and Human Trust Calibration(EduPanel:一种用于教学视频的三智能体LLM评审器——可靠性、互补性与人类信任校准)
[] 🎬 Masked Visual Actions for Unified World Modeling(掩蔽视觉动作:用于统一世界建模)

【关注我们】
您还可以在以下平台找到我们,获得播客内容以外更多信息
小红书: AI速递
