【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:
[] 🧭 Harness Handbook: Making Evolving Agent Harnesses Readable,Navigable, and Editable(驾驭手册:使不断演化的智能体驾驭系统可读、可导航且可编辑)
[] 🖼 Boogu-Image-0.1: Boosting Open-Source Unified Multimodal Understanding and Generation(Boogu-Image-0.1:推动开源统一多模态理解与生成)
[] 🧠 Ring-Zero: Scaling Zero RL to a Trillion Parameters for Emergent Reasoning(环零:将零强化学习扩展至万亿参数以实现涌现推理)
[] 📄 OvisOCR2 Technical Report(OvisOCR2 技术报告)
[] 🤖 KnowAct-GUIClaw: Know Deeply, Act Perfectly, Personal GUI Assistant with Self-Evolving Memory and Skill(知深行准-GUIClaw:具备自进化记忆与技能的深度认知、精准执行个人GUI助手)
[] 🛡 PolicyShiftGuard: Benchmarking and Improving Policy-Adaptive Image Guardrails(政策转变守卫:基准测试与改进策略自适应图像防护机制)
[] 🤖 GigaWorld-Policy-0.5: A Faster and Stronger WAM Empowered by AutoResearch(GigaWorld-Policy-0.5:一种由自动研究驱动的更快更强的世界动作模型)
[] 🖼 MetaView: Monocular Novel View Synthesis with Scale-Aware Implicit Geometry Priors(MetaView:具有尺度感知隐式几何先验的单目新视角合成)
[] ✂ ShortOPD: Recovering Pruned LLMs with Short-to-Long On-Policy Distillation(ShortOPD:通过短到长在线策略蒸馏恢复剪枝后的大语言模型)
[] 🎭 Hallo4D: Multi-Modal Hallucination Mitigation for Consistent Spatio-Temporal Generation(Hallo4D:多模态幻觉缓解实现一致的时空生成)
[] 🔬 Registers Matter for Pixel-Space Diffusion Transformers(寄存器对像素空间扩散Transformer至关重要)
[] 🤖 Vinci2: Providing Proactive Assistance in Continuous Egocentric Videos(Vinci2:在连续自我中心视频中提供主动辅助)
[] 🔍 Tracing Agentic Failure from the Flow of Success(从成功流程中追溯智能体失败)
[] 🔍 From Noisy Traces to Root Causes: Structural Trajectory Analysis and Causal Extraction for Agent Optimization(从噪声轨迹到根因:面向智能体优化的结构化轨迹分析与因果提取)
[] 🧭 AgentCompass: A Unified Evaluation Infrastructure for Agent Capabilities(AgentCompass:一种统一的智能体能力评估基础设施)

【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
