【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:
[] ⚡ TurboVLA: Real-Time Vision-Language-Action Model at 32 Hz on an RTX 4090 with <1 GB VRAM(TurboVLA:在RTX 4090上以32Hz频率运行且显存占用低于1GB的实时视觉-语言-动作模型)
[] 🎯 CoRT: Counterfactual Replay for Token-Level Rubric-Guided Policy Optimization(CoRT:用于令牌级准则引导策略优化的反事实重放)
[] 🤖 HumanCLAW: Can Vision-Language Models Act Through a Body?(HumanCLAW:视觉-语言模型能否通过身体行动?)
[] 🧬 DecoEvo: Score-Decoupled Co-Evolution of Solver and Rubric-Generator Skills in Text Space(DecoEvo:文本空间中求解器与评估生成器技能的解耦协同进化)
[] 🧠 CLBench-V: Evaluating Multimodal Context Learning from Grounding to Knowledge Acquisition(CLBench-V:从基础定位到知识获取的多模态上下文学习评估)
[] 🧩 CAST: Game Solvers as Turn-Level Teachers for LLM Agents(CAST:游戏求解器作为LLM智能体的回合级教师)
[] 🧠 SkillRise: Agentic Reinforcement Learning for Cross-Task Skill Evolution(技能崛起:面向跨任务技能演化的智能体强化学习)
[] 🎮 StatePlay: State-Aware Game World Models for Mechanics-Consistent Generation(StatePlay:状态感知的游戏世界模型用于机制一致的内容生成)
[] 📊 OmegaUse-OfficeVal: Benchmarking LLM Agents on Long-Horizon Office-Suite Tasks with Economic Grounding(OmegaUse-OfficeVal:基于经济基准评估LLM智能体在长期办公套件任务中的表现)
[] 🤖 Can AI agents conduct open-ended AI research? Early evidence from two case studies(AI智能体能否进行开放式的AI研究?来自两个案例研究的早期证据)
[] 🛡 GPT-Red: Automated Red Teaming via Self-Play at Scale(GPT-Red:通过大规模自我对弈实现自动化红队测试)
[] 🎬 Explicit Layer Modeling for Video Object Insertion and Layer Decomposition(显式层建模用于视频对象插入与层分解)
[] 🕵 StealthBench: Measuring Operational Stealth in Autonomous Offensive-Security Agents(StealthBench:衡量自主攻击安全代理的操作隐蔽性)
[] 🧠 Memory for Large Language Models(大型语言模型的记忆机制)
[] 📜 Grading the Narrators: An Isnad-Rijal Framework for Claim-Level Provenance in Multi-Agent Knowledge Systems(为叙述者评级:面向多智能体知识系统中声明级溯源的一种伊斯纳德-里贾尔框架)
【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
