【目录】
本期的 15 篇论文如下:
[] 🎥 Kwai Keye-VL-2.0 Technical Report(快手可灵-VL-2.0技术报告)
[] 🌍 ABot-Earth 0.5: Generative 3D Earth Model(ABot-Earth 0.5:生成式三维地球模型)
[] 🤖 Role-Agent: Bootstrapping LLM Agents via Dual-Role Evolution(角色代理:通过双角色进化引导LLM代理)
[] 🔧 Retrospective Harness Optimization: Improving LLM Agents via Self-Preference over Trajectory Rollouts(回顾性装备优化:通过轨迹展开上的自我偏好改进LLM智能体)
[] 🐝 SearchSwarm: Towards Delegation Intelligence in Agentic LLMs for Long-Horizon Deep Research(搜索蜂群:面向长周期深度研究的代理型大语言模型委派智能)
[] 🎥 MemDreamer: Decoupling Perception and Reasoning for Long Video Understanding via Hierarchical Graph Memory and Agentic Retrieval Mechanism(MemDreamer:通过分层图记忆与智能检索机制解耦感知与推理实现长视频理解)
[] 📊 Data Journalist Agent: Transforming Data into Verifiable Multimodal Stories(数据记者智能体:将数据转化为可验证的多模态故事)
[] 🎭 SCAIL-2: Unifying Controlled Character Animation with End-to-end In-Context Conditioning(SCAIL-2:通过端到端上下文条件控制统一受控角色动画)
[] 🔀 Flow-DPPO: Divergence Proximal Policy Optimization for Flow Matching Models(Flow-DPPO:面向流匹配模型的散度近端策略优化)
[] 🏅 WorldOlympiad: Can Your World Model Survive a Triathlon?(世界奥林匹克:你的世界模型能经受三项赛考验吗?)
[] 🎯 Rethinking the Divergence Regularization in LLM RL(重新思考大语言模型强化学习中的散度正则化)
[] 💋 Lip Forcing: Few-Step Autoregressive Diffusion for Real-time Lip Synchronization(唇部强制:用于实时唇部同步的少步自回归扩散)
[] 🤖 EEVEE: Towards Test-time Prompt Learning in the Real World for Self-Improving Agents(EEVEE:面向真实世界测试时提示学习的自改进智能体)
[] 🧠 One Token per Multimodal Evidence: Latent Memory for Resource-Constrained QA(每多模态证据一个令牌:面向资源受限问答的潜在记忆)
[] 🤖 Workflow-GYM: Towards Long-Horizon Evaluation of Computer-use Agentic tasks in Real-World Professional Fields(工作流健身房:面向真实世界专业领域的长周期计算机使用代理任务评估)

【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
【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
