How to Build a Model-Native Agent That Learns Internal Planning, Memory, and Multi-Tool Reasoning Through End-to-End Reinforcement Learning
"强化学习" 相关文章
解密prompt系列63. Agent训练方案:RStar2 & Early Experience etc
对话郎咸朋：VLA 技术论战、团队换血与不被看好时的自我证明
深度｜Andrej Karpathy：行业对Agent的发展过于乐观，一个能真正帮你工作的Agent还需要十年发展时间
智源具身框架Thor开源：迈向类人级全身控制，让机器人在强对抗中“站稳脚跟”
用更一致的轨迹、更少的解码步数「驯服」掩码扩散语言模型，扩散语言模型的推理性能和效率大幅提升
比NanoBanana更擅长中文和细节控制，兔展&北大Uniworld V2刷新SOTA
用更一致的轨迹、更少的解码步数「驯服」掩码扩散语言模型，扩散语言模型的推理性能和效率大幅提升
用更一致的轨迹、更少的解码步数「驯服」掩码扩散语言模型，扩散语言模型的推理性能和效率大幅提升
数字生命「培养皿」里，AI竟然学会了打架、结盟、抢地盘
Neighboring State-based Exploration for Reinforcement Learning
Interpretable end-to-end Neurosymbolic Reinforcement Learning agents
STACKFEED: Structured Textual Actor-Critic Knowledge Base Editing with FeedBack
GenDexHand: Generative Simulation for Dexterous Hands
RLAC: Reinforcement Learning with Adversarial Critic for Free-Form Generation Tasks
Efficient Reinforcement Learning for Large Language Models with Intrinsic Exploration
Bootstrap Off-policy with World Model
UME-R1: Exploring Reasoning-Driven Generative Multimodal Embeddings
Who Can We Trust? Scope-Aware Video Moment Retrieval with Multi-Agent Conflict
Consistently Simulating Human Personas with Multi-Turn Reinforcement Learning
