DiscoverHuggingFace 每日AI论文速递【月末特辑】11月最火AI论文 | Kandinsky 5.0全家桶开源;视频生成让模型边播边想
【月末特辑】11月最火AI论文 | Kandinsky 5.0全家桶开源;视频生成让模型边播边想

【月末特辑】11月最火AI论文 | Kandinsky 5.0全家桶开源;视频生成让模型边播边想

Update: 2025-12-03
Share

Description

本期的 10 篇论文如下:

[00:35 ] TOP1(🔥219) | 🎨 Kandinsky 5.0: A Family of Foundation Models for Image and Video Generation(Kandinsky 5.0:用于图像和视频生成的基础模型家族)

[02:45 ] TOP2(🔥207) | 🎬 Thinking with Video: Video Generation as a Promising Multimodal Reasoning Paradigm(用视频思考:视频生成作为统一多模态推理新范式)

[04:58 ] TOP3(🔥191) | 🌍 Lumine: An Open Recipe for Building Generalist Agents in 3D Open Worlds(Lumine:在3D开放世界中打造通才智能体的开源方案)

[07:26 ] TOP4(🔥166) | ⚡ ROOT: Robust Orthogonalized Optimizer for Neural Network Training(ROOT:面向神经网络训练的鲁棒正交化优化器)

[09:37 ] TOP5(🔥156) | 🚀 MiroThinker: Pushing the Performance Boundaries of Open-Source Research Agents via Model, Context, and Interactive Scaling(MiroThinker:通过模型、上下文与交互扩展,将开源研究智能体性能推向新边界)

[11:54 ] TOP6(🔥151) | 🧠 General Agentic Memory Via Deep Research(通过深度研究的通用代理记忆)

[13:55 ] TOP7(🔥131) | 🏅 P1: Mastering Physics Olympiads with Reinforcement Learning(用强化学习攻克物理奥赛)

[16:01 ] TOP8(🔥131) | 🍲 Souper-Model: How Simple Arithmetic Unlocks State-of-the-Art LLM Performance(“汤”级模型:简单加权平均即可让大语言模型性能跃升)

[18:03 ] TOP9(🔥126) | 🧠 Tiny Model, Big Logic: Diversity-Driven Optimization Elicits Large-Model Reasoning Ability in VibeThinker-1.5B(小模型大逻辑:多样性驱动优化唤醒VibeThinker-1.5B的大模型推理力)

[20:14 ] TOP10(🔥121) | 🚀 Diffusion Language Models are Super Data Learners(扩散语言模型是超级数据学习者)

<figure></figure>

【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

Comments 
loading
In Channel
loading
00:00
00:00
1.0x

0.5x

0.8x

1.0x

1.25x

1.5x

2.0x

3.0x

Sleep Timer

Off

End of Episode

5 Minutes

10 Minutes

15 Minutes

30 Minutes

45 Minutes

60 Minutes

120 Minutes

【月末特辑】11月最火AI论文 | Kandinsky 5.0全家桶开源;视频生成让模型边播边想

【月末特辑】11月最火AI论文 | Kandinsky 5.0全家桶开源;视频生成让模型边播边想