H2026.05.15 | 30B模型刷奥赛金牌;自蒸馏让3B小模型零外挂超能

2026.05.15 | 30B模型刷奥赛金牌;自蒸馏让3B小模型零外挂超能

HuggingFace 每日AI论文速递·2026 年 5 月 17 日·13 分钟
在小宇宙播放
创作者写的节目说明

【目录】
本期的 15 篇论文如下:
[] 🥇 Achieving Gold-Medal-Level Olympiad Reasoning via Simple and Unified Scaling(通过简单且统一的缩放实现金牌级别的奥赛推理)
[] 🤖 Self-Distilled Agentic Reinforcement Learning(自蒸馏智能体强化学习)
[] 🧠 MemLens: Benchmarking Multimodal Long-Term Memory in Large Vision-Language Models(MemLens:大型视觉语言模型中多模态长期记忆的基准测试)
[] 👁 MemEye: A Visual-Centric Evaluation Framework for Multimodal Agent Memory(MemEye:面向多模态智能体记忆的视觉中心评估框架)
[] 🎬 SANA-WM: Efficient Minute-Scale World Modeling with Hybrid Linear Diffusion Transformer(SANA-WM:高效分钟级世界建模的混合线性扩散Transformer)
[] 🎬 Causal Forcing++: Scalable Few-Step Autoregressive Diffusion Distillation for Real-Time Interactive Video Generation(因果强制++:面向实时交互式视频生成的可扩展少步自回归扩散蒸馏)
[] 🧬 Darwin Family: MRI-Trust-Weighted Evolutionary Merging for Training-Free Scaling of Language-Model Reasoning(达尔文家族:基于MRI信任加权进化合并的无训练语言模型推理扩展)
[] 🐾 WildClawBench: A Benchmark for Real-World, Long-Horizon Agent Evaluation(WildClawBench:面向真实世界长周期智能体评估的基准)
[] 🧠 STALE: Can LLM Agents Know When Their Memories Are No Longer Valid?(STALE:LLM代理能否知晓其记忆何时失效?)
[] 🧠 Beyond Individual Intelligence: Surveying Collaboration, Failure Attribution, and Self-Evolution in LLM-based Multi-Agent Systems(超越个体智能:基于LLM的多智能体系统中的协作、故障归因与自我进化综述)
[] 🎥 Warp-as-History: Generalizable Camera-Controlled Video Generation from One Training Video(扭曲即历史:从单个训练视频实现可泛化的相机控制视频生成)
[] 🧠 PREPING: Building Agent Memory without Tasks(PREPING:无需任务构建智能体记忆)
[] 🧭 RouteProfile: Elucidating the Design Space of LLM Profiles for Routing(RouteProfile:阐明用于路由的LLM配置文件设计空间)
[] 🧠 EvolveMem:Self-Evolving Memory Architecture via AutoResearch for LLM Agents(EvolveMem:面向LLM智能体的自演化记忆架构通过自动研究实现)
[] 🧠 ATLAS: Agentic or Latent Visual Reasoning? One Word is Enough for Both(ATLAS:是智能体推理还是潜在视觉推理?一个词就足够了)

【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


在小宇宙查看该单集文稿

说明里标出的时间点,点一下跳过去

HuggingFace 每日AI论文速递 · 更多单集

看全部 739 期 →

HuggingFace 每日AI论文速递

H2026.05.15 | 30B模型刷奥赛金牌;自蒸馏让3B小模型零外挂超能

HuggingFace 每日AI论文速递 · 5 月 17 日 · 13 分钟

2026.05.15 | 30B模型刷奥赛金牌;自蒸馏让3B小模型零外挂超能

在小宇宙播放
创作者写的节目说明

【目录】
本期的 15 篇论文如下:
[] 🥇 Achieving Gold-Medal-Level Olympiad Reasoning via Simple and Unified Scaling(通过简单且统一的缩放实现金牌级别的奥赛推理)
[] 🤖 Self-Distilled Agentic Reinforcement Learning(自蒸馏智能体强化学习)
[] 🧠 MemLens: Benchmarking Multimodal Long-Term Memory in Large Vision-Language Models(MemLens:大型视觉语言模型中多模态长期记忆的基准测试)
[] 👁 MemEye: A Visual-Centric Evaluation Framework for Multimodal Agent Memory(MemEye:面向多模态智能体记忆的视觉中心评估框架)
[] 🎬 SANA-WM: Efficient Minute-Scale World Modeling with Hybrid Linear Diffusion Transformer(SANA-WM:高效分钟级世界建模的混合线性扩散Transformer)
[] 🎬 Causal Forcing++: Scalable Few-Step Autoregressive Diffusion Distillation for Real-Time Interactive Video Generation(因果强制++:面向实时交互式视频生成的可扩展少步自回归扩散蒸馏)
[] 🧬 Darwin Family: MRI-Trust-Weighted Evolutionary Merging for Training-Free Scaling of Language-Model Reasoning(达尔文家族:基于MRI信任加权进化合并的无训练语言模型推理扩展)
[] 🐾 WildClawBench: A Benchmark for Real-World, Long-Horizon Agent Evaluation(WildClawBench:面向真实世界长周期智能体评估的基准)
[] 🧠 STALE: Can LLM Agents Know When Their Memories Are No Longer Valid?(STALE:LLM代理能否知晓其记忆何时失效?)
[] 🧠 Beyond Individual Intelligence: Surveying Collaboration, Failure Attribution, and Self-Evolution in LLM-based Multi-Agent Systems(超越个体智能:基于LLM的多智能体系统中的协作、故障归因与自我进化综述)
[] 🎥 Warp-as-History: Generalizable Camera-Controlled Video Generation from One Training Video(扭曲即历史:从单个训练视频实现可泛化的相机控制视频生成)
[] 🧠 PREPING: Building Agent Memory without Tasks(PREPING:无需任务构建智能体记忆)
[] 🧭 RouteProfile: Elucidating the Design Space of LLM Profiles for Routing(RouteProfile:阐明用于路由的LLM配置文件设计空间)
[] 🧠 EvolveMem:Self-Evolving Memory Architecture via AutoResearch for LLM Agents(EvolveMem:面向LLM智能体的自演化记忆架构通过自动研究实现)
[] 🧠 ATLAS: Agentic or Latent Visual Reasoning? One Word is Enough for Both(ATLAS:是智能体推理还是潜在视觉推理?一个词就足够了)

【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


在小宇宙查看该单集文稿

说明里标出的时间点,点一下跳过去

HHuggingFace 每日AI论文速递

HuggingFace 每日AI论文速递

面向中文AI从业者与爱好者的每日AI论文速递,工作日高密度速览前沿论文,月末以月度热门论文特辑补充,形成每日速递加月末特…

已收录 739 期 · 最近更新 11 月 10 日

在哪里听

https://feed.xyzfm.space/7jtt3j3lbbra复制

内容版权归创作者,声湃按创作者开放的 RSS 收录。了解收录与 RSS →

更多单集 全部 739 期 →