
2026.11.09 | 高斯间距校准跨场景重建;参数先验加速多教师蒸馏
11 月 10 日 · 11 分钟

本期的 31 篇论文如下:
[] 📊 MixEval-X: Any-to-Any Evaluations from Real-World Data Mixtures(MixEval-X:从现实世界数据混合中进行任意到任意评估)
[] 🎥 Movie Gen: A Cast of Media Foundation Models(电影生成:媒体基础模型集合)
[] 📱 MobA: A Two-Level Agent System for Efficient Mobile Task Automation(MobA:一种高效移动任务自动化的两级代理系统)
[] 🌐 Harnessing Webpage UIs for Text-Rich Visual Understanding(利用网页UI进行丰富的视觉理解)
[] 🔄 Janus: Decoupling Visual Encoding for Unified Multimodal Understanding and Generation(雅努斯:解耦视觉编码以实现统一的多模态理解和生成)
[] 🩺 MMed-RAG: Versatile Multimodal RAG System for Medical Vision Language Models(多功能多模态RAG系统在医学视觉语言模型中的应用)
[] 📊 A Unified View of Delta Parameter Editing in Post-Trained Large-Scale Models(大规模模型后训练中Delta参数编辑的统一视角)
[] 🔄 PopAlign: Diversifying Contrasting Patterns for a More Comprehensive Alignment(PopAlign:多样化对比模式以实现更全面的模型对齐)
[] 🔍 BenTo: Benchmark Task Reduction with In-Context Transferability(BenTo: 基于上下文迁移性的基准任务缩减)
[] 🎥 DreamVideo-2: Zero-Shot Subject-Driven Video Customization with Precise Motion Control(DreamVideo-2:零样本主题驱动视频定制与精确运动控制)
[] 🧠 MoH: Multi-Head Attention as Mixture-of-Head Attention(MoH:多头部注意力机制作为混合头部注意力机制)
[] 🎥 VidPanos: Generative Panoramic Videos from Casual Panning Videos(VidPanos:从随意拍摄的平移视频生成全景视频)
[] 📉 FlatQuant: Flatness Matters for LLM Quantization(FlatQuant:扁平化对LLM量化的重要性)
[] 🔄 Retrospective Learning from Interactions(从交互中回顾学习)
[] 🔄 Failing Forward: Improving Generative Error Correction for ASR with Synthetic Data and Retrieval Augmentation(向前失败:利用合成数据和检索增强改进ASR的生成错误校正)
[] 🖼 Can MLLMs Understand the Deep Implication Behind Chinese Images?(多模态大语言模型能否理解中文图像的深层含义?)
[] 📱 MedMobile: A mobile-sized language model with expert-level clinical capabilities(MedMobile:具备专家级临床能力的移动端语言模型)
[] 🌍 WorldCuisines: A Massive-Scale Benchmark for Multilingual and Multicultural Visual Question Answering on Global Cuisines(世界美食:多语言多文化视觉问答的大规模基准)
[] 🤖 Remember, Retrieve and Generate: Understanding Infinite Visual Concepts as Your Personalized Assistant(记住、检索与生成:理解无限视觉概念作为个性化助手)
[] 🔄 LoLDU: Low-Rank Adaptation via Lower-Diag-Upper Decomposition for Parameter-Efficient Fine-Tuning(LoLDU:通过下三角-对角-上三角分解实现低秩适应的参数高效微调)
[] 🔒 AERO: Softmax-Only LLMs for Efficient Private Inference(AERO:仅使用Softmax的LLM实现高效隐私推断)
[] 🌐 $γ-$MoD: Exploring Mixture-of-Depth Adaptation for Multimodal Large Language Models(γ-MoD:探索多模态大语言模型的深度混合适应)
[] 🌐 Long-LRM: Long-sequence Large Reconstruction Model for Wide-coverage Gaussian Splats(长序列大重建模型:广覆盖高斯点云)
[] 🎶 MuVi: Video-to-Music Generation with Semantic Alignment and Rhythmic Synchronization(MuVi:视频到音乐生成与语义对齐及节奏同步)
[] 🔒 Do LLMs Have Political Correctness? Analyzing Ethical Biases and Jailbreak Vulnerabilities in AI Systems(大型语言模型是否具备政治正确性?分析AI系统中的伦理偏见与越狱漏洞)
[] 📚 SBI-RAG: Enhancing Math Word Problem Solving for Students through Schema-Based Instruction and Retrieval-Augmented Generation(基于模式教学和检索增强生成的数学应用题解决方法)
[] 🗺 Roadmap towards Superhuman Speech Understanding using Large Language Models(基于大型语言模型的超人类语音理解路线图)
[] 🔄 Toward Guidance-Free AR Visual Generation via Condition Contrastive Alignment(面向无指导的AR视觉生成的条件对比对齐)
[] 🤖 TransAgent: Transfer Vision-Language Foundation Models with Heterogeneous Agent Collaboration(TransAgent:异构代理协作迁移视觉语言基础模型)
[] 🔬 Open Materials 2024 (OMat24) Inorganic Materials Dataset and Models(开放材料2024(OMat24)无机材料数据集与模型)
[] 📚 Minimum Tuning to Unlock Long Output from LLMs with High Quality Data as the Key(最小调整解锁LLM长输出:高质量数据的关键)

【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
说明里标出的时间点,点一下跳过去

2026.11.09 | 高斯间距校准跨场景重建;参数先验加速多教师蒸馏
11 月 10 日 · 11 分钟

【周末特辑】11月第2周最火AI论文 | 图形智能体的设计空间;错峰去噪生成视频
11 月 8 日 · 11 分钟

2026.11.06 | 视觉规划整块解码;借兄弟框架练智能体
11 月 7 日 · 9 分钟

【月末特辑】10月最火AI论文 | 把经验变成预见;让音画同步生成
11 月 7 日 · 13 分钟

HuggingFace 每日AI论文速递 · 2024 年 10 月 19 日 · 21 分钟
本期的 31 篇论文如下:
[] 📊 MixEval-X: Any-to-Any Evaluations from Real-World Data Mixtures(MixEval-X:从现实世界数据混合中进行任意到任意评估)
[] 🎥 Movie Gen: A Cast of Media Foundation Models(电影生成:媒体基础模型集合)
[] 📱 MobA: A Two-Level Agent System for Efficient Mobile Task Automation(MobA:一种高效移动任务自动化的两级代理系统)
[] 🌐 Harnessing Webpage UIs for Text-Rich Visual Understanding(利用网页UI进行丰富的视觉理解)
[] 🔄 Janus: Decoupling Visual Encoding for Unified Multimodal Understanding and Generation(雅努斯:解耦视觉编码以实现统一的多模态理解和生成)
[] 🩺 MMed-RAG: Versatile Multimodal RAG System for Medical Vision Language Models(多功能多模态RAG系统在医学视觉语言模型中的应用)
[] 📊 A Unified View of Delta Parameter Editing in Post-Trained Large-Scale Models(大规模模型后训练中Delta参数编辑的统一视角)
[] 🔄 PopAlign: Diversifying Contrasting Patterns for a More Comprehensive Alignment(PopAlign:多样化对比模式以实现更全面的模型对齐)
[] 🔍 BenTo: Benchmark Task Reduction with In-Context Transferability(BenTo: 基于上下文迁移性的基准任务缩减)
[] 🎥 DreamVideo-2: Zero-Shot Subject-Driven Video Customization with Precise Motion Control(DreamVideo-2:零样本主题驱动视频定制与精确运动控制)
[] 🧠 MoH: Multi-Head Attention as Mixture-of-Head Attention(MoH:多头部注意力机制作为混合头部注意力机制)
[] 🎥 VidPanos: Generative Panoramic Videos from Casual Panning Videos(VidPanos:从随意拍摄的平移视频生成全景视频)
[] 📉 FlatQuant: Flatness Matters for LLM Quantization(FlatQuant:扁平化对LLM量化的重要性)
[] 🔄 Retrospective Learning from Interactions(从交互中回顾学习)
[] 🔄 Failing Forward: Improving Generative Error Correction for ASR with Synthetic Data and Retrieval Augmentation(向前失败:利用合成数据和检索增强改进ASR的生成错误校正)
[] 🖼 Can MLLMs Understand the Deep Implication Behind Chinese Images?(多模态大语言模型能否理解中文图像的深层含义?)
[] 📱 MedMobile: A mobile-sized language model with expert-level clinical capabilities(MedMobile:具备专家级临床能力的移动端语言模型)
[] 🌍 WorldCuisines: A Massive-Scale Benchmark for Multilingual and Multicultural Visual Question Answering on Global Cuisines(世界美食:多语言多文化视觉问答的大规模基准)
[] 🤖 Remember, Retrieve and Generate: Understanding Infinite Visual Concepts as Your Personalized Assistant(记住、检索与生成:理解无限视觉概念作为个性化助手)
[] 🔄 LoLDU: Low-Rank Adaptation via Lower-Diag-Upper Decomposition for Parameter-Efficient Fine-Tuning(LoLDU:通过下三角-对角-上三角分解实现低秩适应的参数高效微调)
[] 🔒 AERO: Softmax-Only LLMs for Efficient Private Inference(AERO:仅使用Softmax的LLM实现高效隐私推断)
[] 🌐 $γ-$MoD: Exploring Mixture-of-Depth Adaptation for Multimodal Large Language Models(γ-MoD:探索多模态大语言模型的深度混合适应)
[] 🌐 Long-LRM: Long-sequence Large Reconstruction Model for Wide-coverage Gaussian Splats(长序列大重建模型:广覆盖高斯点云)
[] 🎶 MuVi: Video-to-Music Generation with Semantic Alignment and Rhythmic Synchronization(MuVi:视频到音乐生成与语义对齐及节奏同步)
[] 🔒 Do LLMs Have Political Correctness? Analyzing Ethical Biases and Jailbreak Vulnerabilities in AI Systems(大型语言模型是否具备政治正确性?分析AI系统中的伦理偏见与越狱漏洞)
[] 📚 SBI-RAG: Enhancing Math Word Problem Solving for Students through Schema-Based Instruction and Retrieval-Augmented Generation(基于模式教学和检索增强生成的数学应用题解决方法)
[] 🗺 Roadmap towards Superhuman Speech Understanding using Large Language Models(基于大型语言模型的超人类语音理解路线图)
[] 🔄 Toward Guidance-Free AR Visual Generation via Condition Contrastive Alignment(面向无指导的AR视觉生成的条件对比对齐)
[] 🤖 TransAgent: Transfer Vision-Language Foundation Models with Heterogeneous Agent Collaboration(TransAgent:异构代理协作迁移视觉语言基础模型)
[] 🔬 Open Materials 2024 (OMat24) Inorganic Materials Dataset and Models(开放材料2024(OMat24)无机材料数据集与模型)
[] 📚 Minimum Tuning to Unlock Long Output from LLMs with High Quality Data as the Key(最小调整解锁LLM长输出:高质量数据的关键)

【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
说明里标出的时间点,点一下跳过去

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
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 →

2026.11.09 | 高斯间距校准跨场景重建;参数先验加速多教师蒸馏
11 月 10 日 · 11 分钟

【周末特辑】11月第2周最火AI论文 | 图形智能体的设计空间;错峰去噪生成视频
11 月 8 日 · 11 分钟

2026.11.06 | 视觉规划整块解码;借兄弟框架练智能体
11 月 7 日 · 9 分钟