H2025.02.04 | DAAs性能提升,OmniHuman动画优化。

2025.02.04 | DAAs性能提升,OmniHuman动画优化。

HuggingFace 每日AI论文速递·2025 年 2 月 5 日·16 分钟
在小宇宙播放
创作者写的节目说明

本期的 20 篇论文如下:

[] 🤔 The Differences Between Direct Alignment Algorithms are a Blur(直接对齐算法的差异逐渐模糊)

[] 🤖 OmniHuman-1: Rethinking the Scaling-Up of One-Stage Conditioned Human Animation Models(OmniHuman-1:重新思考单阶段条件式人体动画模型的放大)

[] 💡 Process Reinforcement through Implicit Rewards(基于隐式奖励的过程强化)

[] ⚖ Preference Leakage: A Contamination Problem in LLM-as-a-judge(偏好泄露:LLM即评判器中的污染问题)

[] 🛡 SafeRAG: Benchmarking Security in Retrieval-Augmented Generation of Large Language Model(SafeRAG:评估大语言模型检索增强生成中的安全性)

[] 🚀 FastKV: KV Cache Compression for Fast Long-Context Processing with Token-Selective Propagation(FastKV:通过令牌选择性传播实现快速长文本处理的KV缓存压缩)

[] 🌍 AIN: The Arabic INclusive Large Multimodal Model(AIN:阿拉伯语包容性大型多模态模型)

[] 🧠 DeepRAG: Thinking to Retrieval Step by Step for Large Language Models(DeepRAG:面向大型语言模型的逐步思考检索)

[] 🤔 MM-IQ: Benchmarking Human-Like Abstraction and Reasoning in Multimodal Models(MM-IQ:多模态模型中类人抽象与推理能力的基准测试)

[] 🛡 Almost Surely Safe Alignment of Large Language Models at Inference-Time(大语言模型在推理时近乎完全安全的对齐)

[] 🤔 ZebraLogic: On the Scaling Limits of LLMs for Logical Reasoning(ZebraLogic:关于大型语言模型在逻辑推理中的扩展极限)

[] 🤔 The Jumping Reasoning Curve? Tracking the Evolution of Reasoning Performance in GPT-[n] and o-[n] Models on Multimodal Puzzles(跳跃的推理曲线?追踪GPT-[n]和o-[n]模型在多模态谜题上的推理性能演变)

[] 🎮 Improving Transformer World Models for Data-Efficient RL(改进Transformer世界模型以实现数据高效的强化学习)

[] 💡 Improved Training Technique for Latent Consistency Models(改进的潜在一致性模型训练技术)

[] 🧠 Scaling Embedding Layers in Language Models(语言模型中扩展嵌入层)

[] 🎨 SliderSpace: Decomposing the Visual Capabilities of Diffusion Models(SliderSpace:解构扩散模型的视觉能力)

[] 🤔 PhD Knowledge Not Required: A Reasoning Challenge for Large Language Models(无需博士知识:大型语言模型的推理挑战)

[] 🧠 Lifelong Sequential Knowledge Editing without Model Degradation(终身序列知识编辑,且不降低模型性能)

[] 🔬 Current Pathology Foundation Models are unrobust to Medical Center Differences(当前病理学基础模型对于医疗中心差异不具有鲁棒性)

[] 🫀 A Study on the Performance of U-Net Modifications in Retroperitoneal Tumor Segmentation(U-Net改进模型在腹膜后肿瘤分割中的性能研究)

【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

H2025.02.04 | DAAs性能提升,OmniHuman动画优化。

HuggingFace 每日AI论文速递 · 2025 年 2 月 5 日 · 16 分钟

2025.02.04 | DAAs性能提升,OmniHuman动画优化。

在小宇宙播放
创作者写的节目说明

本期的 20 篇论文如下:

[] 🤔 The Differences Between Direct Alignment Algorithms are a Blur(直接对齐算法的差异逐渐模糊)

[] 🤖 OmniHuman-1: Rethinking the Scaling-Up of One-Stage Conditioned Human Animation Models(OmniHuman-1:重新思考单阶段条件式人体动画模型的放大)

[] 💡 Process Reinforcement through Implicit Rewards(基于隐式奖励的过程强化)

[] ⚖ Preference Leakage: A Contamination Problem in LLM-as-a-judge(偏好泄露:LLM即评判器中的污染问题)

[] 🛡 SafeRAG: Benchmarking Security in Retrieval-Augmented Generation of Large Language Model(SafeRAG:评估大语言模型检索增强生成中的安全性)

[] 🚀 FastKV: KV Cache Compression for Fast Long-Context Processing with Token-Selective Propagation(FastKV:通过令牌选择性传播实现快速长文本处理的KV缓存压缩)

[] 🌍 AIN: The Arabic INclusive Large Multimodal Model(AIN:阿拉伯语包容性大型多模态模型)

[] 🧠 DeepRAG: Thinking to Retrieval Step by Step for Large Language Models(DeepRAG:面向大型语言模型的逐步思考检索)

[] 🤔 MM-IQ: Benchmarking Human-Like Abstraction and Reasoning in Multimodal Models(MM-IQ:多模态模型中类人抽象与推理能力的基准测试)

[] 🛡 Almost Surely Safe Alignment of Large Language Models at Inference-Time(大语言模型在推理时近乎完全安全的对齐)

[] 🤔 ZebraLogic: On the Scaling Limits of LLMs for Logical Reasoning(ZebraLogic:关于大型语言模型在逻辑推理中的扩展极限)

[] 🤔 The Jumping Reasoning Curve? Tracking the Evolution of Reasoning Performance in GPT-[n] and o-[n] Models on Multimodal Puzzles(跳跃的推理曲线?追踪GPT-[n]和o-[n]模型在多模态谜题上的推理性能演变)

[] 🎮 Improving Transformer World Models for Data-Efficient RL(改进Transformer世界模型以实现数据高效的强化学习)

[] 💡 Improved Training Technique for Latent Consistency Models(改进的潜在一致性模型训练技术)

[] 🧠 Scaling Embedding Layers in Language Models(语言模型中扩展嵌入层)

[] 🎨 SliderSpace: Decomposing the Visual Capabilities of Diffusion Models(SliderSpace:解构扩散模型的视觉能力)

[] 🤔 PhD Knowledge Not Required: A Reasoning Challenge for Large Language Models(无需博士知识:大型语言模型的推理挑战)

[] 🧠 Lifelong Sequential Knowledge Editing without Model Degradation(终身序列知识编辑,且不降低模型性能)

[] 🔬 Current Pathology Foundation Models are unrobust to Medical Center Differences(当前病理学基础模型对于医疗中心差异不具有鲁棒性)

[] 🫀 A Study on the Performance of U-Net Modifications in Retroperitoneal Tumor Segmentation(U-Net改进模型在腹膜后肿瘤分割中的性能研究)

【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 →