We're sunsetting PodQuest on 2025-07-28. Thank you for your support!
Export Podcast Subscriptions
cover of episode 2025.06.10 | 强化学习改进语言模型;医学多模态模型提升推理能力。

2025.06.10 | 强化学习改进语言模型;医学多模态模型提升推理能力。

2025/6/10
logo of podcast HuggingFace 每日AI论文速递

HuggingFace 每日AI论文速递

AI Chapters
Chapters

Shownotes Transcript

本期的 15 篇论文如下:

[00:21] 🤖 Reinforcement Pre-Training(强化预训练)

[01:01] 🩺 Lingshu: A Generalist Foundation Model for Unified Multimodal Medical Understanding and Reasoning(灵枢:用于统一多模态医学理解与推理的通用基础模型)

[01:42] 📱 MiniCPM4: Ultra-Efficient LLMs on End Devices(MiniCPM4:终端设备上的超高效大型语言模型)

[02:30] 🛡 Saffron-1: Towards an Inference Scaling Paradigm for LLM Safety Assurance(Saffron-1:面向LLM安全保障的推理扩展范式)

[03:07] 🖼 OneIG-Bench: Omni-dimensional Nuanced Evaluation for Image Generation(OneIG-Bench:用于图像生成的全方位细致评估)

[03:49] 🏠 SpatialLM: Training Large Language Models for Structured Indoor Modeling(SpatialLM:用于结构化室内建模的大型语言模型训练)

[04:35] 🤖 Astra: Toward General-Purpose Mobile Robots via Hierarchical Multimodal Learning(Astra:通过分层多模态学习迈向通用移动机器人)

[05:14] 🖼 Rethinking Cross-Modal Interaction in Multimodal Diffusion Transformers(重新思考多模态扩散Transformer中的跨模态交互)

[06:02] 🖼 Image Reconstruction as a Tool for Feature Analysis(图像重建作为特征分析的工具)

[06:41] 🧪 GTR-CoT: Graph Traversal as Visual Chain of Thought for Molecular Structure Recognition(GTR-CoT:用于分子结构识别的图遍历视觉链式思考)

[07:22] 📉 Through the Valley: Path to Effective Long CoT Training for Small Language Models(穿越低谷:小语言模型有效长链思考训练之路)

[08:04] 🤖 BitVLA: 1-bit Vision-Language-Action Models for Robotics Manipulation(BitVLA:用于机器人操作的1-bit视觉-语言-动作模型)

[08:42] 🧠 Pre-trained Large Language Models Learn Hidden Markov Models In-context(预训练大语言模型上下文学习隐马尔可夫模型)

[09:25] 🤔 The Illusion of Thinking: Understanding the Strengths and Limitations of Reasoning Models via the Lens of Problem Complexity(思考的幻觉:通过问题复杂性的视角理解推理模型的优势与局限性)

[10:04] 🧠 CCI4.0: A Bilingual Pretraining Dataset for Enhancing Reasoning in Large Language Models(CCI4.0:用于增强大型语言模型推理能力的双语预训练数据集) 【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