cs.AI（2024-10-24）

📊 共 14 篇论文 | 🔗 2 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (10) 支柱二：RL算法与架构 (RL & Architecture) (3 🔗1) 支柱四：生成式动作 (Generative Motion) (1 🔗1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (10 篇)

#题目一句话要点标签🔗⭐
1 OSCAR: Operating System Control via State-Aware Reasoning and Re-Planning OSCAR：通过状态感知推理与重规划实现操作系统控制 generalist agent large language model multimodal
2 Demystifying Large Language Models for Medicine: A Primer 针对医疗领域，提出利用大型语言模型（LLMs）的实用指南与最佳实践。 large language model
3 LLM as a code generator in Agile Model Driven Development 提出基于敏捷模型驱动开发（AMDD）的LLM代码生成方法，提升代码生成灵活性和可扩展性。 large language model
4 WASP: A Weight-Space Approach to Detecting Learned Spuriousness WASP：一种权重空间方法，用于检测模型学习到的虚假相关性 foundation model
5 Tailored-LLaMA: Optimizing Few-Shot Learning in Pruned LLaMA Models with Task-Specific Prompts Tailored-LLaMA：利用任务特定提示优化剪枝LLaMA模型中的少样本学习 large language model
6 MMAU: A Massive Multi-Task Audio Understanding and Reasoning Benchmark 提出MMAU：一个大规模多任务音频理解与推理基准 multimodal
7 ReasonAgain: Using Extractable Symbolic Programs to Evaluate Mathematical Reasoning ReasonAgain：利用可提取的符号程序评估数学推理能力 large language model
8 Provably Robust Watermarks for Open-Source Language Models 提出首个开源语言模型可证明鲁棒的水印方案，抵抗参数扰动攻击。 large language model
9 LLM-based Online Prediction of Time-varying Graph Signals 提出基于LLM的时变图信号在线预测框架，提升缺失值填补精度 large language model
10 Smart ETL and LLM-based contents classification: the European Smart Tourism Tools Observatory experience 提出基于智能ETL和LLM的内容分类方法，用于欧洲智能旅游工具观测平台的内容更新。 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (3 篇)

#题目一句话要点标签🔗⭐
11 Improving Small-Scale Large Language Models Function Calling for Reasoning Tasks 提出一种基于RLHF的小型语言模型函数调用优化框架，提升其在推理任务上的性能。 reinforcement learning RLHF DPO
12 Aligning CodeLLMs with Direct Preference Optimization 提出基于DPO的代码大模型对齐方法，提升代码生成任务性能 PPO DPO direct preference optimization
13 SIKeD: Self-guided Iterative Knowledge Distillation for mathematical reasoning 提出SIKeD，通过自引导迭代知识蒸馏提升小模型在数学推理任务上的性能。 distillation large language model ✅

🔬 支柱四：生成式动作 (Generative Motion) (1 篇)

#题目一句话要点标签🔗⭐
14 Scaling up Masked Diffusion Models on Text 提出可扩展的Masked Diffusion模型，在文本生成和理解任务上达到媲美自回归模型的效果。 MDM classifier-free guidance ✅

⬅️ 返回 cs.AI 首页 · 🏠 返回主页