cs.AI（2024-10-14）

📊 共 22 篇论文 | 🔗 4 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (16 🔗3) 支柱二：RL算法与架构 (RL & Architecture) (5 🔗1) 支柱一：机器人控制 (Robot Control) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (16 篇)

#题目一句话要点标签🔗⭐
1 A Scalable Communication Protocol for Networks of Large Language Models 提出Agora元协议，解决大规模LLM网络通信的效率与灵活性问题 large language model
2 From Anchors to Answers: A Novel Node Tokenizer for Integrating Graph Structure into Large Language Models 提出NT-LLM，通过锚点位置编码将图结构融入大语言模型，提升图推理能力。 large language model
3 Embedding Self-Correction as an Inherent Ability in Large Language Models for Enhanced Mathematical Reasoning 提出链式自我修正机制以提升大语言模型的数学推理能力 large language model
4 Enhancing Retrieval-Augmented Audio Captioning with Generation-Assisted Multimodal Querying and Progressive Learning 提出生成辅助多模态查询和渐进学习，提升检索增强的音频描述生成效果。 multimodal
5 Beyond Graphs: Can Large Language Models Comprehend Hypergraphs? 提出LLM4Hypergraph基准，评估并提升大语言模型对超图的理解能力 large language model ✅
6 FLARE: Faithful Logic-Aided Reasoning and Exploration 提出FLARE以解决大型语言模型推理过程的可信性问题 large language model chain-of-thought
7 CoMAT: Chain of Mathematically Annotated Thought Improves Mathematical Reasoning 提出CoMAT：通过数学符号标注的思维链提升LLM的数学推理能力 large language model chain-of-thought
8 VisRAG: Vision-based Retrieval-augmented Generation on Multi-modality Documents 提出VisRAG，一种基于视觉的多模态文档检索增强生成方法，提升信息利用率。 large language model ✅
9 AFlow: Automating Agentic Workflow Generation AFlow：自动化Agentic工作流生成框架，提升LLM复杂任务解决能力 large language model ✅
10 SensorBench: Benchmarking LLMs in Coding-Based Sensor Processing 提出SensorBench，用于评估LLM在基于代码的传感器数据处理中的性能 large language model
11 Overcoming classic challenges for artificial neural networks by providing incentives and practice 通过激励与实践，元学习克服了人工神经网络的经典挑战 large language model
12 Can Structured Data Reduce Epistemic Uncertainty? 利用本体对齐减少认知不确定性，提升深度学习模型性能及RAG效果 large language model
13 SeCodePLT: A Unified Platform for Evaluating the Security of Code GenAI SeCodePLT：用于评估代码生成AI安全性的统一平台，解决现有基准的局限性。 large language model
14 Focused ReAct: Improving ReAct through Reiterate and Early Stop 提出Focused ReAct，通过迭代和早停机制提升ReAct在复杂任务中的性能。 large language model
15 TMGBench: A Systematic Game Benchmark for Evaluating Strategic Reasoning Abilities of LLMs TMGBench：一个系统性的博弈论基准，用于评估大语言模型的策略推理能力 large language model
16 Optimizing Instruction Synthesis: Effective Exploration of Evolutionary Space with Tree Search 提出IDEA-MCTS框架以优化指令合成过程 instruction following

🔬 支柱二：RL算法与架构 (RL & Architecture) (5 篇)

#题目一句话要点标签🔗⭐
17 Innovative Thinking, Infinite Humor: Humor Research of Large Language Models through Structured Thought Leaps LoL框架：通过结构化思维跃迁提升大语言模型幽默感 reinforcement learning HuMoR large language model
18 Large Language Model-Enhanced Reinforcement Learning for Generic Bus Holding Control Strategies 提出LLM增强的强化学习方法，用于通用公交车保持控制策略优化。 reinforcement learning large language model
19 Compositional Shielding and Reinforcement Learning for Multi-Agent Systems 提出基于组合屏蔽与强化学习的多智能体安全策略方法 reinforcement learning deep reinforcement learning
20 CleanUMamba: A Compact Mamba Network for Speech Denoising using Channel Pruning CleanUMamba：一种用于语音降噪的紧凑型Mamba网络，采用通道剪枝实现高效降噪。 Mamba ✅
21 QUIS: Question-guided Insights Generation for Automated Exploratory Data Analysis QUIS：基于问题引导的洞察生成，用于自动化探索性数据分析 reinforcement learning large language model

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#题目一句话要点标签🔗⭐
22 VideoAgent: Self-Improving Video Generation VideoAgent：基于外部反馈自提升的视频生成方法，用于机器人控制。 manipulation

⬅️ 返回 cs.AI 首页 · 🏠 返回主页