cs.AI（2024-06-07）

📊 共 13 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) (8 🔗1) 支柱二：RL算法与架构 (RL & Architecture) (4) 支柱一：机器人控制 (Robot Control) (1 🔗1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (8 篇)

#题目一句话要点标签🔗⭐
1 StackSight: Unveiling WebAssembly through Large Language Models and Neurosymbolic Chain-of-Thought Decompilation 提出StackSight以解决WebAssembly反向工程难题 large language model chain-of-thought
2 Zero, Finite, and Infinite Belief History of Theory of Mind Reasoning in Large Language Models 提出基于信念历史的ToM推理框架，评估LLM在多轮文本游戏中的表现 large language model
3 A Language Model-Guided Framework for Mining Time Series with Distributional Shifts 提出一种基于语言模型引导的时间序列挖掘框架，解决分布偏移下的数据稀疏问题。 large language model foundation model
4 LLM-Enhanced Bayesian Optimization for Efficient Analog Layout Constraint Generation 提出LLANA框架，利用LLM增强贝叶斯优化，高效生成模拟版图约束。 large language model ✅
5 LLM Whisperer: An Inconspicuous Attack to Bias LLM Responses LLM Whisperer：一种隐蔽的攻击方式，通过操纵Prompt来偏置LLM的输出。 large language model
6 Enhancing Large-Scale AI Training Efficiency: The C4 Solution for Real-Time Anomaly Detection and Communication Optimization C4：针对大规模AI训练的实时异常检测与通信优化方案 large language model
7 VCSearch: Bridging the Gap Between Well-Defined and Ill-Defined Problems in Mathematical Reasoning 提出VCSEARCH框架，提升LLM在病态数学问题上的鲁棒推理能力 large language model
8 OCDB: Revisiting Causal Discovery with a Comprehensive Benchmark and Evaluation Framework OCDB：构建全面的因果发现基准与评估框架，提升LLM可解释性 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (4 篇)

#题目一句话要点标签🔗⭐
9 ChatPCG: Large Language Model-Driven Reward Design for Procedural Content Generation ChatPCG：基于大语言模型驱动的程序化内容生成奖励设计 reinforcement learning deep reinforcement learning reward design
10 Massively Multiagent Minigames for Training Generalist Agents Meta MMO：大规模多智能体迷你游戏集，用于训练通用智能体 reinforcement learning generalist agent
11 Online Adaptation for Enhancing Imitation Learning Policies 提出在线自适应方法，提升模仿学习策略在复杂任务中的性能 imitation learning
12 Robust Reward Design for Markov Decision Processes 提出一种鲁棒的马尔可夫决策过程奖励设计方法，解决模型不确定性问题。 reward design

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#题目一句话要点标签🔗⭐
13 SLOPE: Search with Learned Optimal Pruning-based Expansion 提出基于学习最优剪枝扩展的搜索算法SLOPE，提升运动规划效率。 motion planning ✅

⬅️ 返回 cs.AI 首页 · 🏠 返回主页