cs.AI（2024-06-20）

📊 共 31 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) (19 🔗3) 支柱二：RL算法与架构 (RL & Architecture) (10 🔗1) 支柱一：机器人控制 (Robot Control) (2)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (19 篇)

#题目一句话要点标签🔗⭐
1 IWISDM: Assessing instruction following in multimodal models at scale 提出iWISDM：大规模评估多模态模型指令遵循能力的基准 large language model multimodal instruction following ✅
2 PIN: A Knowledge-Intensive Dataset for Paired and Interleaved Multimodal Documents PIN：面向配对与交错多模态文档的知识密集型数据集，促进LMMs发展 multimodal
3 FVEL: Interactive Formal Verification Environment with Large Language Models via Theorem Proving 提出FVEL以解决形式验证中的灵活性与效率问题 large language model ✅
4 CityBench: Evaluating the Capabilities of Large Language Models for Urban Tasks CityBench：构建城市任务评估基准，系统评估大语言模型在城市研究中的能力 large language model
5 A Large Language Model Outperforms Other Computational Approaches to the High-Throughput Phenotyping of Physician Notes 利用大型语言模型GPT-4实现医生笔记的高通量表型分析，性能超越传统方法 large language model
6 SORRY-Bench: Systematically Evaluating Large Language Model Safety Refusal SORRY-Bench：系统性评估大型语言模型安全拒绝能力 large language model
7 APEER: Automatic Prompt Engineering Enhances Large Language Model Reranking 提出APEER，通过自动Prompt工程提升大语言模型重排序效果 large language model
8 LiveMind: Low-latency Large Language Models with Simultaneous Inference LiveMind：一种支持同步推理的低延迟大语言模型框架 large language model
9 SPL: A Socratic Playground for Learning Powered by Large Language Model SPL：基于大型语言模型的苏格拉底式学习平台，提升批判性思维。 large language model
10 DASB - Discrete Audio and Speech Benchmark 发布离散音频和语音基准（DASB），用于全面评估各类音频token化方法。 large language model multimodal
11 RE-AdaptIR: Improving Information Retrieval through Reverse Engineered Adaptation 提出RE-AdaptIR，利用逆向工程自适应提升LLM在信息检索中的性能 large language model
12 How critically can an AI think? A framework for evaluating the quality of thinking of generative artificial intelligence 提出MAGE框架，评估生成式AI在模拟批判性思维能力方面的局限性，辅助教育者设计更鲁棒的评估方案。 large language model
13 Does GPT Really Get It? A Hierarchical Scale to Quantify Human vs AI's Understanding of Algorithms 提出算法理解层次结构，量化评估人类与GPT对算法的理解程度 large language model
14 Qiskit HumanEval: An Evaluation Benchmark For Quantum Code Generative Models 提出Qiskit HumanEval量子代码生成评测基准，评估LLM在量子计算领域的代码生成能力 large language model
15 Artificial Leviathan: Exploring Social Evolution of LLM Agents Through the Lens of Hobbesian Social Contract Theory 基于LLM智能体模拟霍布斯社会契约演化，探索复杂社会关系动态形成 large language model
16 The neural correlates of logical-mathematical symbol systems processing resemble that of spatial cognition more than natural language processing 揭示逻辑数学符号处理的神经机制：空间认知或为基础 large language model
17 EvoAgent: Towards Automatic Multi-Agent Generation via Evolutionary Algorithms EvoAgent：通过进化算法实现自动多智能体生成，提升任务解决能力 large language model ✅
18 TurboSpec: Closed-loop Speculation Control System for Optimizing LLM Serving Goodput TurboSpec：闭环推测控制系统优化LLM服务吞吐量 large language model
19 AspirinSum: an Aspect-based utility-preserved de-identification Summarization framework 提出AspirinSum框架，通过基于方面的方法实现效用保持的去标识化摘要。 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (10 篇)

#题目一句话要点标签🔗⭐
20 ReaL: Efficient RLHF Training of Large Language Models with Parameter Reallocation 提出ReaL，通过参数重分配高效训练大型语言模型的RLHF。 reinforcement learning RLHF large language model ✅
21 CityGPT: Empowering Urban Spatial Cognition of Large Language Models CityGPT：通过城市级世界模型增强大语言模型的城市空间认知能力 world model large language model
22 SeCoKD: Aligning Large Language Models for In-Context Learning with Fewer Shots 提出SeCoKD框架，通过自知识蒸馏提升大语言模型少样本上下文学习能力。 distillation large language model
23 Rewarding What Matters: Step-by-Step Reinforcement Learning for Task-Oriented Dialogue 提出基于步进式奖励的强化学习方法，提升面向任务型对话系统的理解与生成能力。 reinforcement learning policy learning
24 Harvesting Efficient On-Demand Order Pooling from Skilled Couriers: Enhancing Graph Representation Learning for Refining Real-time Many-to-One Assignments 提出SCDN模型，利用骑手经验增强图表示学习，优化即时众包订单分配。 representation learning
25 Identifiable Exchangeable Mechanisms for Causal Structure and Representation Learning 提出IEM框架，统一可识别表征学习与因果结构学习，放宽因果结构识别条件。 representation learning
26 Graph Neural Networks for Job Shop Scheduling Problems: A Survey 综述：图神经网络在车间作业调度问题中的应用 reinforcement learning deep reinforcement learning DRL
27 REVEAL-IT: REinforcement learning with Visibility of Evolving Agent poLicy for InTerpretability REVEAL-IT：利用演化策略可见性的强化学习可解释性框架 reinforcement learning
28 Learning telic-controllable state representations 提出Telic-Controllable状态表示学习框架，平衡目标灵活性与认知复杂性 reinforcement learning representation learning
29 Efficient Strategy Learning by Decoupling Searching and Pathfinding for Object Navigation 针对对象导航，提出解耦搜索与寻路的策略学习方法，提升效率。 masked autoencoder MAE

🔬 支柱一：机器人控制 (Robot Control) (2 篇)

#题目一句话要点标签🔗⭐
30 What Teaches Robots to Walk, Teaches Them to Trade too -- Regime Adaptive Execution using Informed Data and LLMs 利用LLM和市场反馈强化学习，解决金融市场动态切换下的交易执行问题 quadruped locomotion reinforcement learning
31 Adversaries Can Misuse Combinations of Safe Models 揭示组合安全模型中的潜在风险：即使单个模型安全，组合使用仍可能被恶意利用。 manipulation

⬅️ 返回 cs.AI 首页 · 🏠 返回主页