cs.AI（2025-11-21）

📊 共 25 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) (18 🔗2) 支柱二：RL算法与架构 (RL & Architecture) (6) 支柱一：机器人控制 (Robot Control) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (18 篇)

#	题目	一句话要点	标签	🔗	⭐
1	M^3-Bench: Multi-Modal, Multi-Hop, Multi-Threaded Tool-Using MLLM Agent Benchmark	提出M^3-Bench，用于评估多模态工具使用代理在复杂工作流中的性能。	large language model multimodal visual grounding	✅
2	The Rapid Growth of AI Foundation Model Usage in Science	大规模分析揭示AI基础模型在科学领域应用呈指数级增长趋势	foundation model
3	Enhancing Quranic Learning: A Multimodal Deep Learning Approach for Arabic Phoneme Recognition	提出基于Transformer的多模态深度学习框架，用于提升古兰经阿拉伯语音素识别的准确性。	multimodal
4	MusicAIR: A Multimodal AI Music Generation Framework Powered by an Algorithm-Driven Core	MusicAIR：提出一种算法驱动的多模态音乐生成框架，降低版权风险并提升创作效率。	multimodal
5	Episodic Memory in Agentic Frameworks: Suggesting Next Tasks	提出基于情景记忆的Agent框架，辅助LLM进行科研工作流任务推荐	large language model
6	The Impact of Off-Policy Training Data on Probe Generalisation	研究脱策略训练数据对LLM探针泛化能力的影响，揭示意图性行为探针的潜在失效风险。	large language model
7	Learning to Debug: LLM-Organized Knowledge Trees for Solving RTL Assertion Failures	GROVE：利用LLM组织知识树解决RTL断言失败问题	large language model
8	MultiGA: Leveraging Multi-Source Seeding in Genetic Algorithms	MultiGA：利用多源种子的大语言模型遗传算法优化	large language model
9	AI- and Ontology-Based Enhancements to FMEA for Advanced Systems Engineering: Current Developments and Future Directions	利用AI与本体增强FMEA，提升先进系统工程能力	large language model
10	Liberating Logic in the Age of AI: Going Beyond Programming with Computational Thinking	AI时代计算思维解放：超越编程，利用自然语言驾驭计算能力	large language model
11	ARISE: Agentic Rubric-Guided Iterative Survey Engine for Automated Scholarly Paper Generation	ARISE：一种基于Agent和Rubric迭代的学术论文自动生成引擎	large language model	✅
12	The Belief-Desire-Intention Ontology for modelling mental reality and agency	提出BDI本体以解决智能体认知表示不足问题	large language model
13	Datacenters in the Desert: Feasibility and Sustainability of LLM Inference in the Middle East	分析中东沙漠地区部署LLM推理数据中心的可行性与可持续性	large language model
14	Budget-Aware Tool-Use Enables Effective Agent Scaling	提出预算感知工具使用框架BATS，有效提升LLM智能体在受限资源下的性能	large language model
15	LLM and Agent-Driven Data Analysis: A Systematic Approach for Enterprise Applications and System-level Deployment	提出基于LLM和Agent的企业数据分析系统，提升数据访问效率与安全性	large language model
16	Bridging Symbolic Control and Neural Reasoning in LLM Agents: The Structured Cognitive Loop	提出结构化认知循环SCL，提升LLM Agent的可靠性、可解释性和可控性	large language model
17	Cognitive Inception: Agentic Reasoning against Visual Deceptions by Injecting Skepticism	提出Inception框架以解决视觉欺骗问题	large language model
18	Empa: An AI-Powered Virtual Mentor for Developing Global Collaboration Skills in HPC Education	Empa：AI驱动的虚拟导师，提升HPC教育中的全球协作技能	large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (6 篇)

#	题目	一句话要点	标签	🔗	⭐
19	CLLMRec: LLM-powered Cognitive-Aware Concept Recommendation via Semantic Alignment and Prerequisite Knowledge Distillation	CLLMRec：基于LLM的认知感知概念推荐，通过语义对齐和先验知识蒸馏实现	teacher-student distillation large language model
20	QoS-Aware Dynamic CU Selection in O-RAN with Graph-Based Reinforcement Learning	提出基于图神经网络强化学习的O-RAN动态CU选择方法，优化QoS和能耗。	reinforcement learning deep reinforcement learning DRL
21	Morality in AI. A plea to embed morality in LLM architectures and frameworks	提出将道德嵌入LLM架构与框架，提升AI伦理决策能力	reinforcement learning linear attention large language model
22	Quantum Masked Autoencoders for Vision Learning	提出量子掩码自编码器(QMAE)，提升图像掩码重建和分类精度。	masked autoencoder
23	MIR: Efficient Exploration in Episodic Multi-Agent Reinforcement Learning via Mutual Intrinsic Reward	提出互利内在奖励(MIR)方法，解决多智能体强化学习中稀疏奖励探索难题。	reinforcement learning
24	Hybrid Differential Reward: Combining Temporal Difference and Action Gradients for Efficient Multi-Agent Reinforcement Learning in Cooperative Driving	提出混合差分奖励机制，解决合作驾驶中多智能体强化学习的奖励稀疏问题。	reinforcement learning

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#	题目	一句话要点	标签	🔗	⭐
25	Evaluating Adversarial Vulnerabilities in Modern Large Language Models	评估大型语言模型中的对抗性漏洞：针对Gemini和GPT-4的越狱攻击分析	manipulation large language model

⬅️ 返回 cs.AI 首页 · 🏠 返回主页