cs.AI（2025-04-07）

📊 共 35 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) (22 🔗3) 支柱二：RL算法与架构 (RL & Architecture) (12 🔗1) 支柱三：空间感知与语义 (Perception & Semantics) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (22 篇)

#题目一句话要点标签🔗⭐
1 SmolVLM: Redefining small and efficient multimodal models 提出SmolVLM以解决小型多模态模型的资源效率问题 multimodal
2 Large Language Model (LLM) for Software Security: Code Analysis, Malware Analysis, Reverse Engineering 综述性研究：利用大型语言模型提升软件安全，聚焦代码分析、恶意软件分析与逆向工程 large language model
3 The challenge of uncertainty quantification of large language models in medicine 提出一种综合框架，用于量化医学大语言模型的不确定性，提升临床决策的可靠性。 large language model
4 Promoting Security and Trust on Social Networks: Explainable Cyberbullying Detection Using Large Language Models in a Stream-Based Machine Learning Framework 提出基于流式机器学习框架和大型语言模型的可解释网络欺凌检测方案，提升社交网络安全。 large language model
5 Leveraging Label Potential for Enhanced Multimodal Emotion Recognition 提出LSGMER模型，利用标签信息增强多模态情感识别的准确性和稳定性。 multimodal
6 CCSK:Cognitive Convection of Self-Knowledge Based Retrieval Augmentation for Large Language Models 提出CCSK，通过自知识认知对流增强大语言模型的检索增强生成效果 large language model
7 Multimodal Agricultural Agent Architecture (MA3): A New Paradigm for Intelligent Agricultural Decision-Making 提出多模态农业Agent架构MA3，用于智能农业决策，应对气候变化下的生产优化与可持续发展挑战。 multimodal
8 Prism: Dynamic and Flexible Benchmarking of LLMs Code Generation with Monte Carlo Tree Search Prism：基于蒙特卡洛树搜索的LLM代码生成动态灵活基准测试框架 large language model
9 On the Effectiveness and Generalization of Race Representations for Debiasing High-Stakes Decisions 利用种族表征进行偏差校正，但通用性仍具挑战 large language model
10 AccLLM: Accelerating Long-Context LLM Inference Via Algorithm-Hardware Co-Design AccLLM：通过算法-硬件协同设计加速长文本LLM推理 large language model
11 SciSciGPT: Advancing Human-AI Collaboration in the Science of Science SciSciGPT：利用大语言模型赋能科学研究，促进人机协作 large language model
12 A Desideratum for Conversational Agents: Capabilities, Challenges, and Future Directions 构建对话Agent能力图谱，分析挑战与未来方向，助力通用人工智能 large language model ✅
13 Frontier AI's Impact on the Cybersecurity Landscape 前沿AI加剧网络安全攻防失衡，攻击能力超越防御，亟需新基准与防御AI foundation model
14 EduPlanner: LLM-Based Multi-Agent Systems for Customized and Intelligent Instructional Design EduPlanner：基于LLM的多智能体系统，用于定制化和智能化的教学设计 large language model ✅
15 Utility-Focused LLM Annotation for Retrieval and Retrieval-Augmented Generation 利用大语言模型标注文档效用，提升检索和RAG系统性能 large language model
16 Prima.cpp: Fast 30-70B LLM Inference on Heterogeneous and Low-Resource Home Clusters Prima.cpp：在异构低资源家庭集群上实现快速30-70B LLM推理 large language model
17 Debate Only When Necessary: Adaptive Multiagent Collaboration for Efficient LLM Reasoning 提出DOWN框架，通过自适应辩论提升LLM推理效率并降低计算成本 large language model
18 The Dream Within Huang Long Cave: AI-Driven Interactive Narrative for Family Storytelling and Emotional Reflection 提出基于AI的交互叙事系统，用于家庭故事讲述和情感反思 large language model
19 Debate-Feedback: A Multi-Agent Framework for Efficient Legal Judgment Prediction 提出Debate-Feedback框架，利用多智能体辩论高效预测法律判决 large language model
20 BIASINSPECTOR: Detecting Bias in Structured Data through LLM Agents BIASINSPECTOR：利用LLM Agent自动检测结构化数据中的偏见 large language model
21 ELT-Bench: An End-to-End Benchmark for Evaluating AI Agents on ELT Pipelines 提出ELT-Bench，用于评估AI Agent在端到端ELT Pipeline构建中的能力。 large language model ✅
22 Generalising from Self-Produced Data: Model Training Beyond Human Constraints 提出一种AI自主生成数据并训练模型的新框架，突破人类数据和抽象层级的限制。 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (12 篇)

#题目一句话要点标签🔗⭐
23 R2Vul: Learning to Reason about Software Vulnerabilities with Reinforcement Learning and Structured Reasoning Distillation R2Vul：结合强化学习与结构化推理蒸馏提升代码LLM的软件漏洞检测能力 reinforcement learning distillation large language model
24 Deep Reinforcement Learning Algorithms for Option Hedging 对比深度强化学习算法在期权对冲中的表现，MCPG算法表现最佳 reinforcement learning deep reinforcement learning DRL
25 Resource-Efficient Beam Prediction in mmWave Communications with Multimodal Realistic Simulation Framework 提出基于跨模态关系知识蒸馏的毫米波通信波束预测方法，提升资源效率。 distillation multimodal
26 Algorithm Discovery With LLMs: Evolutionary Search Meets Reinforcement Learning 提出基于强化学习微调的LLM进化搜索算法，加速组合优化算法发现 reinforcement learning large language model
27 VAPO: Efficient and Reliable Reinforcement Learning for Advanced Reasoning Tasks VAPO：用于高级推理任务的高效可靠的强化学习框架 reinforcement learning chain-of-thought
28 GAMDTP: Dynamic Trajectory Prediction with Graph Attention Mamba Network 提出GAMDTP以解决动态轨迹预测问题 Mamba SSM
29 Synthetic Data Generation & Multi-Step RL for Reasoning & Tool Use 提出Step-Wise RL，通过合成数据和多步强化学习提升语言模型在推理和工具使用上的性能。 reinforcement learning RLHF large language model
30 HypRL: Reinforcement Learning of Control Policies for Hyperproperties HYPRL：提出一种基于HyperLTL规范引导的多智能体强化学习控制策略框架 reinforcement learning reward shaping
31 Interactive Explanations for Reinforcement-Learning Agents 提出ASQ-IT交互式解释系统，提升用户对强化学习智能体行为的理解和问题定位能力 reinforcement learning
32 Unleashing the Power of LLMs in Dense Retrieval with Query Likelihood Modeling 提出LLM-QL模型，利用查询似然建模增强LLM在稠密检索中的性能 contrastive learning large language model
33 Weak-for-Strong: Training Weak Meta-Agent to Harness Strong Executors 提出W4S框架，利用弱Meta-Agent优化工作流，提升强执行器的性能。 reinforcement learning large language model
34 GOTHAM: Graph Class Incremental Learning Framework under Weak Supervision 提出GOTHAM框架，解决弱监督下图数据的类别增量学习问题。 teacher-student distillation ✅

🔬 支柱三：空间感知与语义 (Perception & Semantics) (1 篇)

#题目一句话要点标签🔗⭐
35 How to evaluate control measures for LLM agents? A trajectory from today to superintelligence 提出LLM Agent控制评估框架，根据Agent能力演进调整红队对抗策略 affordance

⬅️ 返回 cs.AI 首页 · 🏠 返回主页