cs.AI（2025-05-06）

📊 共 32 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 🔗1) 支柱二：RL算法与架构 (RL & Architecture) (7 🔗2) 支柱五：交互与反应 (Interaction & Reaction) (1 🔗1) 支柱三：空间感知与语义 (Perception & Semantics) (1) 支柱一：机器人控制 (Robot Control) (1)

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

#题目一句话要点标签🔗⭐
1 Beyond Retrieval: Improving Evidence Quality for LLM-based Multimodal Fact-Checking Aletheia：提升证据质量，增强LLM在多模态事实核查中的表现 large language model multimodal
2 Can Large Language Models Predict Parallel Code Performance? 利用大型语言模型预测并行GPU代码性能，无需硬件剖析 large language model
3 LogiDebrief: A Signal-Temporal Logic based Automated Debriefing Approach with Large Language Models Integration LogiDebrief：结合时序逻辑与大语言模型的自动化9-1-1呼叫评估框架 large language model
4 OSUniverse: Benchmark for Multimodal GUI-navigation AI Agents OSUniverse：用于多模态GUI导航AI代理的基准测试平台 multimodal ✅
5 Validating the Effectiveness of a Large Language Model-based Approach for Identifying Children's Development across Various Free Play Settings in Kindergarten 提出基于大语言模型的儿童自由玩耍场景发展能力评估方法 large language model
6 Synthline: A Product Line Approach for Synthetic Requirements Engineering Data Generation using Large Language Models Synthline：一种基于产品线的大语言模型合成需求工程数据生成方法 large language model
7 LlamaFirewall: An open source guardrail system for building secure AI agents LlamaFirewall：用于构建安全AI Agent的开源安全防护系统 large language model chain-of-thought
8 AI-Driven Scholarly Peer Review via Persistent Workflow Prompting, Meta-Prompting, and Meta-Reasoning 提出持久工作流提示（PWP）方法，提升LLM在学术同行评议中的表现 large language model multimodal
9 Capability-Driven Skill Generation with LLMs: A RAG-Based Approach for Reusing Existing Libraries and Interfaces 提出基于RAG的LLM能力驱动技能生成方法，复用现有库和接口 large language model
10 Prism: Unleashing GPU Sharing for Cost-Efficient Multi-LLM Serving Prism：释放GPU共享潜力，实现多LLM服务的高性价比 large language model
11 Binding threshold units with artificial oscillatory neurons 提出Hopfield-Kuramoto关联记忆模型，结合阈值单元与振荡神经元，实现低秩权重校正。 large language model
12 am-ELO: A Stable Framework for Arena-based LLM Evaluation 提出am-ELO，一个基于竞技场的稳定LLM评估框架，解决ELO系统的不稳定性问题。 large language model
13 Procedural Memory Is Not All You Need: Bridging Cognitive Gaps in LLM-Based Agents 提出LLM智能体认知增强框架，弥合程序记忆与复杂环境适应性差距 large language model
14 Graph Drawing for LLMs: An Empirical Evaluation 研究图布局对LLM图任务性能的影响，优化视觉模态输入 large language model
15 A Hashgraph-Inspired Consensus Mechanism for Reliable Multi-Model Reasoning 提出Hashgraph启发的共识机制，提升多模型推理的可靠性 large language model
16 STORY2GAME: Generating (Almost) Everything in an Interactive Fiction Game STORY2GAME：利用大型语言模型生成交互式小说游戏，实现故事、世界和游戏逻辑的自动构建。 large language model
17 Domain Adversarial Training for Mitigating Gender Bias in Speech-based Mental Health Detection 提出基于领域对抗训练的语音心理健康检测方法，缓解性别偏见。 foundation model
18 RAG-MCP: Mitigating Prompt Bloat in LLM Tool Selection via Retrieval-Augmented Generation RAG-MCP：通过检索增强生成缓解LLM工具选择中的Prompt膨胀问题 large language model
19 Accelerating Evolution: Integrating PSO Principles into Real-Coded Genetic Algorithm Crossover 提出一种受粒子群优化启发的交叉算子PSOX，加速实数编码遗传算法的收敛。 multimodal
20 DocSpiral: A Platform for Integrated Assistive Document Annotation through Human-in-the-Spiral DocSpiral平台：通过人机协同循环加速图像文档结构化标注。 large language model
21 Patterns and Mechanisms of Contrastive Activation Engineering 对比激活工程(CAE)调控大语言模型，但存在分布外失效、易受攻击等问题。 large language model
22 Assessing and Enhancing the Robustness of LLM-based Multi-Agent Systems Through Chaos Engineering 提出基于混沌工程的框架，提升LLM多智能体系统在真实环境下的鲁棒性。 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (7 篇)

#题目一句话要点标签🔗⭐
23 Artificial Behavior Intelligence: Technology, Challenges, and Future Directions 提出人工行为智能(ABI)框架，利用大模型提升行为识别精度与可解释性，应用于多种AI场景。 distillation large language model foundation model
24 X-Reasoner: Towards Generalizable Reasoning Across Modalities and Domains 提出X-Reasoner，通过通用领域文本后训练实现跨模态和领域的推理泛化。 reinforcement learning multimodal chain-of-thought
25 GRAPE: Heterogeneous Graph Representation Learning for Genetic Perturbation with Coding and Non-Coding Biotype GRAPE：结合编码与非编码生物型的基因扰动异构图表示学习 representation learning large language model
26 Mamba-Diffusion Model with Learnable Wavelet for Controllable Symbolic Music Generation 提出基于Transformer-Mamba块和可学习小波变换的扩散模型，用于可控的符号音乐生成。 Mamba classifier-free guidance ✅
27 Decentralized Distributed Proximal Policy Optimization (DD-PPO) for High Performance Computing Scheduling on Multi-User Systems 提出DD-PPO算法，用于多用户HPC系统高性能计算任务调度，提升效率。 reinforcement learning PPO
28 The Steganographic Potentials of Language Models 探索语言模型的隐写潜力，揭示其在隐藏信息方面的能力与风险 reinforcement learning large language model
29 Rainbow Delay Compensation: A Multi-Agent Reinforcement Learning Framework for Mitigating Delayed Observation 提出Rainbow Delay Compensation框架，解决多智能体系统中观测延迟问题 reinforcement learning ✅

🔬 支柱五：交互与反应 (Interaction & Reaction) (1 篇)

#题目一句话要点标签🔗⭐
30 CombiBench: Benchmarking LLM Capability for Combinatorial Mathematics 提出CombiBench，用于评估LLM在组合数学问题上的推理能力 IMoS large language model ✅

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

#题目一句话要点标签🔗⭐
31 Augmenting Human Cognition through Everyday AR 利用日常AR增强人类认知，打造直观的“思考工具” affordance multimodal

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#题目一句话要点标签🔗⭐
32 Synthesizing Images on Perceptual Boundaries of ANNs for Uncovering and Manipulating Human Perceptual Variability 提出BAM框架，通过合成图像边界样本揭示并操控人类感知变异性 manipulation

⬅️ 返回 cs.AI 首页 · 🏠 返回主页