cs.AI（2025-09-11）

📊 共 20 篇论文 | 🔗 3 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (12 🔗3) 支柱二：RL算法与架构 (RL & Architecture) (5) 支柱五：交互与反应 (Interaction & Reaction) (1) 支柱七：动作重定向 (Motion Retargeting) (1) 支柱六：视频提取与匹配 (Video Extraction) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (12 篇)

#题目一句话要点标签🔗⭐
1 A Modular and Multimodal Generative AI Framework for Urban Building Energy Data: Generating Synthetic Homes 提出模块化多模态生成AI框架，用于生成城市建筑能源数据，合成住宅信息。 multimodal
2 Boosting Embodied AI Agents through Perception-Generation Disaggregation and Asynchronous Pipeline Execution Auras：通过解耦感知-生成和异步流水线执行加速具身智能体 embodied AI
3 LoCoBench: A Benchmark for Long-Context Large Language Models in Complex Software Engineering LoCoBench：用于评估长上下文LLM在复杂软件工程中性能的基准测试 large language model ✅
4 Quality Assessment of Tabular Data using Large Language Models and Code Generation 提出基于大语言模型和代码生成的表格数据质量评估框架 large language model
5 Vibe Check: Understanding the Effects of LLM-Based Conversational Agents' Personality and Alignment on User Perceptions in Goal-Oriented Tasks 研究LLM对话Agent人格表达与用户匹配度对目标导向任务用户感知的影响 large language model
6 LLMs as Agentic Cooperative Players in Multiplayer UNO 提出基于LLM的合作型玩家在UNO游戏中的应用 large language model
7 Towards a Common Framework for Autoformalization 提出Autoformalization通用框架，促进AI系统在形式化推理领域的交叉融合。 large language model
8 The Illusion of Diminishing Returns: Measuring Long Horizon Execution in LLMs 揭示LLM长程执行的衰减假象：通过隔离执行能力评估模型性能 large language model
9 TORSO: Template-Oriented Reasoning Towards General Tasks 提出TORSO：一种面向模板推理的通用任务解决框架，无需人工设计的few-shot示例。 large language model
10 Towards Adaptive ML Benchmarks: Web-Agent-Driven Construction, Domain Expansion, and Metric Optimization 提出TAM Bench：一个基于Web Agent驱动的自适应机器学习基准，用于评估LLM在端到端ML任务中的能力。 large language model
11 LightAgent: Production-level Open-source Agentic AI Framework 提出LightAgent：一个生产级开源Agentic AI框架，旨在简化多智能体系统部署。 large language model ✅
12 Jupiter: Enhancing LLM Data Analysis Capabilities via Notebook and Inference-Time Value-Guided Search Jupiter：通过Notebook和推理时价值引导搜索增强LLM数据分析能力 large language model ✅

🔬 支柱二：RL算法与架构 (RL & Architecture) (5 篇)

#题目一句话要点标签🔗⭐
13 Curriculum-Based Multi-Tier Semantic Exploration via Deep Reinforcement Learning 提出基于课程学习的多层语义探索深度强化学习方法，提升具身智能体在未知环境中的探索效率。 reinforcement learning deep reinforcement learning DRL
14 How well can LLMs provide planning feedback in grounded environments? 评估LLM在具身环境中提供规划反馈的能力，揭示其优势与局限 policy learning reward design large language model
15 Tree-OPO: Off-policy Monte Carlo Tree-Guided Advantage Optimization for Multistep Reasoning 提出Tree-OPO，利用MCTS指导优势函数优化，提升LLM多步推理能力 reinforcement learning policy learning large language model
16 SWE-Effi: Re-Evaluating Software AI Agent System Effectiveness Under Resource Constraints 提出SWE-Effi以解决软件工程AI系统资源约束下的有效性评估问题 reinforcement learning large language model
17 Adaptive Knowledge Distillation using a Device-Aware Teacher for Low-Complexity Acoustic Scene Classification 提出基于设备感知教师的自适应知识蒸馏方法，用于低复杂度声场景分类。 distillation

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

#题目一句话要点标签🔗⭐
18 ENSI: Efficient Non-Interactive Secure Inference for Large Language Models ENSI：面向大语言模型的高效非交互安全推理框架 OMOMO large language model

🔬 支柱七：动作重定向 (Motion Retargeting) (1 篇)

#题目一句话要点标签🔗⭐
19 ProgD: Progressive Multi-scale Decoding with Dynamic Graphs for Joint Multi-agent Motion Forecasting ProgD：基于动态图的渐进多尺度解码，用于多智能体联合运动预测 motion prediction

🔬 支柱六：视频提取与匹配 (Video Extraction) (1 篇)

#题目一句话要点标签🔗⭐
20 Mind Meets Space: Rethinking Agentic Spatial Intelligence from a Neuroscience-inspired Perspective 提出神经科学启发的Agentic空间智能框架，提升智能体在3D环境中的推理能力 egocentric multimodal

⬅️ 返回 cs.AI 首页 · 🏠 返回主页