cs.AI（2025-12-30）

📊 共 19 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) (13 🔗2) 支柱二：RL算法与架构 (RL & Architecture) (4) 支柱一：机器人控制 (Robot Control) (2 🔗1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (13 篇)

#题目一句话要点标签🔗⭐
1 A multimodal Transformer for InSAR-based ground deformation forecasting with cross-site generalization across Europe 提出多模态Transformer，用于InSAR地表形变预测并提升跨区域泛化能力 multimodal
2 Thinking on Maps: How Foundation Model Agents Explore, Remember, and Reason Map Environments 提出交互式评估框架，分析大模型智能体在地图环境中的探索、记忆和推理能力 foundation model
3 Cultural Encoding in Large Language Models: The Existence Gap in AI-Mediated Brand Discovery 揭示大语言模型中的文化编码现象，提出数据护城河框架应对品牌AI可见性挑战 large language model
4 CogRec: A Cognitive Recommender Agent Fusing Large Language Models and Soar for Explainable Recommendation CogRec：融合大语言模型与Soar认知架构的可解释推荐智能体 large language model
5 ProSoftArena: Benchmarking Hierarchical Capabilities of Multimodal Agents in Professional Software Environments ProSoftArena：构建专业软件环境多模态Agent能力分级评估基准 multimodal
6 A Proof-of-Concept for Explainable Disease Diagnosis Using Large Language Models and Answer Set Programming McCoy：结合LLM与ASP，实现可解释的疾病诊断概念验证 large language model
7 Evaluating the Reasoning Abilities of LLMs on Underrepresented Mathematics Competition Problems 利用欠代表性数学竞赛题评估大语言模型的推理能力 large language model
8 PackKV: Reducing KV Cache Memory Footprint through LLM-Aware Lossy Compression PackKV：通过LLM感知的有损压缩降低KV缓存内存占用 large language model ✅
9 LoongFlow: Directed Evolutionary Search via a Cognitive Plan-Execute-Summarize Paradigm LoongFlow：基于认知Plan-Execute-Summarize范式的定向进化搜索 large language model
10 Jailbreaking Attacks vs. Content Safety Filters: How Far Are We in the LLM Safety Arms Race? 系统评估LLM安全：关注越狱攻击在完整推理流程中的绕过情况 large language model
11 SPARK: Search Personalization via Agent-Driven Retrieval and Knowledge-sharing SPARK：通过Agent驱动的检索和知识共享实现搜索个性化 large language model
12 TESO Tabu Enhanced Simulation Optimization for Noisy Black Box Problems 提出TESO，一种结合禁忌搜索和精英记忆的噪声黑盒问题优化方法 multimodal ✅
13 Coding With AI: From a Reflection on Industrial Practices to Future Computer Science and Software Engineering Education 基于行业实践反思，探讨AI编码对计算机科学与软件工程教育的未来影响 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (4 篇)

#题目一句话要点标签🔗⭐
14 Deep Reinforcement Learning for Solving the Fleet Size and Mix Vehicle Routing Problem 提出基于深度强化学习的FRIPN网络，解决车队规模和车型组合车辆路径问题 reinforcement learning deep reinforcement learning DRL
15 Automated Classification of First-Trimester Fetal Heart Views Using Ultrasound-Specific Self-Supervised Learning 提出基于超声自监督学习的USF-MAE模型，用于自动分类妊娠早期胎儿心脏视图 MAE foundation model
16 ROAD: Reflective Optimization via Automated Debugging for Zero-Shot Agent Alignment ROAD：通过自动化调试进行反思优化，实现零样本Agent对齐。 reinforcement learning large language model
17 PhyAVBench: A Challenging Audio Physics-Sensitivity Benchmark for Physically Grounded Text-to-Audio-Video Generation 提出PhyAVBench基准，评估文本到音视频生成模型对物理规律的理解能力 world model physically plausible

🔬 支柱一：机器人控制 (Robot Control) (2 篇)

#题目一句话要点标签🔗⭐
18 The Silicon Psyche: Anthropomorphic Vulnerabilities in Large Language Models 提出心理防火墙以应对大型语言模型的脆弱性问题 manipulation large language model
19 What Drives Success in Physical Planning with Joint-Embedding Predictive World Models? 提出基于联合嵌入预测世界模型的物理规划方法，优化模型架构与训练目标。 manipulation world model ✅

⬅️ 返回 cs.AI 首页 · 🏠 返回主页