cs.AI（2024-12-20）

📊 共 20 篇论文 | 🔗 1 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (15) 支柱二：RL算法与架构 (RL & Architecture) (5 🔗1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (15 篇)

#题目一句话要点标签🔗⭐
1 Aria-UI: Visual Grounding for GUI Instructions Aria-UI：提出纯视觉GUI指令理解模型，无需HTML/AXTree输入，实现更强的任务自动化。 multimodal visual grounding
2 Social Science Is Necessary for Operationalizing Socially Responsible Foundation Models 强调社会科学在负责任的基础模型落地中的必要性，构建社会技术系统框架。 foundation model
3 Can LLMs Obfuscate Code? A Systematic Analysis of Large Language Models into Assembly Code Obfuscation 利用LLM生成混淆汇编代码：MetamorphASM基准测试与系统分析 large language model
4 Less is More: Towards Green Code Large Language Models via Unified Structural Pruning 提出Flab-Pruner，通过统一结构剪枝实现绿色代码大语言模型 large language model
5 VirusT5: Harnessing Large Language Models to Predicting SARS-CoV-2 Evolution VirusT5：利用大型语言模型预测SARS-CoV-2病毒进化 large language model
6 AI-in-the-loop: The future of biomedical visual analytics applications in the era of AI 探讨AI驱动下生物医学可视化分析的未来，强调“AI-in-the-loop”的人机协作模式。 large language model foundation model
7 AutoLife: Automatic Life Journaling with Smartphones and LLMs AutoLife：利用智能手机和LLM自动生成生活日志 large language model multimodal
8 Multi-modal Agent Tuning: Building a VLM-Driven Agent for Efficient Tool Usage 提出多模态Agent调优方法，构建VLM驱动的工具高效使用Agent large language model
9 Formal Mathematical Reasoning: A New Frontier in AI 倡导形式化数学推理以推动AI4Math发展 large language model
10 The Evolution of LLM Adoption in Industry Data Curation Practices 探索LLM在工业界数据治理实践中的演进：从启发式到洞察驱动 large language model
11 MetaScientist: A Human-AI Synergistic Framework for Automated Mechanical Metamaterial Design MetaScientist：一种人机协同的自动化机械超材料设计框架 foundation model
12 Trust Calibration in IDEs: Paving the Way for Widespread Adoption of AI Refactoring 在IDE中实现信任校准，促进AI重构的广泛应用 large language model
13 Collaborative Gym: A Framework for Enabling and Evaluating Human-Agent Collaboration 提出Collaborative Gym框架，用于人机协作Agent的开发与评估 large language model
14 Level-Navi Agent: A Framework and benchmark for Chinese Web Search Agents 提出Level-Navi Agent框架与Web24基准，用于评估中文Web搜索Agent能力 large language model
15 JailPO: A Novel Black-box Jailbreak Framework via Preference Optimization against Aligned LLMs 提出JailPO，通过偏好优化实现针对对齐LLM的黑盒越狱攻击 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🔗⭐
16 APIRL: Deep Reinforcement Learning for REST API Fuzzing 提出APIRL以解决REST API模糊测试中的性能与精度问题 reinforcement learning deep reinforcement learning
17 Align Anything: Training All-Modality Models to Follow Instructions with Language Feedback 提出Align-Anything框架，利用语言反馈对齐多模态模型与人类意图 reinforcement learning RLHF large language model ✅
18 AIR: Unifying Individual and Collective Exploration in Cooperative Multi-Agent Reinforcement Learning 提出AIR，统一个体与集体探索，提升合作多智能体强化学习效果 reinforcement learning
19 Tacit Learning with Adaptive Information Selection for Cooperative Multi-Agent Reinforcement Learning 提出基于自适应信息选择的隐式学习框架，解决通信受限多智能体强化学习问题 reinforcement learning
20 Autonomous Option Invention for Continual Hierarchical Reinforcement Learning and Planning 提出一种持续分层强化学习和规划的自主选项发明方法，提升样本效率。 reinforcement learning

⬅️ 返回 cs.AI 首页 · 🏠 返回主页