cs.AI（2025-12-29）

📊 共 16 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) (11) 支柱二：RL算法与架构 (RL & Architecture) (3 🔗3) 支柱一：机器人控制 (Robot Control) (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) (11 篇)

#题目一句话要点标签🔗⭐
1 Agentic Physical AI toward a Domain-Specific Foundation Model for Nuclear Reactor Control 提出Agentic Physical AI，用于核反应堆控制的领域特定基础模型。 foundation model multimodal
2 Toward Trustworthy Agentic AI: A Multimodal Framework for Preventing Prompt Injection Attacks 提出跨Agent多模态溯源防御框架，防范Agentic AI中的提示注入攻击。 large language model multimodal
3 Divergent-Convergent Thinking in Large Language Models for Creative Problem Generation CreativeDC：利用大语言模型中的发散-收敛思维生成多样化创意问题 large language model
4 EquaCode: A Multi-Strategy Jailbreak Approach for Large Language Models via Equation Solving and Code Completion 提出 EquaCode，利用数学方程求解与代码补全实现大语言模型越狱攻击 large language model
5 SPIRAL: Symbolic LLM Planning via Grounded and Reflective Search SPIRAL：通过具身和反思搜索实现符号LLM规划 large language model chain-of-thought
6 AKG kernel Agent: A Multi-Agent Framework for Cross-Platform Kernel Synthesis 提出AKG kernel Agent，一个用于跨平台内核合成的多智能体框架。 multimodal
7 The Gaining Paths to Investment Success: Information-Driven LLM Graph Reasoning for Venture Capital Prediction MIRAGE-VC：信息增益驱动的LLM图推理，用于风险投资预测 large language model
8 Securing the AI Supply Chain: What Can We Learn From Developer-Reported Security Issues and Solutions of AI Projects? 通过分析开发者报告的安全问题与解决方案，提升AI供应链安全性 large language model
9 TCEval: Using Thermal Comfort to Assess Cognitive and Perceptual Abilities of AI 提出TCEval框架以评估AI的认知与感知能力 large language model
10 From Model Choice to Model Belief: Establishing a New Measure for LLM-Based Research 提出“模型置信度”：一种更高效的LLM数据利用方法，提升LLM模拟研究的统计效率。 large language model
11 It's a TRAP! Task-Redirecting Agent Persuasion Benchmark for Web Agents 提出TRAP基准测试，评估Web Agent在提示注入攻击下的任务重定向脆弱性 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (3 篇)

#题目一句话要点标签🔗⭐
12 Replay Failures as Successes: Sample-Efficient Reinforcement Learning for Instruction Following 提出HiR，通过回溯式指令重放提升指令跟随任务中强化学习的样本效率 reinforcement learning preference learning large language model ✅
13 Web World Models 提出Web World Model，结合Web代码的逻辑一致性和LLM的生成能力，构建可控且开放的Agent环境。 world model large language model ✅
14 Alpha-R1: Alpha Screening with LLM Reasoning via Reinforcement Learning 提出Alpha-R1，利用强化学习训练LLM进行上下文感知的Alpha筛选，提升量化投资策略的鲁棒性。 reinforcement learning large language model ✅

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#题目一句话要点标签🔗⭐
15 MindWatcher: Toward Smarter Multimodal Tool-Integrated Reasoning 提出MindWatcher，一种集成多模态工具的智能推理Agent，解决复杂决策问题。 manipulation multimodal chain-of-thought

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

#题目一句话要点标签🔗⭐
16 An Inference-Based Architecture for Intent and Affordance Saturation in Decision-Making 提出基于推理的架构，解决决策中意图和可供性饱和导致的决策瘫痪问题 affordance

⬅️ 返回 cs.AI 首页 · 🏠 返回主页