cs.AI（2025-07-03）

📊 共 21 篇论文 | 🔗 2 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (18 🔗2) 支柱二：RL算法与架构 (RL & Architecture) (2) 支柱三：空间感知与语义 (Perception & Semantics) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (18 篇)

#	题目	一句话要点	标签	🔗	⭐
1	MateInfoUB: A Real-World Benchmark for Testing LLMs in Competitive, Multilingual, and Multimodal Educational Tasks	MateInfoUB：用于评估LLM在多语言多模态竞赛教育任务中的真实世界基准	large language model multimodal
2	SI-Agent: An Agentic Framework for Feedback-Driven Generation and Tuning of Human-Readable System Instructions for Large Language Models	SI-Agent：一种基于反馈驱动的Agent框架，用于生成和优化LLM的人类可读系统指令	large language model
3	Self-DANA: A Resource-Efficient Channel-Adaptive Self-Supervised Approach for ECG Foundation Models	提出Self-DANA，一种资源高效的通道自适应自监督心电图基础模型方法	foundation model
4	Strategic Intelligence in Large Language Models: Evidence from evolutionary Game Theory	利用演化博弈论评估大型语言模型的战略智能水平	large language model
5	Autonomous Control Leveraging LLMs: An Agentic Framework for Next-Generation Industrial Automation	提出基于LLM的Agent框架，用于下一代工业自动化中的离散规划与连续控制统一。	large language model instruction following
6	Symbiosis: Multi-Adapter Inference and Fine-Tuning	Symbiosis：多适配器推理与微调框架，解决资源管理和隐私问题	large language model
7	Bourbaki: Self-Generated and Goal-Conditioned MDPs for Theorem Proving	提出基于自生成目标条件MDP的Bourbaki框架，用于提升定理证明能力	large language model
8	Content filtering methods for music recommendation: A review	综述：面向音乐推荐的内容过滤方法，解决交互稀疏性问题	large language model
9	The Impact of LLM-Assistants on Software Developer Productivity: A Systematic Literature Review	系统性文献综述揭示LLM助手对软件开发者生产力的影响：收益与风险并存	large language model
10	LLM-Driven Auto Configuration for Transient IoT Device Collaboration	CollabIoT：利用LLM驱动的自动配置实现瞬态IoT设备的安全协作	large language model
11	Moral Responsibility or Obedience: What Do We Want from AI?	重新审视AI安全：从服从到道德推理的范式转变	large language model
12	Knowledge Protocol Engineering: A New Paradigm for AI in Domain-Specific Knowledge Work	提出知识协议工程(KPE)，赋能LLM在特定领域知识工作中进行深度推理。	large language model
13	Meta SecAlign: A Secure Foundation LLM Against Prompt Injection Attacks	Meta SecAlign：首个完全开源的、防御提示注入攻击的安全LLM	instruction following	✅
14	LLMs and Fuzzing in Tandem: A New Approach to Automatically Generating Weakest Preconditions	提出结合LLM与模糊测试的Fuzzing Guidance方法，自动生成最弱前置条件	large language model
15	Hey AI, Generate Me a Hardware Code! Agentic AI-based Hardware Design & Verification	提出基于Agentic AI的硬件设计与验证方法，提升效率和覆盖率。	large language model
16	FlowSpec: Continuous Pipelined Speculative Decoding for Efficient Distributed LLM Inference	FlowSpec：面向高效分布式LLM推理的连续流水线式推测解码框架	large language model	✅
17	Improving LLM Reasoning for Vulnerability Detection via Group Relative Policy Optimization	提出基于GRPO的LLM微调方法，提升软件漏洞检测的推理能力	large language model
18	Clarifying Before Reasoning: A Coq Prover with Structural Context	提出基于结构化上下文的Coq定理证明器，显著提升LLM推理能力	large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (2 篇)

#	题目	一句话要点	标签	🔗	⭐
19	StepHint: Multi-level Stepwise Hints Enhance Reinforcement Learning to Reason	StepHint：多层次逐步提示增强强化学习的推理能力	reinforcement learning large language model
20	ASDA: Audio Spectrogram Differential Attention Mechanism for Self-Supervised Representation Learning	提出ASDA：音频频谱差分注意力机制，提升自监督表征学习性能	representation learning

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

#	题目	一句话要点	标签	🔗	⭐
21	Early Signs of Steganographic Capabilities in Frontier LLMs	评估前沿LLM的隐写能力，揭示其潜在的安全风险	affordance large language model

⬅️ 返回 cs.AI 首页 · 🏠 返回主页