cs.AI（2025-01-14）

📊 共 22 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) 支柱一：机器人控制 (Robot Control) (1) 支柱八：物理动画 (Physics-based Animation) (1)

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

#	题目	一句话要点	标签	🔗	⭐
1	CG-MER: A Card Game-based Multimodal dataset for Emotion Recognition	CG-MER：一个基于卡牌游戏的多模态情感识别法语数据集	multimodal
2	SEAL: Speaker Error Correction using Acoustic-conditioned Large Language Models	SEAL：利用声学条件大语言模型进行说话人错误纠正	large language model
3	Evaluating Computational Accuracy of Large Language Models in Numerical Reasoning Tasks for Healthcare Applications	评估大型语言模型在医疗数值推理任务中的计算准确性	large language model
4	I Can Find You in Seconds! Leveraging Large Language Models for Code Authorship Attribution	利用大型语言模型进行代码作者身份溯源，实现快速精准定位	large language model
5	Hierarchical Autoscaling for Large Language Model Serving with Chiron	Chiron：面向LLM服务的层级自适应伸缩方案，优化SLO达成率和GPU效率。	large language model
6	Large Language Model Interface for Home Energy Management Systems	提出基于大语言模型的界面以优化家庭能源管理系统	large language model
7	The Invisible Hand: Unveiling Provider Bias in Large Language Models for Code Generation	揭示大型语言模型代码生成中的服务提供商偏见问题	large language model
8	A Driver Advisory System Based on Large Language Model for High-speed Train	提出基于大语言模型的智能驾驶咨询系统，提升高铁故障处理能力	large language model
9	Agent-Centric Projection of Prompting Techniques and Implications for Synthetic Training Data for Large Language Models	提出基于Agent视角的Prompting技术框架，为LLM合成训练数据提供新思路	large language model
10	ADAM: An AI Reasoning and Bioinformatics Model for Alzheimer's Disease Detection and Microbiome-Clinical Data Integration	ADAM：一种用于阿尔茨海默病检测和多模态数据融合的AI推理模型	large language model multimodal
11	LLM-Ehnanced Holonic Architecture for Ad-Hoc Scalable SoS	提出LLM增强的Holonic架构，用于自适应可扩展的SoS系统，解决互操作性与重配置难题。	large language model multimodal
12	Visual Language Models as Operator Agents in the Space Domain	提出基于视觉语言模型（VLM）的航天领域操作代理，用于软件仿真和硬件控制。	large language model multimodal
13	Towards Best Practices for Open Datasets for LLM Training	针对LLM训练，提出开放数据集的最佳实践方案，旨在提高透明度、责任性和创新性。	large language model
14	Flow: Modularized Agentic Workflow Automation	Flow：模块化Agent工作流自动化框架，通过动态调整和模块化设计提升效率。	large language model	✅
15	PRESERVE: Prefetching Model Weights and KV-Cache in Distributed LLM Serving	PRESERVE：通过预取模型权重和KV-Cache优化分布式LLM服务	large language model
16	A Roadmap to Guide the Integration of LLMs in Hierarchical Planning	提出LLM融入层级规划的路线图，并构建基准数据集与初步实验。	large language model
17	Self-Instruct Few-Shot Jailbreaking: Decompose the Attack into Pattern and Behavior Learning	提出Self-Instruct-FSJ框架，通过分解攻击模式和行为学习，提升少样本越狱LLM的效率。	large language model	✅
18	Leveraging Metamemory Mechanisms for Enhanced Data-Free Code Generation in LLMs	提出M^2WF框架，利用元记忆机制增强LLM在无数据环境下的代码生成能力	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	Optimization of Link Configuration for Satellite Communication Using Reinforcement Learning	利用强化学习优化卫星通信链路配置，对比PPO与模拟退火算法性能。	reinforcement learning PPO
20	Self-Attentive Spatio-Temporal Calibration for Precise Intermediate Layer Matching in ANN-to-SNN Distillation	提出自注意力时空校准方法，解决ANN到SNN蒸馏中的中间层匹配问题	distillation

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#	题目	一句话要点	标签	🔗	⭐
21	Playing Devil's Advocate: Unmasking Toxicity and Vulnerabilities in Large Vision-Language Models	通过对抗性提示揭示大型视觉语言模型中的毒性和脆弱性	manipulation HuMoR multimodal

🔬 支柱八：物理动画 (Physics-based Animation) (1 篇)

#	题目	一句话要点	标签	🔗	⭐
22	Training Hybrid Neural Networks with Multimode Optical Nonlinearities Using Digital Twins	利用数字孪生训练混合神经网络，结合多模态光学非线性实现高效AI。	PULSE

⬅️ 返回 cs.AI 首页 · 🏠 返回主页