cs.LG（2025-06-30）
📊 共 15 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) (7 🔗1)
支柱二：RL算法与架构 (RL & Architecture) (6 🔗2)
支柱八：物理动画 (Physics-based Animation) (2)
🔬 支柱九：具身大模型 (Embodied Foundation Models) (7 篇)
| # | 题目 | 一句话要点 | 标签 | 🔗 | ⭐ |
|---|---|---|---|---|---|
| 1 | Can We Predict the Unpredictable? Leveraging DisasterNet-LLM for Multimodal Disaster Classification | 提出DisasterNet-LLM以解决多模态灾害分类问题 | large language model multimodal | ||
| 2 | Beyond Sensor Data: Foundation Models of Behavioral Data from Wearables Improve Health Predictions | 提出行为数据基础模型以提升健康预测准确性 | foundation model | ||
| 3 | Chain of Thought in Order: Discovering Learning-Friendly Orders for Arithmetic | 提出学习友好的顺序以优化Transformer的算术推理 | chain-of-thought | ||
| 4 | Are AI-Generated Fixes Secure? Analyzing LLM and Agent Patches on SWE-bench | 分析LLM生成补丁的安全性以应对软件开发中的风险 | large language model | ||
| 5 | Data Uniformity Improves Training Efficiency and More, with a Convergence Framework Beyond the NTK Regime | 提出数据均匀性选择以提升训练效率和性能 | large language model | ✅ | |
| 6 | Agent.xpu: Efficient Scheduling of Agentic LLM Workloads on Heterogeneous SoC | 提出Agent.xpu以高效调度异构SoC上的智能LLM工作负载 | large language model | ||
| 7 | Federated Learning-Enabled Hybrid Language Models for Communication-Efficient Token Transmission | 提出FedHLM以解决边缘设备通信效率低的问题 | large language model |
🔬 支柱二：RL算法与架构 (RL & Architecture) (6 篇)
| # | 题目 | 一句话要点 | 标签 | 🔗 | ⭐ |
|---|---|---|---|---|---|
| 8 | $μ^2$Tokenizer: Differentiable Multi-Scale Multi-Modal Tokenizer for Radiology Report Generation | 提出$μ^2$Tokenizer以解决放射学报告生成中的信息提取与评估问题 | DPO direct preference optimization large language model | ✅ | |
| 9 | Double Q-learning for Value-based Deep Reinforcement Learning, Revisited | 提出深度双Q学习以解决Q学习过度估计问题 | reinforcement learning deep reinforcement learning | ||
| 10 | Teaching Time Series to See and Speak: Forecasting with Aligned Visual and Textual Perspectives | 提出多模态对比学习框架以提升时间序列预测能力 | contrastive learning large language model multimodal | ✅ | |
| 11 | Reinforcement Learning for Synchronised Flow Control in a Dual-Gate Resin Infusion System | 提出基于强化学习的同步流动控制策略以解决树脂注入系统问题 | reinforcement learning PPO | ||
| 12 | Gym4ReaL: A Suite for Benchmarking Real-World Reinforcement Learning | 提出Gym4ReaL以解决现实世界强化学习的基准测试问题 | reinforcement learning | ||
| 13 | Optimizing Conversational Product Recommendation via Reinforcement Learning | 提出基于强化学习的对话产品推荐优化方法 | reinforcement learning |
🔬 支柱八：物理动画 (Physics-based Animation) (2 篇)
| # | 题目 | 一句话要点 | 标签 | 🔗 | ⭐ |
|---|---|---|---|---|---|
| 14 | EFPI: Elastic Formation and Position Identification in Football (Soccer) using Template Matching and Linear Assignment | 提出EFPI方法以解决足球战术分析中的阵型识别问题 | spatiotemporal | ||
| 15 | A Joint Topology-Data Fusion Graph Network for Robust Traffic Speed Prediction with Data Anomalism | 提出GFEN以解决交通速度预测中的数据异常问题 | spatiotemporal |