cs.LG（2025-07-20）

📊 共 14 篇论文 | 🔗 4 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (7 🔗3) 支柱二：RL算法与架构 (RL & Architecture) (5 🔗1) 支柱五：交互与反应 (Interaction & Reaction) (1) 支柱一：机器人控制 (Robot Control) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (7 篇)

#题目一句话要点标签🔗⭐
1 Benchmarking Foundation Models with Multimodal Public Electronic Health Records 构建多模态电子病历基准测试，评估并提升医学Foundation Model性能。 foundation model multimodal ✅
2 MMCircuitEval: A Comprehensive Multimodal Circuit-Focused Benchmark for Evaluating LLMs MMCircuitEval：首个面向电路的多模态大语言模型综合评估基准 large language model multimodal ✅
3 Rethinking Memorization Measures and their Implications in Large Language Models 重新审视大语言模型中的记忆化度量及其影响，揭示现有度量标准的不一致性。 large language model
4 TD-Interpreter: Enhancing the Understanding of Timing Diagrams with Visual-Language Learning TD-Interpreter：利用视觉-语言学习增强时序图理解能力 large language model multimodal
5 Quantizing Text-attributed Graphs for Semantic-Structural Integration 提出STAG框架以解决图结构信息量化问题 large language model zero-shot transfer
6 LibLMFuzz: LLM-Augmented Fuzz Target Generation for Black-box Libraries LibLMFuzz：利用LLM增强的模糊测试目标生成，用于黑盒库漏洞挖掘 large language model
7 MultiKernelBench: A Multi-Platform Benchmark for Kernel Generation MultiKernelBench：首个面向多平台的大模型内核生成综合基准测试 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🔗⭐
8 Reinforcement Learning for Flow-Matching Policies 提出基于强化学习的Flow-Matching策略，提升通用机器人任务性能 reinforcement learning imitation learning flow matching
9 Hierarchical Multi-Agent Reinforcement Learning with Control Barrier Functions for Safety-Critical Autonomous Systems 提出基于控制屏障函数的分层多智能体强化学习方法，用于安全关键自主系统。 reinforcement learning policy learning
10 The Tsetlin Machine Goes Deep: Logical Learning and Reasoning With Graphs 提出Graph Tsetlin Machine，用于图结构数据的可解释深度逻辑学习与推理。 reinforcement learning representation learning multimodal
11 eMargin: Revisiting Contrastive Learning with Margin-Based Separation 提出eMargin：基于边距分离改进对比学习的时间序列表示 representation learning contrastive learning ✅
12 Omni-Thinker: Scaling Multi-Task RL in LLMs with Hybrid Reward and Task Scheduling Omni-Thinker：通过混合奖励与任务调度扩展LLM中的多任务强化学习 reinforcement learning large language model

🔬 支柱五：交互与反应 (Interaction & Reaction) (1 篇)

#题目一句话要点标签🔗⭐
13 A Privacy-Centric Approach: Scalable and Secure Federated Learning Enabled by Hybrid Homomorphic Encryption 提出基于混合同态加密的可扩展安全联邦学习方法 OMOMO

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#题目一句话要点标签🔗⭐
14 Application-Specific Component-Aware Structured Pruning of Deep Neural Networks in Control via Soft Coefficient Optimization 提出一种面向控制的深度神经网络结构化剪枝方法，通过软系数优化实现应用特定性能保持。 model predictive control

⬅️ 返回 cs.LG 首页 · 🏠 返回主页