cs.LG（2025-09-20）

📊 共 19 篇论文 | 🔗 2 篇有代码

🎯 兴趣领域导航

支柱二：RL算法与架构 (RL & Architecture) (9 🔗2) 支柱九：具身大模型 (Embodied Foundation Models) (9) 支柱一：机器人控制 (Robot Control) (1)

🔬 支柱二：RL算法与架构 (RL & Architecture) (9 篇)

#题目一句话要点标签🔗⭐
1 Conversational Orientation Reasoning: Egocentric-to-Allocentric Navigation with Multimodal Chain-of-Thought 提出多模态链式思考框架，解决复杂环境下基于语音和地标的室内定位问题。 curriculum learning egocentric multimodal
2 HypeMARL: Multi-Agent Reinforcement Learning For High-Dimensional, Parametric, and Distributed Systems HypeMARL：面向高维参数化分布式系统的多智能体强化学习 reinforcement learning deep reinforcement learning
3 Towards Universal Debiasing for Language Models-based Tabular Data Generation 提出通用去偏见框架以解决语言模型生成表格数据中的公平性问题 DPO direct preference optimization large language model
4 Learning from Observation: A Survey of Recent Advances 综述学习自观察模仿学习：回顾最新进展与未来方向 offline RL imitation learning model-based RL
5 Knowledge Distillation for Variational Quantum Convolutional Neural Networks on Heterogeneous Data 提出一种异构数据下VQCNN知识蒸馏框架，解决分布式量子机器学习中的模型聚合难题。 distillation
6 $\boldsymbolλ$-Orthogonality Regularization for Compatible Representation Learning 提出λ-正交正则化，用于兼容表征学习，提升模型更新后的零样本性能。 representation learning ✅
7 Causality-Induced Positional Encoding for Transformer-Based Representation Learning of Non-Sequential Features 提出CAPE，通过因果关系建模增强Transformer对非序列化特征的表征学习。 representation learning ✅
8 Bayesian Ego-graph Inference for Networked Multi-Agent Reinforcement Learning 提出BayesG，通过贝叶斯推断学习稀疏交互结构，解决网络化多智能体强化学习问题 reinforcement learning
9 Self-Supervised Learning of Graph Representations for Network Intrusion Detection 提出GraphIDS，通过自监督图表示学习进行网络入侵检测。 representation learning masked autoencoder

🔬 支柱九：具身大模型 (Embodied Foundation Models) (9 篇)

#题目一句话要点标签🔗⭐
10 DISCO: Disentangled Communication Steering for Large Language Models DISCO：面向大语言模型的解耦通信引导，提升控制粒度 large language model
11 Multi-level Diagnosis and Evaluation for Robust Tabular Feature Engineering with Large Language Models 提出多层次诊断评估框架，提升大语言模型在表格特征工程中的鲁棒性 large language model
12 mmExpert: Integrating Large Language Models for Comprehensive mmWave Data Synthesis and Understanding mmExpert：集成大语言模型，实现毫米波数据综合生成与理解 large language model
13 GRIL: Knowledge Graph Retrieval-Integrated Learning with Large Language Models 提出GRIL，一种知识图谱检索增强的LLM学习框架，解决现有方法检索与推理分离及依赖人工标注问题。 large language model
14 Geometric Mixture Classifier (GMC): A Discriminative Per-Class Mixture of Hyperplanes 提出几何混合分类器(GMC)，用每类超平面混合模型解决多模态分类问题。 multimodal
15 SCAN: Self-Denoising Monte Carlo Annotation for Robust Process Reward Learning 提出SCAN自降噪蒙特卡洛标注方法，用于稳健的过程奖励学习。 large language model
16 LLM-Guided Co-Training for Text Classification 提出LLM引导的协同训练方法，提升文本分类在半监督学习中的性能 large language model
17 Federated Learning with Ad-hoc Adapter Insertions: The Case of Soft-Embeddings for Training Classifier-as-Retriever 提出一种基于联邦学习和Ad-hoc适配器插入的检索增强生成方法，用于资源受限的边缘设备。 large language model
18 FairTune: A Bias-Aware Fine-Tuning Framework Towards Fair Heart Rate Prediction from PPG FairTune：一种偏见感知的微调框架，用于从PPG信号中实现公平的心率预测 foundation model

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#题目一句话要点标签🔗⭐
19 Improving User Interface Generation Models from Designer Feedback 提出设计师反馈驱动的UI生成模型，显著提升UI设计质量 manipulation RLHF

⬅️ 返回 cs.LG 首页 · 🏠 返回主页