cs.LG（2026-04-14）

📊 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) (13 🔗1) 支柱二：RL算法与架构 (RL & Architecture) (8 🔗1) 支柱一：机器人控制 (Robot Control) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (13 篇)

#	题目	一句话要点	标签	🔗	⭐
1	KumoRFM-2: Scaling Foundation Models for Relational Learning	KumoRFM-2：扩展关系学习的基础模型，提升小样本学习性能并支持十亿级数据。	foundation model
2	Cross-Domain Transfer with Particle Physics Foundation Models: From Jets to Neutrino Interactions	提出跨领域转移学习模型以提升粒子物理实验的敏感性	foundation model
3	Scaffold-Conditioned Preference Triplets for Controllable Molecular Optimization with Large Language Models	提出SCPT框架，利用LLM实现支架约束下的可控分子优化	large language model
4	RoleMAG: Learning Neighbor Roles in Multimodal Graphs	提出RoleMAG，学习多模态图中邻居角色以提升跨模态补全性能。	multimodal
5	LLM-Enhanced Log Anomaly Detection: A Comprehensive Benchmark of Large Language Models for Automated System Diagnostics	提出LLM驱动的日志异常检测基准，用于自动化系统诊断。	large language model
6	TimeSAF: Towards LLM-Guided Semantic Asynchronous Fusion for Time Series Forecasting	TimeSAF：面向LLM引导的语义异步融合时间序列预测	large language model zero-shot transfer
7	GCA Framework: A Gulf-Grounded Dataset and Agentic Pipeline for Climate Decision Support	GCA框架：构建海湾地区气候决策支持数据集与智能代理	large language model multimodal
8	Token Encoding for Semantic Recovery	提出TokCode框架，通过token编码实现恶劣信道下可靠的语义恢复。	foundation model
9	Understanding and Improving Continuous Adversarial Training for LLMs via In-context Learning Theory	基于上下文学习理论，改进LLM的连续对抗训练，提升其鲁棒性与实用性。	large language model	✅
10	Interpretable Relational Inference with LLM-Guided Symbolic Dynamics Modeling	COSINE：利用LLM引导的符号动力学建模实现可解释的关系推断	large language model
11	Adaptive Budget Allocation in LLM-Augmented Surveys	提出自适应预算分配算法，优化LLM增强型调查中人工标注资源的利用率。	large language model
12	Analyzing the Effect of Noise in LLM Fine-tuning	研究噪声对LLM微调的影响：揭示不同噪声类型对模型学习动态的影响	large language model
13	PipeLive: Efficient Live In-place Pipeline Parallelism Reconfiguration for Dynamic LLM Serving	PipeLive：用于动态LLM服务的实时、高效、原地流水线并行重配置	large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (8 篇)

#	题目	一句话要点	标签	🔗	⭐
14	Rethinking On-Policy Distillation of Large Language Models: Phenomenology, Mechanism, and Recipe	深入剖析On-Policy蒸馏机制，提出有效策略以提升大语言模型性能	distillation large language model
15	EEG-Based Multimodal Learning via Hyperbolic Mixture-of-Curvature Experts	提出EEG-MoCE，利用双曲混合曲率专家网络进行脑电多模态学习。	representation learning multimodal
16	Lightning OPD: Efficient Post-Training for Large Reasoning Models with Offline On-Policy Distillation	Lightning OPD：通过离线On-Policy蒸馏高效后训练大型推理模型	distillation large language model
17	Nemotron 3 Super: Open, Efficient Mixture-of-Experts Hybrid Mamba-Transformer Model for Agentic Reasoning	Nemotron 3 Super：面向Agentic Reasoning的开放高效混合专家Mamba-Transformer模型	reinforcement learning Mamba
18	SubFlow: Sub-mode Conditioned Flow Matching for Diverse One-Step Generation	提出SubFlow，通过子模态条件Flow Matching解决单步生成模型的多样性退化问题	flow matching	✅
19	From Imitation to Discrimination: Progressive Curriculum Learning for Robust Web Navigation	提出渐进式课程学习以解决网页导航的鲁棒性问题	curriculum learning
20	Labeled TrustSet Guided: Batch Active Learning with Reinforcement Learning	提出TrustSet与强化学习结合的批量主动学习方法以提升数据标注效率	reinforcement learning
21	TCL: Enabling Fast and Efficient Cross-Hardware Tensor Program Optimization via Continual Learning	TCL：通过持续学习实现快速高效的跨硬件张量程序优化	Mamba distillation

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#	题目	一句话要点	标签	🔗	⭐
22	OSC: Hardware Efficient W4A4 Quantization via Outlier Separation in Channel Dimension	提出OSC，通过通道维度异常值分离实现硬件高效的W4A4量化	OSC large language model

⬅️ 返回 cs.LG 首页 · 🏠 返回主页