cs.LG(2024-02-19)

📊 共 22 篇论文 | 🔗 4 篇有代码

🎯 兴趣领域导航

支柱二:RL算法与架构 (RL & Architecture) (10 🔗1) 支柱九:具身大模型 (Embodied Foundation Models) (10 🔗2) 支柱八:物理动画 (Physics-based Animation) (1) 支柱六:视频提取与匹配 (Video Extraction) (1 🔗1)

🔬 支柱二:RL算法与架构 (RL & Architecture) (10 篇)

#题目一句话要点标签🔗
1 A Critical Evaluation of AI Feedback for Aligning Large Language Models 评估AI反馈在大型语言模型对齐中的有效性 reinforcement learning large language model instruction following
2 Revisiting Data Augmentation in Deep Reinforcement Learning 提出数据增强方法以提升深度强化学习的样本效率与泛化能力 reinforcement learning deep reinforcement learning DRL
3 Multimodal Fusion of EHR in Structures and Semantics: Integrating Clinical Records and Notes with Hypergraph and LLM 提出MINGLE框架以解决EHR多模态数据融合问题 representation learning multimodal
4 In value-based deep reinforcement learning, a pruned network is a good network 提出逐步剪枝技术以提升价值基深度强化学习网络性能 reinforcement learning deep reinforcement learning
5 The Edge-of-Reach Problem in Offline Model-Based Reinforcement Learning 提出Reach-Aware Value Learning以解决离边界状态问题 reinforcement learning offline reinforcement learning
6 Reinforcement Learning as a Parsimonious Alternative to Prediction Cascades: A Case Study on Image Segmentation 提出PaSeR以解决低资源环境下图像分割问题 reinforcement learning
7 Offline Multi-task Transfer RL with Representational Penalization 提出一种新算法以解决离线多任务强化学习中的表示转移问题 reinforcement learning offline RL offline reinforcement learning
8 On the Byzantine-Resilience of Distillation-Based Federated Learning 提出新防御机制以增强蒸馏基础联邦学习的拜占庭鲁棒性 distillation
9 DB-LLM: Accurate Dual-Binarization for Efficient LLMs 提出DB-LLM以解决超低位量化带来的准确性下降问题 distillation large language model
10 Induced Model Matching: Restricted Models Help Train Full-Featured Models 提出诱导模型匹配方法以提升全特征模型训练效果 predictive model distillation

🔬 支柱九:具身大模型 (Embodied Foundation Models) (10 篇)

#题目一句话要点标签🔗
11 DeepCode AI Fix: Fixing Security Vulnerabilities with Large Language Models 提出深度代码AI修复方法以解决安全漏洞问题 large language model
12 WKVQuant: Quantizing Weight and Key/Value Cache for Large Language Models Gains More 提出WKVQuant以解决大语言模型的量化问题 large language model
13 Microstructures and Accuracy of Graph Recall by Large Language Models 系统研究大型语言模型的图回忆能力及其偏差特征 large language model
14 Robust CLIP: Unsupervised Adversarial Fine-Tuning of Vision Embeddings for Robust Large Vision-Language Models 提出无监督对抗微调方案以增强CLIP视觉编码器的鲁棒性 foundation model
15 Uncertainty quantification in fine-tuned LLMs using LoRA ensembles 提出低秩适应集成方法以量化微调大语言模型的不确定性 large language model
16 Mafin: Enhancing Black-Box Embeddings with Model Augmented Fine-Tuning 提出Mafin以增强黑箱嵌入模型的微调效果 large language model
17 Self-AMPLIFY: Improving Small Language Models with Self Post Hoc Explanations 提出Self-AMPLIFY以提升小型语言模型性能 large language model
18 LoRA Training in the NTK Regime has No Spurious Local Minima 在NTK范式下,LoRA训练消除虚假局部极小值 large language model
19 UniST: A Prompt-Empowered Universal Model for Urban Spatio-Temporal Prediction 提出UniST以解决城市时空预测的通用性问题 large language model
20 Dynamic Multi-Network Mining of Tensor Time Series 提出动态多网络挖掘方法以解决张量时间序列子序列聚类问题 TAMP

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

#题目一句话要点标签🔗
21 PARCv2: Physics-aware Recurrent Convolutional Neural Networks for Spatiotemporal Dynamics Modeling 提出PARCv2以解决非稳态物理问题建模挑战 spatiotemporal

🔬 支柱六:视频提取与匹配 (Video Extraction) (1 篇)

#题目一句话要点标签🔗
22 Generative Semi-supervised Graph Anomaly Detection 提出生成式半监督图异常检测方法以提升检测性能 egocentric

⬅️ 返回 cs.LG 首页 · 🏠 返回主页