cs.LG（2025-06-13）

📊 共 32 篇论文 | 🔗 5 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (15 🔗3) 支柱二：RL算法与架构 (RL & Architecture) (13 🔗1) 支柱一：机器人控制 (Robot Control) (2) 支柱八：物理动画 (Physics-based Animation) (1) 支柱五：交互与反应 (Interaction & Reaction) (1 🔗1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (15 篇)

#题目一句话要点标签🔗⭐
1 Explaining Recovery Trajectories of Older Adults Post Lower-Limb Fracture Using Modality-wise Multiview Clustering and Large Language Models 提出多模态聚类与大语言模型以解释老年人下肢骨折恢复轨迹 large language model multimodal
2 RollingQ: Reviving the Cooperation Dynamics in Multimodal Transformer 提出RollingQ以解决多模态Transformer中的合作动态问题 multimodal ✅
3 A Survey of Foundation Models for IoT: Taxonomy and Criteria-Based Analysis 提出基础模型分类与评估标准以解决IoT任务比较难题 foundation model
4 Fed-HeLLo: Efficient Federated Foundation Model Fine-Tuning with Heterogeneous LoRA Allocation 提出Fed-HeLLo以解决异构资源下的联邦模型微调问题 foundation model
5 Learn to Preserve Personality: Federated Foundation Models in Recommendations 提出联邦基础模型以解决个性化推荐中的个性保持问题 foundation model
6 Improving Multimodal Learning Balance and Sufficiency through Data Remixing 提出多模态数据重混合以解决模态不平衡问题 multimodal ✅
7 EMLoC: Emulator-based Memory-efficient Fine-tuning with LoRA Correction 提出EMLoC框架以解决大模型微调的内存开销问题 foundation model
8 Mind the XAI Gap: A Human-Centered LLM Framework for Democratizing Explainable AI 提出人本中心的LLM框架以解决可解释AI的透明性问题 large language model
9 Uncovering Bias Paths with LLM-guided Causal Discovery: An Active Learning and Dynamic Scoring Approach 提出LLM引导的因果发现框架以解决公平性路径识别问题 large language model
10 CLEAN-MI: A Scalable and Efficient Pipeline for Constructing High-Quality Neurodata in Motor Imagery Paradigm 提出CLEAN-MI以解决脑机接口中神经数据构建问题 foundation model
11 SEC-bench: Automated Benchmarking of LLM Agents on Real-World Software Security Tasks 提出SEC-bench以解决LLM代理在软件安全任务中的评估问题 large language model
12 Convergent Linear Representations of Emergent Misalignment 提出新方法以理解和缓解模型的紧急失调现象 large language model
13 Model Organisms for Emergent Misalignment 提出新模型生物以解决新兴不对齐问题 large language model
14 SWE-Bench-CL: Continual Learning for Coding Agents 提出SWE-Bench-CL以解决持续学习中的知识遗忘问题 large language model ✅
15 LoRA Users Beware: A Few Spurious Tokens Can Manipulate Your Finetuned Model 揭示LoRA模型在微调中易受短路攻击的脆弱性 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (13 篇)

#题目一句话要点标签🔗⭐
16 LearnAlign: Reasoning Data Selection for Reinforcement Learning in Large Language Models Based on Improved Gradient Alignment 提出LearnAlign以解决大语言模型强化学习中的数据选择问题 reinforcement learning large language model
17 Visual Pre-Training on Unlabeled Images using Reinforcement Learning 提出基于强化学习的无标签图像预训练方法以提升特征学习 reinforcement learning visual pre-training
18 Automated Treatment Planning for Interstitial HDR Brachytherapy for Locally Advanced Cervical Cancer using Deep Reinforcement Learning 提出基于深度强化学习的自动化HDR近距离放疗计划框架以解决宫颈癌治疗问题 reinforcement learning deep reinforcement learning
19 Growing with Experience: Growing Neural Networks in Deep Reinforcement Learning 提出GrowNN以解决深度强化学习中网络训练困难问题 reinforcement learning deep reinforcement learning
20 Task-Driven Discrete Representation Learning 提出任务驱动的离散表示学习框架以提升下游任务性能 DRL representation learning VQ-VAE
21 Brewing Knowledge in Context: Distillation Perspectives on In-Context Learning 提出知识蒸馏视角以理解上下文学习机制 distillation large language model
22 Understanding Input Selectivity in Mamba: Impact on Approximation Power, Memorization, and Associative Recall Capacity 揭示Mamba中的输入选择性对近似能力和记忆的影响 Mamba SSM
23 From Emergence to Control: Probing and Modulating Self-Reflection in Language Models 提出反思诱导探测方法以增强语言模型自我反思能力 reinforcement learning large language model
24 Interpretable representation learning of quantum data enabled by probabilistic variational autoencoders 提出基于变分自编码器的量子数据可解释表示学习方法 representation learning
25 TreeRL: LLM Reinforcement Learning with On-Policy Tree Search 提出TreeRL框架以解决传统RL方法的探索不足问题 reinforcement learning ✅
26 Attention-based Adversarial Robust Distillation in Radio Signal Classifications for Low-Power IoT Devices 提出基于注意力的对抗鲁棒蒸馏方法以解决低功耗IoT设备中的信号分类问题 distillation
27 ReVeal: Self-Evolving Code Agents via Reliable Self-Verification 提出ReVeal以解决自我验证不可靠的问题 reinforcement learning large language model
28 An Explainable AI Framework for Dynamic Resource Management in Vehicular Network Slicing 提出可解释的深度强化学习框架以解决车载网络切片中的动态资源管理问题 reinforcement learning deep reinforcement learning

🔬 支柱一：机器人控制 (Robot Control) (2 篇)

#题目一句话要点标签🔗⭐
29 TrustGLM: Evaluating the Robustness of GraphLLMs Against Prompt, Text, and Structure Attacks 提出TrustGLM以评估GraphLLMs对对抗性攻击的鲁棒性 manipulation large language model
30 Bias Amplification in RAG: Poisoning Knowledge Retrieval to Steer LLMs 提出BRRA框架以解决RAG系统中的偏见放大问题 manipulation large language model

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

#题目一句话要点标签🔗⭐
31 Delayformer: spatiotemporal transformation for predicting high-dimensional dynamics 提出Delayformer以解决高维动态预测问题 spatiotemporal

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

#题目一句话要点标签🔗⭐
32 SecONNds: Secure Outsourced Neural Network Inference on ImageNet 提出SecONNds以解决安全外包神经网络推理隐私问题 OMOMO ✅

⬅️ 返回 cs.LG 首页 · 🏠 返回主页