cs.LG（2026-02-21）

📊 共 10 篇论文

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (5) 支柱二：RL算法与架构 (RL & Architecture) (2) 支柱一：机器人控制 (Robot Control) (2) 支柱八：物理动画 (Physics-based Animation) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (5 篇)

#	题目	一句话要点	标签	🔗	⭐
1	TRUE: A Trustworthy Unified Explanation Framework for Large Language Model Reasoning	提出TRUE框架，用于提升大语言模型推理过程的可信性和可解释性	large language model
2	Could Large Language Models work as Post-hoc Explainability Tools in Credit Risk Models?	利用大语言模型提升信贷风险模型事后可解释性	large language model
3	Prior Aware Memorization: An Efficient Metric for Distinguishing Memorization from Generalization in Large Language Models	提出Prior-Aware Memorization，高效区分大语言模型中的记忆与泛化	large language model
4	Transformers for dynamical systems learn transfer operators in-context	Transformer通过上下文学习动力系统传递算子，实现零样本预测。	foundation model zero-shot transfer
5	Limits of Convergence-Rate Control for Open-Weight Safety	提出SpecDef算法，通过谱重参数化实现开放权重模型的收敛速度控制，防御恶意微调。	foundation model

🔬 支柱二：RL算法与架构 (RL & Architecture) (2 篇)

#	题目	一句话要点	标签	🔗	⭐
6	VariBASed: Variational Bayes-Adaptive Sequential Monte-Carlo Planning for Deep Reinforcement Learning	提出VariBASeD，融合变分贝叶斯和序列蒙特卡洛规划，提升深度强化学习的探索效率。	reinforcement learning deep reinforcement learning
7	DeepInterestGR: Mining Deep Multi-Interest Using Multi-Modal LLMs for Generative Recommendation	DeepInterestGR：利用多模态LLM挖掘深度多兴趣，用于生成式推荐	reinforcement learning chain-of-thought

🔬 支柱一：机器人控制 (Robot Control) (2 篇)

#	题目	一句话要点	标签	🔗	⭐
8	In-Context Planning with Latent Temporal Abstractions	提出I-TAP，结合上下文适应与在线规划，解决连续控制中随机动态和部分可观测问题。	manipulation reinforcement learning offline RL
9	Issues with Measuring Task Complexity via Random Policies in Robotic Tasks	评估基于随机策略的任务复杂度度量方法在机器人任务中的有效性	manipulation reinforcement learning

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

#	题目	一句话要点	标签	🔗	⭐
10	Incremental Transformer Neural Processes	提出增量Transformer神经过程(incTNP)，加速序列数据建模并保持性能。	spatiotemporal large language model

⬅️ 返回 cs.LG 首页 · 🏠 返回主页