cs.AI（2024-07-18）

📊 共 16 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) (8 🔗1) 支柱九：具身大模型 (Embodied Foundation Models) (7 🔗1) 支柱六：视频提取与匹配 (Video Extraction) (1)

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

#题目一句话要点标签🔗⭐
1 Thought-Like-Pro: Enhancing Reasoning of Large Language Models through Self-Driven Prolog-based Chain-of-Thought 提出Thought-Like-Pro框架，通过自驱动Prolog增强LLM的推理能力 imitation learning large language model chain-of-thought
2 ROLeR: Effective Reward Shaping in Offline Reinforcement Learning for Recommender Systems ROLeR：离线强化学习中基于奖励塑造的推荐系统 reinforcement learning offline RL offline reinforcement learning ✅
3 Multiobjective Vehicle Routing Optimization with Time Windows: A Hybrid Approach Using Deep Reinforcement Learning and NSGA-II 提出一种混合方法以解决带时间窗的多目标车辆路径优化问题 reinforcement learning deep reinforcement learning DRL
4 MetaSumPerceiver: Multimodal Multi-Document Evidence Summarization for Fact-Checking 提出MetaSumPerceiver模型，用于多模态多文档证据总结，辅助事实核查。 reinforcement learning multimodal
5 On Causally Disentangled State Representation Learning for Reinforcement Learning based Recommender Systems 提出CIDS方法，用于强化学习推荐系统中因果解耦的状态表示学习。 reinforcement learning representation learning
6 Correcting the Mythos of KL-Regularization: Direct Alignment without Overoptimization via Chi-Squared Preference Optimization 提出χPO算法，通过χ²散度正则化解决离线对齐中的过优化问题 reinforcement learning offline reinforcement learning RLHF
7 LLM-Empowered State Representation for Reinforcement Learning 提出LLM赋能的状态表示方法LESR，提升强化学习样本效率。 reinforcement learning large language model
8 DeepClair: Utilizing Market Forecasts for Effective Portfolio Selection DeepClair：利用市场预测优化投资组合选择，提升投资策略。 reinforcement learning deep reinforcement learning

🔬 支柱九：具身大模型 (Embodied Foundation Models) (7 篇)

#题目一句话要点标签🔗⭐
9 Training Foundation Models as Data Compression: On Information, Model Weights and Copyright Law 将大模型训练视为数据压缩，探讨信息论、模型权重与版权法问题 foundation model
10 Handling Numeric Expressions in Automatic Speech Recognition 提出一种结合数据生成策略的端到端方法，用于自动语音识别中数值表达式的正确格式化。 large language model TAMP
11 Generative AI Augmented Induction-based Formal Verification 利用生成式AI增强基于归纳的硬件形式化验证，提升验证效率 large language model
12 CellularLint: A Systematic Approach to Identify Inconsistent Behavior in Cellular Network Specifications CellularLint：利用自然语言处理技术系统性识别蜂窝网络规范中的不一致性 large language model
13 CoDefeater: Using LLMs To Find Defeaters in Assurance Cases CoDefeater：利用大型语言模型自动发现保障案例中的反驳论证 large language model
14 DISCOVER: A Data-driven Interactive System for Comprehensive Observation, Visualization, and ExploRation of Human Behaviour DISCOVER：一个数据驱动的交互式系统，用于全面观察、可视化和探索人类行为 multimodal
15 MMAU: A Holistic Benchmark of Agent Capabilities Across Diverse Domains 提出MMAU：一个综合性的多领域Agent能力评估基准 large language model ✅

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

#题目一句话要点标签🔗⭐
16 Visuospatial navigation from the bottom-up: without vestibular integration, distance prediction, or maps 提出一种无需前庭整合、距离预测或地图构建的自下而上视觉空间导航方法 egocentric

⬅️ 返回 cs.AI 首页 · 🏠 返回主页