cs.AI（2026-02-15）

📊 共 18 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) (15 🔗2) 支柱二：RL算法与架构 (RL & Architecture) (3)

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

#题目一句话要点标签🔗⭐
1 HyMem: Hybrid Memory Architecture with Dynamic Retrieval Scheduling HyMem：一种基于动态检索调度的混合记忆架构，提升LLM Agent长时记忆效率。 large language model
2 FMMD: A multimodal open peer review dataset based on F1000Research FMMD：一个基于F1000Research的多模态开放同行评审数据集 multimodal
3 Anticipating Adversary Behavior in DevSecOps Scenarios through Large Language Models 利用大语言模型预测DevSecOps场景中的对抗行为，提升云安全 large language model ✅
4 TabTracer: Monte Carlo Tree Search for Complex Table Reasoning with Large Language Models TabTracer：基于蒙特卡洛树搜索的LLM复杂表格推理框架 large language model
5 Beyond Static Snapshots: Dynamic Modeling and Forecasting of Group-Level Value Evolution with Large Language Models 提出基于大语言模型的动态建模框架，预测群体价值观随时间演变 large language model
6 Toward Autonomous O-RAN: A Multi-Scale Agentic AI Framework for Real-Time Network Control and Management 提出多尺度Agentic AI框架，实现O-RAN实时网络控制与管理 large language model foundation model
7 NEST: Nascent Encoded Steganographic Thoughts NEST：探索大型语言模型中隐写术思维链的风险与防范 large language model chain-of-thought
8 Benchmarking at the Edge of Comprehension 提出抗批判基准测试框架，解决大模型超越人类理解能力后的评测难题 large language model
9 Algebraic Quantum Intelligence: A New Framework for Reproducible Machine Creativity 提出代数量子智能框架，通过非交换代数扩展语义空间，提升机器创造力。 large language model
10 GUI-GENESIS: Automated Synthesis of Efficient Environments with Verifiable Rewards for GUI Agent Post-Training GUI-GENESIS：自动合成高效且具有可验证奖励的GUI Agent后训练环境 multimodal
11 Choosing How to Remember: Adaptive Memory Structures for LLM Agents 提出FluxMem以解决LLM代理记忆结构选择问题 large language model
12 Cognitive Chunking for Soft Prompts: Accelerating Compressor Learning via Block-wise Causal Masking 提出并行迭代压缩PIC，通过分块因果掩码加速软提示压缩器学习。 large language model
13 A Rational Analysis of the Effects of Sycophantic AI 揭示奉承型AI对认知的影响：强化现有信念，阻碍发现真理 large language model
14 ForesightSafety Bench: A Frontier Risk Evaluation and Governance Framework towards Safe AI 提出ForesightSafety Bench，用于全面评估前沿AI的潜在风险与安全治理。 embodied AI ✅
15 Plan-MCTS: Plan Exploration for Action Exploitation in Web Navigation Plan-MCTS：通过规划空间探索提升Web导航中的动作利用 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (3 篇)

#题目一句话要点标签🔗⭐
16 GRAIL: Goal Recognition Alignment through Imitation Learning GRAIL：通过模仿学习实现目标识别对齐，解决次优行为下的目标推断问题 reinforcement learning imitation learning inverse reinforcement learning
17 REDSearcher: A Scalable and Cost-Efficient Framework for Long-Horizon Search Agents REDSearcher：一种可扩展且经济高效的长程搜索代理框架 reinforcement learning large language model multimodal
18 Eureka-Audio: Triggering Audio Intelligence in Compact Language Models Eureka-Audio：在紧凑型语言模型中激发音频智能 Eureka instruction following

⬅️ 返回 cs.AI 首页 · 🏠 返回主页