cs.AI（2026-06-18）

📊 共 34 篇论文 | 🔗 1 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (19 🔗1) 支柱二：RL算法与架构 (RL & Architecture) (10) 支柱一：机器人控制 (Robot Control) (3) 支柱三：空间感知与语义 (Perception & Semantics) (2)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (19 篇)

#题目一句话要点标签🔗⭐
1 AI Economist Agent: An Agentic Framework for Model-Grounded Economic Analysis with RAG, Knowledge Graphs, and Large Language Models 提出基于RAG的AI经济学家框架以解决经济分析问题 large language model
2 QMFOL: Benchmarking Large Language Model Reasoning via Quantifiable Monadic First-Order Logic Test Case Generation 提出QMFOL以解决现有推理基准评估不足的问题 large language model
3 Apparent Psychological Profiles of Large Language Models are Largely a Measurement Artifact 揭示大型语言模型心理特征的测量伪影问题 large language model
4 BIM-Edit: Benchmarking Large Language Models for IFC-Based Building Information Modeling 提出BIM-Edit基准以评估LLM在建筑信息建模中的编辑能力 large language model
5 A Systematic Evaluation of Black-Box Uncertainty Estimation Methods for Large Language Models 系统评估黑箱不确定性估计方法以提升大型语言模型的可靠性 large language model
6 CombEval: A Framework for Evaluating Combinatorial Counting in Large Language Models 提出CombEval框架以评估大语言模型的组合计数能力 large language model ✅
7 Library-Aware Doubles and Iterative Repair for Large Language Model-Generated Unit Tests in OpenSIL Firmware 提出基于大语言模型的自动化单元测试生成与修复方法以解决固件测试问题 large language model
8 Human-on-the-Loop Orchestration for AI-Assisted Legal Discovery 提出四层验证架构以解决法律信息检索中的代理失败问题 large language model
9 AURA: Adaptive Uncertainty-aware Refinement for LLM-as-a-Judge Auditing 提出AURA框架以解决LLM审计中的不确定性问题 large language model
10 Multi-LCB: Extending LiveCodeBench to Multiple Programming Languages 提出Multi-LCB以解决多编程语言评估问题 large language model
11 Multi-View Decompilation for LLM-Based Malware Classification 提出多视图反编译方法以提升恶意软件分类精度 large language model
12 LLM agent safety, multi-turn red-teaming, jailbreak benchmarks, adversarial robustness, safety-critical systems 提出NRT-Bench以评估LLM代理在安全关键系统中的鲁棒性 large language model
13 AutoPass: Evidence-Guided LLM Agents for Compiler Performance Tuning 提出AutoPass以解决编译器性能调优问题 large language model
14 Navigating Unreliable Parametric and Contextual Knowledge: Explicit Knowledge Conflict Resolution for LLM Inference 提出MACR框架以解决LLM推理中的知识冲突问题 large language model
15 Thermodynamic Measure of Intelligence 提出热力学智能测量方法以量化智能系统表现 large language model
16 Advancing DialNav through Automatic Embodied Dialog Augmentation 通过自动化对话增强推进DialNav以解决训练数据不足问题 VLN
17 ORAgentBench: Can LLM Agents Solve Challenging Operations Research Tasks End to End? 提出ORAgentBench以评估自主代理在运筹学任务中的表现 large language model
18 Towards Engineering Scaling Laws with Pretraining Data Composition 通过预训练数据组合工程化缩放规律以提升粒子物理模型性能 large language model
19 SafeSpec: Fast and Safe LLM via Dynamic Reflective Sampling 提出SafeSpec以解决大语言模型推理安全性问题 large language model

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

#题目一句话要点标签🔗⭐
20 ELVA: Exploring Ranking-Driven Universal Multimodal Retrieval 提出ELVA以解决多模态检索中的粒度盲点问题 reinforcement learning contrastive learning large language model
21 Evaluation of EEG Foundation Models for Event-Based Burst-Suppression Detection in ICU 评估EEG基础模型以解决ICU中事件驱动的突发抑制检测问题 Mamba foundation model
22 Augmenting Game AI with Deep Reinforcement Learning 提出深度强化学习框架以增强游戏AI的真实性 reinforcement learning deep reinforcement learning
23 Reward as An Agent for Embodied World Models 提出奖励作为代理以解决探索不足问题 world model world models
24 Multi-Head Attention-Based Feature Extractor Integration with Soft Actor-Critic for Porosity Prediction and Process Parameter Optimization in Additive Manufacturing 提出基于多头注意力机制的SAC方法以优化增材制造中的孔隙率预测 reinforcement learning PPO SAC
25 Beyond Accuracy: Measuring Logical Compliance of Predictive Models 提出规则违规评分（RVS）以解决模型逻辑一致性问题 predictive model
26 Hybrid Diffusion Transformer for Instruction-Guided Audio Editing via Rectified Flow 提出混合扩散变换器以解决指令引导的音频编辑问题 flow matching multimodal
27 Process-Verified Reinforcement Learning for Theorem Proving via Lean 提出基于Lean的过程验证强化学习以提升定理证明效率 reinforcement learning
28 MetaResearcher: Scaling Deep Research via Self-Reflective Reinforcement Learning in Adversarial Virtual Environments 提出MetaResearcher以解决深度研究代理训练的局限性 reinforcement learning
29 Beyond Entropy: Learning from Token-Level Distributional Deviations for LLM Reasoning 提出独立组合标记框架以解决LLM推理中的优化不稳定问题 reinforcement learning large language model

🔬 支柱一：机器人控制 (Robot Control) (3 篇)

#题目一句话要点标签🔗⭐
30 PhysDrift: Bridging the Embodiment Gap in Humanoid Co-Speech Motion Generation 提出PhysDrift以解决人形机器人共语动作生成中的体现差距问题 humanoid humanoid robot motion generation
31 ENPIRE: Agentic Robot Policy Self-Improvement in the Real World 提出ENPIRE框架以实现机器人自主政策自我改进 manipulation dexterous manipulation
32 Learning to Prompt: Improving Student Engagement with Adaptive LLM-based High-School Tutoring 提出基于适应性提示的LLM高中新型辅导系统以提升学生参与度 sim-to-real

🔬 支柱三：空间感知与语义 (Perception & Semantics) (2 篇)

#题目一句话要点标签🔗⭐
33 Lagrange: An Open-Vocabulary, Energy-Based Sparse Framework for Generalized End-to-End Driving 提出Lagrange框架以解决开放世界驾驶中的感知与规划问题 open-vocabulary open vocabulary vision-language-action
34 Editorial Alignment: A Participatory Approach to Engaging Editorial Expertise in LLM-mediated Knowledge Dissemination 提出编辑对齐方法以增强LLM知识传播中的编辑参与 affordance

⬅️ 返回 cs.AI 首页 · 🏠 返回主页