cs.AI（2026-04-14）

📊 共 33 篇论文 | 🔗 4 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (23 🔗3) 支柱二：RL算法与架构 (RL & Architecture) (6 🔗1) 支柱一：机器人控制 (Robot Control) (2) 支柱五：交互与反应 (Interaction & Reaction) (1) 支柱八：物理动画 (Physics-based Animation) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (23 篇)

#题目一句话要点标签🔗⭐
1 MISID: A Multimodal Multi-turn Dataset for Complex Intent Recognition in Strategic Deception Games MISID：用于策略欺骗游戏中复杂意图识别的多模态多轮数据集 large language model multimodal
2 Heuristic Classification of Thoughts Prompting (HCoT): Integrating Expert System Heuristics for Structured Reasoning into Large Language Models 提出启发式思维分类提示（HCoT），将专家系统启发式推理融入大语言模型。 large language model chain-of-thought
3 MultiDocFusion: Hierarchical and Multimodal Chunking Pipeline for Enhanced RAG on Long Industrial Documents 提出MultiDocFusion以解决长工业文档处理中的信息损失问题 large language model multimodal
4 Beyond Output Correctness: Benchmarking and Evaluating Large Language Model Reasoning in Coding Tasks CodeRQ-Bench：用于评估LLM在代码任务中推理质量的基准测试与VERA评估器 large language model ✅
5 Preventing Safety Drift in Large Language Models via Coupled Weight and Activation Constraints 提出耦合权重与激活约束(CWAC)方法，防止大语言模型微调过程中的安全性漂移 large language model
6 A Scoping Review of Large Language Model-Based Pedagogical Agents 综述基于大型语言模型的教学代理以推动教育创新 large language model
7 Modality-Native Routing in Agent-to-Agent Networks: A Multimodal A2A Protocol Extension 提出MMA2A，通过模态原生路由提升多智能体系统跨模态推理任务准确率。 multimodal
8 Modeling Co-Pilots for Text-to-Model Translation 提出Text2Model和Text2Zinc，用于文本到组合优化模型的自动翻译。 large language model chain-of-thought
9 RPRA: Predicting an LLM-Judge for Efficient but Performant Inference 提出RPRA框架，提升小模型推理效率，通过预测LLM判决结果实现自适应推理。 large language model
10 PAL: Personal Adaptive Learner PAL：提出一种个性化自适应学习平台，通过实时互动提升学习体验 multimodal
11 LogicEval: A Systematic Framework for Evaluating Automated Repair Techniques for Logical Vulnerabilities in Real-World Software LogicEval：系统性评估真实软件中逻辑漏洞的自动修复技术 large language model
12 CoDe-R: Refining Decompiler Output with LLMs via Rationale Guidance and Adaptive Inference CoDe-R：通过LLM、理由引导和自适应推理改进反编译器输出，显著提升代码可执行性。 large language model ✅
13 BEAM: Bi-level Memory-adaptive Algorithmic Evolution for LLM-Powered Heuristic Design 提出BEAM以解决现有LHH在启发式设计中的局限性 large language model
14 AISafetyBenchExplorer: A Metric-Aware Catalogue of AI Safety Benchmarks Reveals Fragmented Measurement and Weak Benchmark Governance AISafetyBenchExplorer：构建AI安全基准评测体系，揭示碎片化测量和薄弱的基准治理问题 large language model
15 DeepTest Tool Competition 2026: Benchmarking an LLM-Based Automotive Assistant DeepTest 2026：LLM汽车助手评测竞赛，评估故障检测工具 large language model
16 IDEA: An Interpretable and Editable Decision-Making Framework for LLMs via Verbal-to-Numeric Calibration IDEA框架：通过Verbal-to-Numeric校准，实现LLM决策过程的可解释与可编辑性 large language model ✅
17 A Two-Stage LLM Framework for Accessible and Verified XAI Explanations 提出双阶段LLM框架，提升可解释AI解释的可访问性和可靠性 large language model
18 Operationalising the Right to be Forgotten in LLMs: A Lightweight Sequential Unlearning Framework for Privacy-Aligned Deployment in Politically Sensitive Environments 提出轻量级序列化遗忘框架，用于在政治敏感环境中部署符合隐私法规的大语言模型。 large language model
19 Is Vibe Coding the Future? An Empirical Assessment of LLM Generated Codes for Construction Safety 评估LLM生成代码在建筑安全中的可靠性，揭示“氛围编程”的潜在风险 large language model
20 GAM: Hierarchical Graph-based Agentic Memory for LLM Agents 提出GAM：一种层级图结构的Agent记忆框架，解决LLM Agent长期交互中的知识保留与适应性问题。 large language model
21 Designing Reliable LLM-Assisted Rubric Scoring for Constructed Responses: Evidence from Physics Exams 设计可靠的LLM辅助物理考试评分系统，提升评分一致性与效率 large language model
22 Beyond Scores: Diagnostic LLM Evaluation via Fine-Grained Abilities 提出认知诊断框架以解决大语言模型评估的细粒度能力问题 large language model
23 EMBER: Autonomous Cognitive Behaviour from Learned Spiking Neural Network Dynamics in a Hybrid LLM Architecture EMBER：混合LLM架构中基于学习的脉冲神经网络动态的自主认知行为 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (6 篇)

#题目一句话要点标签🔗⭐
24 DocSeeker: Structured Visual Reasoning with Evidence Grounding for Long Document Understanding DocSeeker：提出一种基于证据 grounding 的结构化视觉推理方法，用于长文档理解。 distillation large language model multimodal
25 KnowRL: Boosting LLM Reasoning via Reinforcement Learning with Minimal-Sufficient Knowledge Guidance KnowRL：通过最小充分知识引导的强化学习提升LLM推理能力 reinforcement learning large language model ✅
26 A hierarchical spatial-aware algorithm with efficient reinforcement learning for human-robot task planning and allocation in production 提出一种层级空间感知算法，结合高效强化学习，解决生产中人机任务规划与分配问题。 reinforcement learning
27 Safe reinforcement learning with online filtering for fatigue-predictive human-robot task planning and allocation in production 提出PF-CD3Q安全强化学习算法，解决人机协作中疲劳预测的任务规划与分配问题。 reinforcement learning
28 Human-Centric Topic Modeling with Goal-Prompted Contrastive Learning and Optimal Transport 提出Human-TM，通过目标提示对比学习和最优传输实现以人为中心的Topic Modeling contrastive learning
29 HintMR: Eliciting Stronger Mathematical Reasoning in Small Language Models HintMR：通过提示辅助增强小语言模型中的数学推理能力 distillation large language model

🔬 支柱一：机器人控制 (Robot Control) (2 篇)

#题目一句话要点标签🔗⭐
30 RePAIR: Interactive Machine Unlearning through Prompt-Aware Model Repair RePAIR：通过提示感知的模型修复实现交互式机器遗忘 manipulation large language model foundation model
31 Security and Resilience in Autonomous Vehicles: A Proactive Design Approach 提出一种主动设计方法，增强自动驾驶汽车的安全性和韧性 manipulation

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

#题目一句话要点标签🔗⭐
32 Fully Homomorphic Encryption on Llama 3 model for privacy preserving LLM inference 在Llama 3模型上实现全同态加密，保护LLM推理过程中的隐私 OMOMO large language model

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

#题目一句话要点标签🔗⭐
33 FRTSearch: Unified Detection and Parameter Inference of Fast Radio Transients using Instance Segmentation FRTSearch：利用实例分割统一快速射电瞬变的检测与参数推断 PULSE

⬅️ 返回 cs.AI 首页 · 🏠 返回主页