cs.AI（2026-04-10）

📊 共 18 篇论文 | 🔗 3 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (13 🔗3) 支柱二：RL算法与架构 (RL & Architecture) (5)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (13 篇)

#题目一句话要点标签🔗⭐
1 Constraint-Aware Corrective Memory for Language-Based Drug Discovery Agents 提出CACM框架，通过约束感知的修正记忆提升语言驱动的药物发现Agent性能。 large language model multimodal
2 Beyond Relevance: Utility-Centric Retrieval in the LLM Era 面向LLM的效用驱动检索：超越相关性，提升检索增强生成质量 large language model
3 Strategic Algorithmic Monoculture:Experimental Evidence from Coordination Games 研究表明LLM在协调博弈中表现出策略性算法趋同，但维持异质性方面不如人类 large language model
4 E3-TIR: Enhanced Experience Exploitation for Tool-Integrated Reasoning 提出E3-TIR，通过增强经验利用解决工具集成推理中的训练难题。 large language model ✅
5 LLM-Rosetta: A Hub-and-Spoke Intermediate Representation for Cross-Provider LLM API Translation LLM-Rosetta：一种用于跨LLM API提供商翻译的Hub-and-Spoke中间表示 large language model ✅
6 SAGE: A Service Agent Graph-guided Evaluation Benchmark 提出SAGE：一个服务代理图引导的评估基准，用于评估LLM在客服场景中的性能 large language model
7 Structuring versus Problematizing: How LLM-based Agents Scaffold Learning in Diagnostic Reasoning 提出PharmaSim Switch以解决药学技术人员诊断推理问题 large language model
8 DeepGuard: Secure Code Generation via Multi-Layer Semantic Aggregation DeepGuard：通过多层语义聚合实现安全的代码生成。 large language model ✅
9 Watt Counts: Energy-Aware Benchmark for Sustainable LLM Inference on Heterogeneous GPU Architectures Watt Counts：针对异构GPU上可持续LLM推理的能耗感知基准测试 large language model
10 Noise-Aware In-Context Learning for Hallucination Mitigation in ALLMs 提出噪声感知上下文学习方法，缓解听觉大语言模型中的幻觉问题 large language model
11 Enhancing LLM Problem Solving via Tutor-Student Multi-Agent Interaction 提出PETITE框架，通过导师-学生多智能体交互提升LLM代码问题求解能力 large language model
12 AudioGuard: Toward Comprehensive Audio Safety Protection Across Diverse Threat Models AudioGuard：面向多样化威胁模型的全面音频安全防护方案 foundation model
13 Hidden in Plain Sight: Visual-to-Symbolic Analytical Solution Inference from Field Visualizations 提出ViSA-R2，从场可视化中推断物理场解析解，解决AI辅助科学推理难题。 chain-of-thought

🔬 支柱二：RL算法与架构 (RL & Architecture) (5 篇)

#题目一句话要点标签🔗⭐
14 PilotBench: A Benchmark for General Aviation Agents with Safety Constraints PilotBench：面向通用航空代理，带安全约束的基准测试 MAE embodied AI large language model
15 SPPO: Sequence-Level PPO for Long-Horizon Reasoning Tasks 提出SPPO以解决长时间推理任务中的PPO不稳定问题 PPO large language model chain-of-thought
16 Advantage-Guided Diffusion for Model-Based Reinforcement Learning 提出Advantage引导的扩散模型(AGD-MBRL)，提升基于扩散模型的模型强化学习性能。 reinforcement learning PPO world model
17 On the Representational Limits of Quantum-Inspired 1024-D Document Embeddings: An Experimental Evaluation Framework 评估量子启发式1024维文档嵌入的表征能力极限，揭示其在信息检索中的局限性 teacher-student distillation large language model
18 StaRPO: Stability-Augmented Reinforcement Policy Optimization 提出StaRPO，通过增强推理稳定性提升大型语言模型在复杂推理任务中的性能。 reinforcement learning large language model

⬅️ 返回 cs.AI 首页 · 🏠 返回主页