cs.AI（2026-05-29）

📊 共 26 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) (16 🔗1) 支柱二：RL算法与架构 (RL & Architecture) (8) 支柱七：动作重定向 (Motion Retargeting) (1 🔗1) 支柱一：机器人控制 (Robot Control) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (16 篇)

#题目一句话要点标签🔗⭐
1 AMix-2: Establishing Protein as a Native Modality in Large Language Models AMix-2：构建蛋白质原生模态的大语言模型，统一蛋白质理解与设计 large language model foundation model
2 Differentially Private Preference Data Synthesis for Large Language Model Alignment 提出DPPrefSyn算法，用于生成差分隐私偏好数据，以对齐大语言模型。 large language model ✅
3 FAM-Bench: A Multimodal Benchmark for Condition-Aware Food-as-Medicine Reasoning 提出FAM-Bench多模态基准，用于评估模型在特定健康状况下的膳食推荐能力 multimodal
4 ImmersiveTTS: Environment-Aware Text-to-Speech with Multimodal Diffusion Transformer and Domain-Specific Representation Alignment ImmersiveTTS：提出环境感知的TTS模型，通过多模态扩散Transformer实现沉浸式语音生成。 multimodal
5 BilliardPhys-Bench: Benchmarking Physical Reasoning and Visual Dynamics of Multimodal LLMs BilliardPhys-Bench：多模态LLM物理推理与视觉动力学评测基准 multimodal
6 Learning to Adapt: Self-Improving Web Agent via Cognitive-Aware Exploration 提出SCALE框架，通过认知探索提升Web Agent在动态环境中的自适应能力 large language model multimodal
7 LinTree: Improving LLM Reasoning with Explicitly Structured Search Histories LinTree：通过显式结构化搜索历史提升LLM推理能力 large language model
8 Neither Replacement nor Panacea: Comparing LLM-Based Conversational and Graphical Decision Support in Industrial Tasks 对比LLM对话式与图形化决策支持在工业任务中的应用，发现其各有优劣。 large language model
9 LLM-FACETS: A Privacy-Preserving Framework for Evaluating LLM Transparency and Accountability LLM-FACETS：一个保护隐私的LLM透明性和问责性评估框架 large language model
10 Developing a UXR Point of View for Cognitive Accessibility in Mobile Learning with Generative AI 利用生成式AI，为移动学习中认知可访问性开发UXR视角 large language model
11 SpecDB: LLM-Generated Customized Databases via Feature-Oriented Decomposition SpecDB：利用LLM和面向特征分解生成定制化数据库 large language model
12 GraphARC: A Comprehensive Benchmark for Graph-Based Abstract Reasoning 提出GraphARC：一个用于图结构抽象推理的综合基准测试。 foundation model
13 TUX: Measuring Human--AI Tacit Understanding 提出TUX指标，衡量人与AI在无明确目标下的隐性理解能力 large language model
14 A Unified and Reproducible Experimentation Framework for Speech Understanding SURE：统一且可复现的语音理解实验框架，提升模型选型效率。 foundation model
15 UniScale: Adaptive Unified Inference Scaling via Online Joint Optimization of Model Routing and Test-Time Scaling 提出UniScale，通过在线联合优化模型路由和测试时缩放，自适应地统一推理加速。 large language model
16 MAVEN: Improving Generalization in Agentic Tool Calling MAVEN：通过模块化验证执行网络提升Agentic工具调用中的泛化能力 large language model

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

#题目一句话要点标签🔗⭐
17 Dreaming Of Others: Latent Teammate Modeling In World Models For Multi-Agent Reinforcement Learning 提出基于世界模型的潜在队友建模方法，解决多智能体强化学习中的协作问题。 reinforcement learning world model world models
18 SLAT: Segment-Level Adaptive Trimming for Efficient CoT Reasoning 提出SLAT：一种分段自适应修剪方法，用于提升CoT推理效率 reinforcement learning large language model chain-of-thought
19 GaMi: Geometry-Agnostic Material Identification via Cross-Modal Subtractive Disentanglement 提出GaMi以解决几何变化引起的材料识别挑战 contrastive learning geometric consistency multimodal
20 COLLEAGUE.SKILL: Automated AI Skill Generation via Expert Knowledge Distillation 提出COLLEAGUE.SKILL以解决AI技能生成的挑战 distillation
21 Answer-Set-Programming-based Abstractions for Reinforcement Learning 提出基于ASP的抽象方法，提升强化学习在复杂状态空间下的泛化能力 reinforcement learning
22 Planner-Centric Reinforcement Learning for Deep Research with Structure-Aware Reward DecomposeR：提出面向规划的强化学习框架，提升LLM在深度研究任务中的表现。 reinforcement learning
23 MIMO: Multilingual Information Retrieval via Monolingual Objectives MIMO：通过单语目标实现多语信息检索，提升跨语言对齐与检索性能。 contrastive learning distillation
24 Distilling LLM Feedback for Lean Theorem Proving 提出Feedback Distillation，利用LLM反馈提升Lean定理证明性能 reinforcement learning distillation

🔬 支柱七：动作重定向 (Motion Retargeting) (1 篇)

#题目一句话要点标签🔗⭐
25 Vector Linking via Cross-Model Local Isometric Consistency 提出基于局部等距一致性的向量链接方法，用于跨模型对象对应恢复。 geometric consistency ✅

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#题目一句话要点标签🔗⭐
26 Choosing the Lens: Strategic Perspective Activation in Context-Dependent Argumentation 提出上下文相关论证框架，解决论证评估中外部环境影响及策略选择问题。 manipulation

⬅️ 返回 cs.AI 首页 · 🏠 返回主页