cs.AI（2026-09-02）

📊 共 23 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) (19 🔗2) 支柱二：RL算法与架构 (RL & Architecture) (3 🔗1) 支柱八：物理动画 (Physics-based Animation) (1)

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

#	题目	一句话要点	标签	🔗	⭐
1	Transfer Safety Awareness for Cross-Modal Safety Drift in Multimodal Large Language Models	提出安全意识表示转移方法以解决多模态安全漂移问题	large language model multimodal	✅
2	Large Language Models (LLMs) for Telecom Root Cause Analysis (RCA): A Structured Reasoning Framework for Evidence-Grounded Diagnosis	提出结构化推理框架以提升电信根因分析的准确性	large language model
3	Door-in-the-Face Requests and Refusal Behaviour in Large Language Models	探讨语言模型中的门槛请求与拒绝行为	large language model
4	Automated Vulnerability Injection in Smart Contracts Using Large Language Models	利用大型语言模型自动注入智能合约漏洞	large language model
5	ASCII Attack: Recontextualising Harmful Requests as Artistic Critique in Large Language Models	提出ASCII Attack以解决大型语言模型对有害请求的脆弱性问题	large language model
6	Beyond Modality Harmony: Orthogonal Purification and Topology-Guided MoE for Conflict-Aware Multimodal Recommendation	提出OrthoRec以解决多模态推荐中的冲突问题	multimodal
7	DocHop: Benchmarking Out-of-domain Multi-hop Reasoning in Information-Dense Documents	提出DocHop以解决信息密集型文档中的多跳推理问题	large language model multimodal
8	Competitive Market Behavior of LLMs	探讨大型语言模型在市场机制中的竞争行为	large language model chain-of-thought
9	CoMerge: Conflict-Driven Preference Optimization for Multi-Task Model Merging	提出CoMerge以解决多任务模型合并中的参数干扰问题	large language model instruction following
10	OmegaUse-SOP: SOP Engineering for Professional Computer Use from Human Demonstrations	提出OmegaUse-SOP以解决专业计算机使用中的SOP工程问题	large language model multimodal
11	Loom: Weaving Diagnostic Strands into Free-Text Consensus via Embedding-Space Reweighting	提出Loom框架以解决工业环境中的文本假设聚合问题	large language model
12	UTP-Bench: Uncertainty-aware Travel Planning Benchmark	提出UTP-Bench以解决不确定性旅行规划问题	large language model
13	SCX Router: Streaming Zero-Shot Model Selection with a Decoder-KV Classifier and a Real-World Task Ontology	提出SCX Router以优化大语言模型的任务选择问题	large language model
14	Task-Level Natural Language Priors as Learning Signals for Low-Resource LLM Training	提出Prior-Guided Tuning以解决低资源LLM训练中的信号不足问题	large language model
15	PEARL: Path-Entity Aligned Relational Learning with Contextual Subgraphs for Inductive Knowledge Graph Completion	提出PEARL框架以解决归纳知识图谱补全问题	large language model
16	Semantic Signal-Assisted Inspection and Recovery Allocation in Reverse Logistics	提出语义信号辅助决策支持以优化逆向物流中的检查与回收分配	large language model
17	Git4Data: Database-Native Version Control for AI Agents	提出Git4Data以解决AI代理版本控制问题	large language model
18	Benchmarking Language Models for Statistical Problem Formulation	提出StatFormBench以解决统计问题表述的挑战	large language model	✅
19	Knowing Is Not Enough: Information Retrievability as a Precondition to Effective LLM Oversight	提出信息可检索性以增强大型语言模型的人工监督效果	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	🔗	⭐
20	PhoenixNest-Video: Evidence-Grounded Multimodal Agent Framework for Automated Video Interview Assessment	提出PhoenixNest-Video框架以解决视频面试评估的自动化问题	reinforcement learning multimodal
21	Discriminative World Models for Web Agents	提出预测状态匹配以提升网络代理的世界模型性能	world model world models	✅
22	APEx: Distillation of Agent Procedural Experience for Adaptive Deep Research Question Answering	提出APEx框架以解决深度研究代理的经验利用问题	reinforcement learning distillation large language model

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

#	题目	一句话要点	标签	🔗	⭐
23	Post-Training Ternarization of Qwen3-4B Capability, Effective Bit Budget, Storage Compression, and Deployment	提出后训练三值化方法以优化Qwen3-4B模型的存储与部署	AMP

⬅️ 返回 cs.AI 首页 · 🏠 返回主页