cs.AI（2025-08-04）

📊 共 7 篇论文 | 🔗 1 篇有代码

🎯 兴趣领域导航

#	题目	一句话要点	标签	🔗
1	Defend LLMs Through Self-Consciousness	提出自我意识防御机制以应对大语言模型的提示注入攻击	large language model
2	Noosemia: toward a Cognitive and Phenomenological Account of Intentionality Attribution in Human-Generative AI Interaction	提出Noosemia框架以解释人机交互中的意向性归属	multimodal
3	Test-time Prompt Intervention	提出测试时提示干预框架以解决推理冗余问题	large language model
4	Dialogue Systems Engineering: A Survey and Future Directions	提出对话系统工程以提升对话系统的开发与应用效率	large language model
5	Trainable Dynamic Mask Sparse Attention	提出可训练动态掩码稀疏注意力机制以解决长上下文建模问题	large language model	✅

#	题目	一句话要点	标签	🔗	⭐
6	Communicating Plans, Not Percepts: Scalable Multi-Agent Coordination with Embodied World Models	提出基于世界模型的通信策略以解决多智能体协调问题	reinforcement learning world model predictive model
7	SecoustiCodec: Cross-Modal Aligned Streaming Single-Codecbook Speech Codec	提出SecoustiCodec以解决语音编码中的语义与副语言信息分离问题	contrastive learning multimodal