cs.AI（2025-03-24）

📊 共 9 篇论文 | 🔗 1 篇有代码

🎯 兴趣领域导航

#	题目	一句话要点	标签	🔗
1	Fundamental Safety-Capability Trade-offs in Fine-tuning Large Language Models	理论分析揭示LLM微调中安全性和能力之间的根本性权衡	large language model
2	Rankers, Judges, and Assistants: Towards Understanding the Interplay of LLMs in Information Retrieval Evaluation	揭示LLM在信息检索评估中的相互影响：排序器、评判器与助手	large language model
3	REALM: A Dataset of Real-World LLM Use Cases	构建REALM数据集，揭示LLM在现实世界的应用场景与用户画像	large language model
4	BitDecoding: Unlocking Tensor Cores for Long-Context LLMs with Low-Bit KV Cache	BitDecoding：利用Tensor Core加速低比特KV缓存长文本LLM推理	large language model	✅
5	Verbal Process Supervision Elicits Better Coding Agents	CURA：通过口头过程监督提升代码生成Agent性能	large language model
6	VeriSafe Agent: Safeguarding Mobile GUI Agent via Logic-based Action Verification	提出VeriSafe Agent，通过逻辑验证保障移动GUI Agent的可靠性	foundation model
7	Bridging Writing Manner Gap in Visual Instruction Tuning by Creating LLM-aligned Instructions	提出LLM对齐指令，弥合视觉指令调优中的写作风格差距，提升多模态模型性能。	large language model
8	Improving RAG for Personalization with Author Features and Contrastive Examples	提出结合作者特征和对比样本的RAG方法，提升个性化文本生成效果	large language model

#	题目	一句话要点	标签	🔗	⭐
9	AdaWorld: Learning Adaptable World Models with Latent Actions	AdaWorld：通过潜在动作学习可适应的World Model，提升泛化能力	world model