cs.CL（2026-02-14）

📊 共 10 篇论文 | 🔗 2 篇有代码

🎯 兴趣领域导航

#	题目	一句话要点	标签	🔗
1	On Calibration of Large Language Models: From Response To Capability	提出能力校准方法，解决大语言模型在通用问题求解中置信度估计不准确的问题	large language model
2	Tutoring Large Language Models to be Domain-adaptive, Precise, and Safe	通过领域自适应、精确性和安全性指导大型语言模型	large language model
3	RMPL: Relation-aware Multi-task Progressive Learning with Stage-wise Training for Multimedia Event Extraction	提出RMPL框架，通过关系感知多任务渐进学习解决多媒体事件抽取中的低资源问题。	multimodal
4	LiveNewsBench: Evaluating LLM Web Search Capabilities with Freshly Curated News	LiveNewsBench：利用新鲜新闻评估LLM的Web搜索能力	large language model
5	ADAB: Arabic Dataset for Automated Politeness Benchmarking -- A Large-Scale Resource for Computational Sociopragmatics	提出ADAB：一个大规模阿拉伯语礼貌基准数据集，用于计算社会语用学研究。	large language model
6	Bridging the Multilingual Safety Divide: Efficient, Culturally-Aware Alignment for Global South Languages	针对全球南方语言，提出高效且具文化敏感性的对齐方法，弥合多语言安全差距。	large language model
7	PrivAct: Internalizing Contextual Privacy Preservation via Multi-Agent Preference Training	PrivAct：通过多智能体偏好训练实现上下文隐私保护内化	large language model	✅

#	题目	一句话要点	标签	🔗
8	DistillLens: Symmetric Knowledge Distillation Through Logit Lens	DistillLens：通过Logit Lens实现对称知识蒸馏，提升LLM指令跟随能力	distillation large language model instruction following	✅
9	Elo-Evolve: A Co-evolutionary Framework for Language Model Alignment	提出Elo-Evolve框架以解决大语言模型对齐问题	curriculum learning large language model
10	Beyond Words: Evaluating and Bridging Epistemic Divergence in User-Agent Interaction via Theory of Mind	提出基准测试与强化学习方法，提升LLM在用户交互中理解和解决认知差异的能力	reinforcement learning large language model