cs.CL（2026-07-07）

📊 共 9 篇论文 | 🔗 2 篇有代码

🎯 兴趣领域导航

#	题目	一句话要点	标签	🔗
1	Data Analysis in the Wild: Benchmarking Large Language Models Against Real-World Data Complexities	提出DataGovBench以解决LLM在真实数据分析中的不足问题	large language model	✅
2	Pluralis v0.1: Towards a Multicultural, Multimodal, Multilingual Benchmark for AI Risk and Reliability	提出Pluralis v0.1以解决AI安全评估中的文化偏见问题	multimodal
3	MemDefrag: Latent Memory Defragmentation for Large Language Models	提出MemDefrag以解决大语言模型的潜在记忆碎片化问题	large language model
4	SpanUQ: Span-Level Uncertainty Quantification for Large Language Model Generation	提出SpanUQ以解决大语言模型生成中的不确定性量化问题	large language model
5	From Voting to Agent Collaboration: Answer-Type-Aware LLM Pipelines for BioASQ 14b	提出针对生物医学问答的LLM框架以提升答案稳健性	large language model chain-of-thought
6	PluraMath: Extending Mathematical Reasoning Evaluation Beyond High-Resource Languages	提出PluraMath以解决数学推理评估的语言偏见问题	large language model instruction following
7	LongCrafter: Towards Diverse Long-Context Understanding via Evidence-Graph-Guided Instruction Synthesis	提出LongCrafter以解决长上下文理解的多样性问题	large language model
8	LLM Agents for Deliberative Collaboration: A Study on Joint Decision Making Under Partial Observability	提出LLM代理以解决部分可观测联合决策问题	large language model

#	题目	一句话要点	标签	🔗	⭐
9	Improving LLM-Generated Process Model Quality Through Reinforcement Learning: The Role of Reward Function Design	通过强化学习优化LLM生成的流程模型质量	reinforcement learning large language model	✅