cs.CL（2026-03-23）

📊 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	SemEval-2026 Task 12: Abductive Event Reasoning: Towards Real-World Event Causal Inference for Large Language Models	SemEval-2026任务12：提出一个基于证据的归纳事件推理基准，用于评估大语言模型在现实世界事件因果推断中的能力。	large language model	✅
2	DATASHI: A Parallel English-Tashlhiyt Corpus for Orthography Normalization and Low-Resource Language Processing	构建平行英-塔什利特语料库DATASHI，用于正字法标准化和低资源语言处理。	large language model multimodal
3	TaigiSpeech: A Low-Resource Real-World Speech Intent Dataset and Preliminary Results with Scalable Data Mining In-the-Wild	提出TaigiSpeech：一个低资源闽南语真实语音意图数据集，并探索可扩展的数据挖掘方法。	multimodal
4	Probing How Scalable Table Data Enhances General Long-Context Reasoning	TableLong：利用可扩展表格数据增强大语言模型的长文本推理能力	large language model
5	Optimizing Multi-Agent Weather Captioning via Text Gradient Descent: A Training-Free Approach with Consensus-Aware Gradient Fusion	提出WeatherTGD，一种基于文本梯度下降的免训练多智能体天气描述优化框架	large language model
6	Generalizable Self-Evolving Memory for Automatic Prompt Optimization	提出MemAPO，通过自进化记忆实现大语言模型自动提示优化，提升泛化能力。	large language model

#	题目	一句话要点	标签	🔗
7	Dual-Space Knowledge Distillation with Key-Query Matching for Large Language Models with Vocabulary Mismatch	提出DSKD-CMA-GA，通过生成对抗学习解决LLM蒸馏中词表不匹配问题。	distillation large language model
8	DRTriton: Large-Scale Synthetic Data Reinforcement Learning for Triton Kernel Generation	DRTriton：利用大规模合成数据强化学习生成Triton内核，显著提升CUDA内核效率。	reinforcement learning large language model
9	TAMTRL: Teacher-Aligned Reward Reshaping for Multi-Turn Reinforcement Learning in Long-Context Compression	提出TAMTRL，通过教师对齐奖励重塑解决长文本压缩中的多轮强化学习问题。	reinforcement learning large language model
10	Gumbel Distillation for Parallel Text Generation	提出Gumbel蒸馏，提升并行文本生成模型质量，缩小与自回归模型的差距	distillation	✅