cs.CL（2024-07-06）

📊 共 10 篇论文 | 🔗 1 篇有代码

🎯 兴趣领域导航

#	题目	一句话要点	标签	🔗
1	Enhance the Robustness of Text-Centric Multimodal Alignments	提出更鲁棒的文本中心多模态对齐方法，提升模型在噪声和模态缺失下的性能。	large language model multimodal
2	MMSci: A Dataset for Graduate-Level Multi-Discipline Multimodal Scientific Understanding	MMSci：一个用于研究生级别多学科多模态科学理解的数据集	multimodal
3	Applicability of Large Language Models and Generative Models for Legal Case Judgement Summarization	探索大型语言模型和生成模型在法律判决摘要中的适用性，并分析其幻觉问题。	large language model
4	AI Safety in Generative AI Large Language Models: A Survey	针对生成式AI大语言模型，综述其AI安全风险与对齐方法，为构建安全模型提供参考。	large language model
5	Large language models are good medical coders, if provided with tools	提出基于检索-排序的两阶段系统，显著提升LLM在ICD-10-CM医学编码中的准确率	large language model
6	LLMCloudHunter: Harnessing LLMs for Automated Extraction of Detection Rules from Cloud-Based CTI	LLMCloudHunter：利用LLM自动从云CTI中提取检测规则	large language model
7	Beyond Perplexity: Multi-dimensional Safety Evaluation of LLM Compression	针对LLM压缩，提出多维度安全评估框架，超越困惑度指标。	large language model	✅

#	题目	一句话要点	标签
8	TRACE: TRansformer-based Attribution using Contrastive Embeddings in LLMs	提出TRACE：一种基于Transformer和对比嵌入的LLM溯源框架	contrastive learning large language model
9	Cross-Lingual Word Alignment for ASEAN Languages with Contrastive Learning	提出基于对比学习的跨语言词对齐方法，提升东盟语言在低资源场景下的准确率。	contrastive learning
10	Progress or Regress? Self-Improvement Reversal in Post-training	提出全面评估框架以解决自我提升逆转问题	preference learning large language model