Domain Specific
Domain-specific adaptation of large language models (LLMs) focuses on enhancing their performance and reliability within specialized fields by overcoming limitations stemming from data scarcity and domain-specific terminology. Current research emphasizes developing effective methods for data curation, including synthetic data generation and techniques like knowledge distillation to transfer knowledge from domain-specific to general-purpose models, alongside novel architectures like graph-oriented databases for improved performance and maintenance. This work is crucial for broadening the applicability of LLMs to diverse sectors, improving efficiency in areas like finance, healthcare, and scientific research, and addressing concerns about bias and hallucination in sensitive domains.
Papers
Evaluation Ethics of LLMs in Legal Domain
Ruizhe Zhang, Haitao Li, Yueyue Wu, Qingyao Ai, Yiqun Liu, Min Zhang, Shaoping Ma
ProgGen: Generating Named Entity Recognition Datasets Step-by-step with Self-Reflexive Large Language Models
Yuzhao Heng, Chunyuan Deng, Yitong Li, Yue Yu, Yinghao Li, Rongzhi Zhang, Chao Zhang
MediSwift: Efficient Sparse Pre-trained Biomedical Language Models
Vithursan Thangarasa, Mahmoud Salem, Shreyas Saxena, Kevin Leong, Joel Hestness, Sean Lie
Zero-Shot Topic Classification of Column Headers: Leveraging LLMs for Metadata Enrichment
Margherita Martorana, Tobias Kuhn, Lise Stork, Jacco van Ossenbruggen