
대규모 언어 모델의 효과적인 인스트럭션 튜닝을 위해서는 데이터셋의 양보다는 다양한 도메인을 포괄하는 인스트럭션의 다양성이 중요하다.


coremsg

인스트럭션-다양성이-일반화에-미치는-결정적-영향-only-if-연구


인스트럭션 다양성이 일반화에 미치는 결정적 영향: ONLY-IF 연구


title_rewrite


Commonality-aware Instruction Tuning (CommonIT) improves the instruction-following abilities of large language models by grouping similar training data into batches, enhancing the model's ability to learn and generalize across various tasks.


commonit-a-new-instruction-tuning-method-for-large-language-models-based-on-data-partitioning-


CommonIT: A New Instruction Tuning Method for Large Language Models Based on Data Partitioning



Proposing a novel continual instruction tuning method based on Key-part Information Gain (KPIG) to improve LLMs' instruction-following and generalization abilities.


don-t-half-listen-key-part-information-in-continual-instruction-tuning


Don't Half-listen: Key-part Information in Continual Instruction Tuning
