تقرير
LIMIT: Less Is More for Instruction Tuning Across Evaluation Paradigms
| العنوان: | LIMIT: Less Is More for Instruction Tuning Across Evaluation Paradigms |
|---|---|
| المؤلفون: | Jha, Aditi, Havens, Sam, Dohmann, Jeremy, Trott, Alex, Portes, Jacob |
| سنة النشر: | 2023 |
| المجموعة: | Computer Science |
| مصطلحات موضوعية: | Computer Science - Machine Learning, Computer Science - Artificial Intelligence, Computer Science - Computation and Language |
| الوصف: | Large Language Models are traditionally finetuned on large instruction datasets. However recent studies suggest that small, high-quality datasets can suffice for general purpose instruction following. This lack of consensus surrounding finetuning best practices is in part due to rapidly diverging approaches to LLM evaluation. In this study, we ask whether a small amount of diverse finetuning samples can improve performance on both traditional perplexity-based NLP benchmarks, and on open-ended, model-based evaluation. We finetune open-source MPT-7B and MPT-30B models on instruction finetuning datasets of various sizes ranging from 1k to 60k samples. We find that subsets of 1k-6k instruction finetuning samples are sufficient to achieve good performance on both (1) traditional NLP benchmarks and (2) model-based evaluation. Finally, we show that mixing textbook-style and open-ended QA finetuning datasets optimizes performance on both evaluation paradigms. Comment: 36 pages, 12 figures, NeurIPS 2023 Workshop on Instruction Tuning and Instruction Following |
| نوع الوثيقة: | Working Paper |
| URL الوصول: | http://arxiv.org/abs/2311.13133 |
| رقم الأكسشن: | edsarx.2311.13133 |
| قاعدة البيانات: | arXiv |
| الوصف غير متاح. |