2023 SelfInstructAligningLanguageMod
- (Wang et al., 2023) ⇒ Yizhong Wang, Yeganeh Kordi, Swaroop Mishra, Alisa Liu, Noah A. Smith, Daniel Khashabi, and Hannaneh Hajishirzi. (2023). “Self-Instruct: Aligning Language Models with Self-Generated Instructions.” doi:10.48550/arXiv.2212.10560
Subject Headings: Large "Instruction-Tuned" Language Model.
Notes
- It can generate a large and diverse synthetic instruction dataset by prompting a language model.
- It can improve language models' ability to follow instructions by finetuning them on the synthetic data.
- It can help build better instruction-following models with minimal human effort.
- It provides a new benchmark for evaluating instruction-following.
- ...
Cited By
Quotes
Abstract
Large "instruction-tuned" language models (i.e., finetuned to respond to instructions) have demonstrated a remarkable ability to generalize zero-shot to new tasks. Nevertheless, they depend heavily on human-written instruction data that is often limited in quantity, diversity, and creativity, therefore hindering the generality of the tuned model. We introduce Self-Instruct, a framework for improving the instruction-following capabilities of pretrained language models by bootstrapping off their own generations. Our pipeline generates instructions, input, and output samples from a language model, then filters invalid or similar ones before using them to finetune the original model. Applying our method to the vanilla GPT3, we demonstrate a 33% absolute improvement over the original model on Super-NaturalInstructions, on par with the performance of InstructGPT-001, which was trained with private user data and human annotations. For further evaluation, we curate a set of expert-written instructions for novel tasks, and show through human evaluation that tuning GPT3 with Self-Instruct outperforms using existing public instruction datasets by a large margin, leaving only a 5% absolute gap behind InstructGPT-001. Self-Instruct provides an almost annotation-free method for aligning pre-trained language models with instructions, and we release our large synthetic dataset to facilitate future studies on instruction tuning. Our code and data are available at this https URL.
References
;
| Author | volume | Date Value | title | type | journal | titleUrl | doi | note | year | |
|---|---|---|---|---|---|---|---|---|---|---|
| 2023 SelfInstructAligningLanguageMod | Noah A. Smith Hannaneh Hajishirzi Yizhong Wang Yeganeh Kordi Swaroop Mishra Alisa Liu Daniel Khashabi | Self-Instruct: Aligning Language Models with Self-Generated Instructions | 10.48550/arXiv.2212.10560 | 2023 |