OPT-R: Exploring the Role of Explanations in Finetuning and Prompting for Reasoning Skills of Large Language Models

Badr Alkhamissi, Siddharth Verma, Ping Yu, Zhijing Jin, Asli Celikyilmaz, Mona Diab


Abstract
We conduct a thorough investigation into the reasoning capabilities of Large Language Models (LLMs), focusing specifically on the Open Pretrained Transformers (OPT) models as a representative of such models. Our study entails finetuning three different sizes of OPT on a carefully curated reasoning corpus, resulting in two sets of finetuned models: OPT-R, finetuned without explanations, and OPT-RE, finetuned with explanations. We then evaluate all models on 57 out-of-domain tasks drawn from the Super-NaturalInstructions benchmark, covering 26 distinct reasoning skills, utilizing three prompting techniques. Through a comprehensive grid of 27 configurations and 6,156 test evaluations, we investigate the dimensions of finetuning, prompting, and scale to understand the role of explanations on different reasoning skills. Our findings reveal that having explanations in the fewshot exemplar has no significant impact on the model’s performance when the model is finetuned, while positively affecting the non-finetuned counterpart. Moreover, we observe a slight yet consistent increase in classification accuracy as we incorporate explanations during prompting and finetuning, respectively. Finally, we offer insights on which reasoning skills benefit the most from incorporating explanations during finetuning and prompting, such as Numerical (+20.4%) and Analogical (+13.9%) reasoning, as well as skills that exhibit negligible or negative effects.
Anthology ID:
2023.nlrse-1.10
Volume:
Proceedings of the 1st Workshop on Natural Language Reasoning and Structured Explanations (NLRSE)
Month:
June
Year:
2023
Address:
Toronto, Canada
Editors:
Bhavana Dalvi Mishra, Greg Durrett, Peter Jansen, Danilo Neves Ribeiro, Jason Wei
Venue:
NLRSE
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
128–138
Language:
URL:
https://aclanthology.org/2023.nlrse-1.10
DOI:
10.18653/v1/2023.nlrse-1.10
Bibkey:
Cite (ACL):
Badr Alkhamissi, Siddharth Verma, Ping Yu, Zhijing Jin, Asli Celikyilmaz, and Mona Diab. 2023. OPT-R: Exploring the Role of Explanations in Finetuning and Prompting for Reasoning Skills of Large Language Models. In Proceedings of the 1st Workshop on Natural Language Reasoning and Structured Explanations (NLRSE), pages 128–138, Toronto, Canada. Association for Computational Linguistics.
Cite (Informal):
OPT-R: Exploring the Role of Explanations in Finetuning and Prompting for Reasoning Skills of Large Language Models (Alkhamissi et al., NLRSE 2023)
Copy Citation:
PDF:
https://aclanthology.org/2023.nlrse-1.10.pdf