Democratizing LLMs: An Exploration of Cost-Performance Trade-offs in Self-Refined Open-Source Models

Sumuk Shashidhar, Abhinav Chinta, Vaibhav Sahai, Zhenhailong Wang, Heng Ji


Abstract
The dominance of proprietary LLMs has led to restricted access and raised information privacy concerns. The SoTA open-source alternatives are crucial for information-sensitive and high-volume applications but often lag behind in performance. To address this gap, we propose (1) A generalized variant of iterative self-critique and self-refinement devoid of external influence. (2) A novel ranking metric - Performance, Refinement, and Inference Cost Score (PeRFICS) - to find the optimal model for a given task considering refined performance and cost. Our experiments show that SoTA open source models of varying sizes from 7B - 65B, on average, improve 8.2% from their baseline performance. Strikingly, even models with extremely small memory footprints, such as Vicuna-7B, show a 11.74% improvement overall and up to a 25.39% improvement in high-creativity, open ended tasks on the Vicuna benchmark. Vicuna-13B takes it a step further and outperforms ChatGPT post-refinement. This work has profound implications for resource-constrained and information-sensitive environments seeking to leverage LLMs without incurring prohibitive costs, compromising on performance and privacy. The domain-agnostic self-refinement process coupled with our novel ranking metric facilitates informed decision-making in model selection, thereby reducing costs and democratizing access to high-performing language models, as evidenced by three case studies on personal computing, gaming and enterprise solutions.
Anthology ID:
2023.findings-emnlp.608
Volume:
Findings of the Association for Computational Linguistics: EMNLP 2023
Month:
December
Year:
2023
Address:
Singapore
Editors:
Houda Bouamor, Juan Pino, Kalika Bali
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
9070–9084
Language:
URL:
https://aclanthology.org/2023.findings-emnlp.608
DOI:
10.18653/v1/2023.findings-emnlp.608
Bibkey:
Cite (ACL):
Sumuk Shashidhar, Abhinav Chinta, Vaibhav Sahai, Zhenhailong Wang, and Heng Ji. 2023. Democratizing LLMs: An Exploration of Cost-Performance Trade-offs in Self-Refined Open-Source Models. In Findings of the Association for Computational Linguistics: EMNLP 2023, pages 9070–9084, Singapore. Association for Computational Linguistics.
Cite (Informal):
Democratizing LLMs: An Exploration of Cost-Performance Trade-offs in Self-Refined Open-Source Models (Shashidhar et al., Findings 2023)
Copy Citation:
PDF:
https://preview.aclanthology.org/ingest-acl-2023-videos/2023.findings-emnlp.608.pdf