Blending Is All You Need: Cheaper, Better Alternative to Trillion-Parameters LLM
This work addresses the computational inefficiency of large-scale conversational AI models, offering a potentially cheaper and more accessible alternative for developers and researchers, though it appears incremental as it builds on existing ensemble-like techniques.
This paper tackles the problem of high computational costs in large language models by proposing 'blending', a method that combines multiple smaller models to achieve performance comparable to or better than a single large model like ChatGPT, with results showing that blending three moderate-sized models (6B/13B parameters) can rival or surpass a 175B+ parameter model in A/B tests over 30 days.
In conversational AI research, there's a noticeable trend towards developing models with a larger number of parameters, exemplified by models like ChatGPT. While these expansive models tend to generate increasingly better chat responses, they demand significant computational resources and memory. This study explores a pertinent question: Can a combination of smaller models collaboratively achieve comparable or enhanced performance relative to a singular large model? We introduce an approach termed "blending", a straightforward yet effective method of integrating multiple chat AIs. Our empirical evidence suggests that when specific smaller models are synergistically blended, they can potentially outperform or match the capabilities of much larger counterparts. For instance, integrating just three models of moderate size (6B/13B paramaeters) can rival or even surpass the performance metrics of a substantially larger model like ChatGPT (175B+ paramaters). This hypothesis is rigorously tested using A/B testing methodologies with a large user base on the Chai research platform over a span of thirty days. The findings underscore the potential of the "blending" strategy as a viable approach for enhancing chat AI efficacy without a corresponding surge in computational demands.