Reducing Reasoning Costs - The Path of Optimization for Chain of Thought via Sparse Attention Mechanism
Keywords: sparse attention mechanism, chain of thought reasoning, transformer architecture, costs
TL;DR: This research proves through experimental results that the proposed sparse attention mechanism can effectively chain the costs of thought reasoning in the transformer architecture.
Abstract: In order to address the chain of thought in the large language model inference cost surge, this research proposes to use a sparse attention mechanism that only focuses on a few relevant tokens. The researcher constructed a new attention mechanism and used GiantRabbit trained with custom GPTs as an experimental tool. The experiment tested and compared the reasoning time, correctness score and chain of thought length of this model and o1 Preview in solving the linear algebra test questions of MIT OpenCourseWare. The results show that GiantRabbit's reasoning time and chain of thought length are significantly lower than o1 Preview. It verifies the feasibility of sparse attention mechanism for optimizing chain of thought reasoning. Detailed architectural details and experimental process have been uploaded to Github, the link is:https://github.com/brucewang123456789/GeniusTrail.git.
Submission Number: 4
Loading