Authors: Shaun Stoltz
There have been significant improvements in directing large language models (LLM) to answer logic-based question such as mathematical reasoning tasks. This has resulted in near perfect performance on these types of problems with accuracy levels in the mid ninety percentile level using state of the art models (GPT-4). The achievement of this level of accuracy has previously needed a multi-prompt approach to elicit better performances from LLM’s. This paper introduces a new prompt paradigm termed "Mega prompt" and further introduces Proteus, a state of the art mega prompt, that has been used to achieve a new level of accuracy on the GSM8K math data set of 97%.
Comments: 10 Pages.
Download: PDF
[v1] 2023-06-12 02:41:42
[v2] 2023-10-10 01:20:34
Unique-IP document downloads: 1483 times
Vixra.org is a pre-print repository rather than a journal. Articles hosted may not yet have been verified by peer-review and should be treated as preliminary. In particular, anything that appears to include financial or legal advice or proposed medical treatments should be treated with due caution. Vixra.org will not be responsible for any consequences of actions that result from any form of use of any documents on this website.
Add your own feedback and questions here:
You are equally welcome to be positive or negative about any paper but please be polite. If you are being critical you must mention at least one specific error, otherwise your comment will be deleted as unhelpful.