Saved in:
Bibliographic Details
Main Authors: Spangher, Lucas, Li, Tianle, Arnold, William F., Masiewicki, Nick, Dotiwalla, Xerxes, Parusmathi, Rama, Grabowski, Peter, Ie, Eugene, Gruhl, Dan
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2410.22368
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910677803728896
author Spangher, Lucas
Li, Tianle
Arnold, William F.
Masiewicki, Nick
Dotiwalla, Xerxes
Parusmathi, Rama
Grabowski, Peter
Ie, Eugene
Gruhl, Dan
author_facet Spangher, Lucas
Li, Tianle
Arnold, William F.
Masiewicki, Nick
Dotiwalla, Xerxes
Parusmathi, Rama
Grabowski, Peter
Ie, Eugene
Gruhl, Dan
contents There exists an extremely wide array of LLM benchmarking tasks, whereas oftentimes a single number is the most actionable for decision-making, especially by non-experts. No such aggregation schema exists that is not Elo-based, which could be costly or time-consuming. Here we propose a method to aggregate performance across a general space of benchmarks, nicknamed Project "MPG," dubbed Model Performance and Goodness, additionally referencing a metric widely understood to be an important yet inaccurate and crude measure of car performance. Here, we create two numbers: a "Goodness" number (answer accuracy) and a "Fastness" number (cost or QPS). We compare models against each other and present a ranking according to our general metric as well as subdomains. We find significant agreement between the raw Pearson correlation of our scores and those of Chatbot Arena, even improving on the correlation of the MMLU leaderboard to Chatbot Arena.
format Preprint
id arxiv_https___arxiv_org_abs_2410_22368
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Project MPG: towards a generalized performance benchmark for LLM capabilities
Spangher, Lucas
Li, Tianle
Arnold, William F.
Masiewicki, Nick
Dotiwalla, Xerxes
Parusmathi, Rama
Grabowski, Peter
Ie, Eugene
Gruhl, Dan
Software Engineering
Artificial Intelligence
There exists an extremely wide array of LLM benchmarking tasks, whereas oftentimes a single number is the most actionable for decision-making, especially by non-experts. No such aggregation schema exists that is not Elo-based, which could be costly or time-consuming. Here we propose a method to aggregate performance across a general space of benchmarks, nicknamed Project "MPG," dubbed Model Performance and Goodness, additionally referencing a metric widely understood to be an important yet inaccurate and crude measure of car performance. Here, we create two numbers: a "Goodness" number (answer accuracy) and a "Fastness" number (cost or QPS). We compare models against each other and present a ranking according to our general metric as well as subdomains. We find significant agreement between the raw Pearson correlation of our scores and those of Chatbot Arena, even improving on the correlation of the MMLU leaderboard to Chatbot Arena.
title Project MPG: towards a generalized performance benchmark for LLM capabilities
topic Software Engineering
Artificial Intelligence
url https://arxiv.org/abs/2410.22368