AgriBench: A Hierarchical Agriculture Benchmark for Multimodal Large Language Models

Yutong Zhou, Masahiro Ryo · 2024

Read AgriBench: A Hierarchical Agriculture Benchmark for Multimodal Large Language Models by Yutong Zhou, Masahiro Ryo (2024) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2412-00465.txt.

Description

Yutong Zhou and Masahiro Ryo build the first farm benchmark for multimodal language models in a paper from two thousand twenty-four. Such models blend sight and text, but farm knowledge tests were missing. A named data set for training was missing too.

Their benchmark, AgriBench, grades models on farm tasks in layers. The companion set, MM-LUCAS, holds over seventeen hundred landscape photos with masks, depth maps, and rich notes on place, date, land cover, and quality. It grows out of a European land survey. The authors frame both as groundwork, still in progress, for expert models in farming and other specialist fields.

Catalog record

CREATORYutong Zhou, Masahiro Ryo
YEAR2024
COLLECTIONPAPERS
FORMATTXT
FILE SIZE35 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2412-00465

Browse the full library · About the Tower of Babel