English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Multimodal Large Language Model (MLLM) relies on the powerful LLM to performmultimodal tasks, showing amazing emergent abilities in recent studies, such aswriting poems based on an image. However, it is difficult for these casestudies to fully reflect the performance of MLLM, lacking a comprehensiveevaluation. In this paper, we fill in this blank, presenting the firstcomprehensive MLLM Evaluation benchmark MME. It measures both perception andcognition abilities on a total of 14 subtasks. In order to avoid data leakagethat may arise from direct use of public datasets for evaluation, theannotations of instruction-answer pairs are all manually designed. The conciseinstruction design allows us to fairly compare MLLMs, instead of struggling inprompt engineering. Besides, with such an instruction, we can also easily carryout quantitative statistics. A total of 30 advanced MLLMs are comprehensivelyevaluated on our MME, which not only suggests that existing MLLMs still have alarge room for improvement, but also reveals the potential directions for thesubsequent model optimization.

MME: A Comprehensive Evaluation Benchmark for Multimodal Large Language Models