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 Models (MLLMs) have shown their remarkableabilities in visual perception and understanding recently. However, how tocomprehensively evaluate the capabilities of MLLMs remains a challenge. Most ofthe existing benchmarks predominantly focus on assessing perception, cognition,and reasoning, neglecting the abilities of self-awareness, referring to themodel's recognition of its own capability boundary. In our study, we focus onself-awareness in image perception and introduce the knowledge quadrant forMLLMs, which clearly defines the knowns and unknowns in perception. Based onthis, we propose a novel benchmark specifically designed to evaluate theSelf-Aware capabilities in Perception for MLLMs(MM-SAP). MM-SAP encompassesthree distinct sub-datasets, each focusing on different aspects ofself-awareness. We evaluated eight well-known MLLMs using MM-SAP, analyzingtheir self-awareness and providing detailed insights. Code and data areavailable at https://github.com/YHWmz/MM-SAP

MM-SAP: A Comprehensive Benchmark for Assessing Self-Awareness of Multimodal Large Language Models in Perception