AVMeme Exam: A Multimodal Multilingual Multicultural Benchmark for LLMs' Contextual and Cultural Knowledge and Thinking
This work addresses the problem of limited contextual and cultural knowledge in AI models for researchers and developers in multimodal AI, though it is incremental as it builds on existing benchmarking efforts.
The authors introduced AVMeme Exam, a benchmark of over 1,000 internet audio-visual memes to test AI models' understanding of contextual and cultural signals, finding that current multimodal large language models perform poorly on textless music and sound effects and struggle with contextual and cultural thinking compared to surface content.
Internet audio-visual clips convey meaning through time-varying sound and motion, which extend beyond what text alone can represent. To examine whether AI models can understand such signals in human cultural contexts, we introduce AVMeme Exam, a human-curated benchmark of over one thousand iconic Internet sounds and videos spanning speech, songs, music, and sound effects. Each meme is paired with a unique Q&A assessing levels of understanding from surface content to context and emotion to usage and world knowledge, along with metadata such as original year, transcript, summary, and sensitivity. We systematically evaluate state-of-the-art multimodal large language models (MLLMs) alongside human participants using this benchmark. Our results reveal a consistent limitation: current models perform poorly on textless music and sound effects, and struggle to think in context and in culture compared to surface content. These findings highlight a key gap in human-aligned multimodal intelligence and call for models that can perceive contextually and culturally beyond the surface of what they hear and see. Project page: avmemeexam.github.io/public