Kimi.ai(@Kimi_Moonshot ):We're introducing WorldVQA, a new benchmark to measure atomic vision-centric world knowledge in Multimodal Large Language Models. Current evaluations often conflate visual knowledge retrieval with reasoning. In contrast, WorldVQA decouples these capabilities to strictly measure "what the model memorizes." The benchmark consists of 3,500 VQA pairs across 9 categories, with careful attention to linguistic and cultural diversity:

Kimi.ai

@Kimi_Moonshot

Built by Moonshot AI to empower everyone to be superhuman. ⚡️API: @KimiProduct where we share cool use cases and prompts.

Joined December 2024

136 Following 168.5K Followers

Kimi.ai@Kimi_Moonshot

2026.02.03 14:46

We're introducing WorldVQA, a new benchmark to measure atomic vision-centric world knowledge in Multimodal Large Language Models. Current evaluations often conflate visual knowledge retrieval with reasoning. In contrast, WorldVQA decouples these capabilities to strictly measure "what the model memorizes." The benchmark consists of 3,500 VQA pairs across 9 categories, with careful attention to linguistic and cultural diversity: