WISE
Emerging8papers using it
2025first seen
WISE is a benchmark used to evaluate multimodal understanding and generation models, containing a dataset of 23 million image-text pairs.
Papers using WISE (8)
- UniCorn: Towards Self-Improving Unified Multimodal Models through Self-Generated SupervisionRethinking UMM Visual Generation: Masked Modeling for Efficient Image-Only Pre-trainingMural: Transferring LLM knowledge to image generation via Mixture-of-TransformersHYDRA: Unifying Multi-modal Generation and Understanding via Representation-Harmonized TokenizationDeepGen 1.0: A Lightweight Unified Multimodal Model for Advancing Image Generation and EditingGenAgent: Scaling Text-to-Image Generation via Agentic Multimodal ReasoningOpenUni: A Simple Baseline for Unified Multimodal Understanding and GenerationHarmonizing Visual Representations for Unified Multimodal Understanding
and Generation