3 papers
cs.CV2026
A Simple Baseline for Unifying Understanding, Generation, and Editing via Vanilla Next-token Prediction
Jie Zhu, Hanghang Ma, Jia Wang +6
In this work, we introduce Wallaroo, a simple autoregressive baseline that leverages next-token prediction to unify multi-modal understanding, image generation, and editing at the…
cs.CV2026
Forge-and-Quench: Enhancing Image Generation for Higher Fidelity in Unified Multimodal Models
Yanbing Zeng, Jia Wang, Hanghang Ma +4
Integrating image generation and understanding into a single framework has become a pivotal goal in the multimodal domain. However, how understanding can effectively assist generat…
cs.CV2025
MagicMirror: A Large-Scale Dataset and Benchmark for Fine-Grained Artifacts Assessment in Text-to-Image Generation
Jia Wang, Jie Hu, Xiaoqi Ma +3
Text-to-image (T2I) generation has achieved remarkable progress in instruction following and aesthetics. However, a persistent challenge is the prevalence of physical artifacts, su…