Enhancing Alignment for Unified Multimodal Models via Semantically-Grounded Supervision

Open paper
Benchmark Model Rank Results
text-to-image-generation-on-dpg-benchSeGroS-Harmon-1.5B (extra data)–DPG-Bench Score: 86.58
text-to-image-generation-on-genevalSeGroS-Harmon-1.5B (extra data)–Overall: 0.8866

Rank counts only results with a code link.