Research map / Image generation / Text-to-image diffusion
UM-Text: A Unified Multimodal Model for Image Understanding and Visual Text Editing
Lichen Ma, Xiaolong Fu, , GaojingZhou, Zipeng Guo, Ting Zhu, Yichun Liu, Yu Shi, Jason Li, Junshi Huang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add character-level visual encoder (ablation) | FLUX-Fill baseline · AnyText-benchmark Chinese (100k subset) | Sen.ACC: 0.029 -> 0.676 | Improved |
| add character-level visual encoder (ablation) | FLUX-Fill baseline · AnyText-benchmark English (100k subset) | Sen.ACC: 0.309 -> 0.759 | Improved |
| add VLM embedding (ablation) | UM-Text · AnyText-benchmark English (100k subset) | Sen.ACC: 0.759 -> 0.782 | Improved |
| add VLM embedding (ablation) | UM-Text · AnyText-benchmark Chinese (100k subset) | Sen.ACC: 0.676 -> 0.698 | Improved |
| add latent-space regional consistency loss (ablation) | UM-Text · AnyText-benchmark English (100k subset) | Sen.ACC: 0.782 -> 0.799 | Improved |
| add RGB-space regional consistency loss (ablation) | UM-Text · AnyText-benchmark Chinese (100k subset) | Sen.ACC: 0.725 -> 0.746 | Improved |
| UM-Text vs FLUX-Text on LPIPS | UM-Text · AnyText-benchmark Chinese | LPIPS: 0.0487 -> 0.0481 | No clear effect |
| UM-Text vs DreamText on reconstruction LPIPS | UM-Text · UDiffText benchmark Recon | LPIPS: 0.0328 -> 0.0479 | Hurt |
| UM-Text vs DreamText on SeqAcc and FID | UM-Text · UDiffText benchmark Recon | FID: 12.13 -> 6.57 | Improved |
| use UM-Designer layout/text with UM-Text vs AnyText-2 | UM-Text · UMT-benchmark Chinese | Sen.ACC: 0.720 -> 0.956 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- DreamText — DreamText: High Fidelity Scene Text Synthesis (CVPR 2025) · better on 3 of 4 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- The Narrow Gate: Localized Image-Text Communication in Native Multimodal Models (NeurIPS 2025 · similarity 0.89)
- CoMM: A Coherent Interleaved Image-Text Dataset for Multimodal Understanding and Generation (CVPR 2025 · similarity 0.89)
- MMKE-Bench: A Multimodal Editing Benchmark for Diverse Visual Knowledge (ICLR 2025 · similarity 0.89)
- AnyText: Multilingual Visual Text Generation and Editing (ICLR 2024 · similarity 0.88)
- OpenING: A Comprehensive Benchmark for Judging Open-ended Interleaved Image-Text Generation (CVPR 2025 · similarity 0.88)
- Towards Visual Text Design Transfer Across Languages (NeurIPS 2024 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
