Research map / Vision-language and multimodal / Human and embodied modeling
Discrete Diffusion Models with MLLMs for Unified Medical Multimodal Generation
Jiawei Mao, Yuhan Wang, Lifeng Chen, Can Zhao, Yucheng Tang, Dong Yang, Liangqiong Qu, Daguang Xu, Yuyin Zhou
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 |
|---|---|---|---|
| use MLLM backbone instead of DiT (ablation) | MeDiM · MIMIC-CXR + PathGen ablation (mean) | mFID: 63.22 -> 20.40 | Improved |
| use MLLM backbone instead of DiT (ablation) | MeDiM · MIMIC-CXR + PathGen ablation (mean) | mMETEOR: 0.214 -> 0.262 | Improved |
| use MLLM backbone instead of UniDisc backbone (ablation) | MeDiM · MIMIC-CXR + PathGen ablation (mean) | mFID: 51.59 -> 20.40 | Improved |
| use MLLM backbone instead of UniDisc backbone (ablation) | MeDiM · MIMIC-CXR + PathGen ablation (mean) | mMETEOR: 0.255 -> 0.262 | Improved |
| add pretrained MLLM weights (ablation) | MeDiM · MIMIC-CXR + PathGen ablation (mean) | mFID: 68.27 -> 20.40 | Improved |
| add timestep embedding (ablation) | MeDiM · MIMIC-CXR + PathGen ablation (mean) | mFID: 40.03 -> 20.40 | Improved |
| add timestep embedding (ablation) | MeDiM · MIMIC-CXR + PathGen ablation (mean) | mMETEOR: 0.246 -> 0.262 | Improved |
| add AdaLN designs (ablation) | MeDiM · MIMIC-CXR + PathGen ablation (mean) | mFID: 32.68 -> 20.40 | Improved |
| add AdaLN designs (ablation) | MeDiM · MIMIC-CXR + PathGen ablation (mean) | mMETEOR: 0.247 -> 0.262 | Improved |
| remove causal mask (bidirectional attention) (ablation) | MeDiM · MIMIC-CXR + PathGen ablation (mean) | mFID: 143.72 -> 20.40 | Improved |
| remove causal mask (bidirectional attention) (ablation) | MeDiM · MIMIC-CXR + PathGen ablation (mean) | mMETEOR: 0.142 -> 0.262 | Improved |
| MeDiM instead of SDM (SFT) for image generation | MeDiM · MIMIC-CXR image generation | FID: 78.97 -> 16.60 | 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.
- MMaDA* — MMaDA: Multimodal Large Diffusion Language Models (NeurIPS 2025) · better on 4 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
- LLaDA-MedV: Exploring Large Language Diffusion Models for Biomedical Image Understanding (CVPR 2026 · similarity 0.90)
- MedMO: Grounding and Understanding Multimodal Large Language Model for Medical Images (CVPR 2026 · similarity 0.89)
- MedMax: Mixed-Modal Instruction Tuning for Training Biomedical Assistants (NeurIPS 2025 · similarity 0.89)
- MedUnifier: Unifying Vision-and-Language Pre-training on Medical Data with Vision Generation Task using Discrete Visual Representations (CVPR 2025 · similarity 0.88)
- How Do Medical MLLMs Fail? A Study on Visual Grounding in Medical Images (ICLR 2026 · similarity 0.88)
- Toward a Vision-Language Foundation Model for Medical Data: Multimodal Dataset and Benchmarks for Vietnamese PET/CT Report Generation (NeurIPS 2025 · 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.
