◇ ─── ◈ ─── ◇
✧ · ꕤ · ✧
This repository collects open-source research from DAMO Academy, Alibaba Group in customized video generation, currently including works to customize identities/attributes and lighting for videos, and reasoning-driven unified video generation. Code is organized into self-contained subprojects for separate setup and reproduction.
[2025/9/19] Accepted by NeurIPS 2025 !
[2025/10/29] Code is available now!
[2026/1/26] Accepted by ICLR 2026 !
[2026/3/21] Code is available now!
[2026/6/19] Accepted by ECCV 2026 !
[2026/5/31] Code is available now!
If you are interested in our foundational video generation research, please refer to the Lumos project.
| Project | Venue | In one sentence | Code & docs |
|---|---|---|---|
| Lumos-Nexus | ECCV 2026 | Lumos-Nexus advances reasoning-driven unified video generation through lightweight connector training and Unified Progressive Frequency Bridging at inference. | Lumos-Nexus/ · README |
| LumosX | ICLR 2026 | LumosX advances personalized multi-subject video generation through relational data design and relational attention modeling. | LumosX/ · README |
| UniLumos | NeurIPS 2025 | UniLumos advances unified image and video relighting through RGB-space geometry feedback on a flow-matching backbone. | UniLumos/ · README |
✦ arXiv preprint ✦
Lumos-Nexus: Efficient Frequency Bridging with Homogeneous Latent Space for Video Unified Models
Reasoning-driven video generation · VR-Bench tasks
✧ High-level physical world reasoning (ETV)
Task prompt
|
Lumos-Nexus
|
✧ High-level commonsense reasoning (CCR)
Task prompt
|
Lumos-Nexus
|
✧ Embodied physical reasoning (BBR)
Task prompt
|
Lumos-Nexus
|
➜ Task prompts: Lumos-Nexus/asserts/readme_png/ · Result GIFs: Lumos-Nexus/asserts/readme_gif/ · more in Lumos-Nexus/README.md
- Venue: arXiv preprint (2603.20192; not yet peer-reviewed)
- Summary: We propose Lumos-Nexus, a training-efficient unified video generation framework. A lightweight generator is aligned with the understanding block during training; at inference, Unified Progressive Frequency Bridging (UPFB) hands off to a high-capacity pretrained generator in a shared latent space for high-fidelity output. Companion evaluation is provided by VR-Bench (see
Lumos-Nexus/vr_bench_eval/).
Quick links
- Paper: arXiv · Project page
- Documentation: Lumos-Nexus/README.md — installation, OmniVideo checkpoints, batch inference, and VR-Bench evaluation
✦ ICLR 2026 ✦
LumosX: Relate Any Identities with Their Attributes for Personalized Video Generation
Identity-consistent · Subject-consistent personalized generation
✧ Identity consistency
Reference
|
Result
|
✧ Subject consistency
Reference
|
Result
|
➜ Reference: LumosX/asserts/images/ · Result GIFs: LumosX/asserts/videos/ · more in LumosX/README.md
- Venue: ICLR 2026
- Summary: We propose LumosX, a framework that advances both data and model design for personalized video generation. The data pipeline builds relational structure from captions and MLLM-derived priors; the model uses Relational Self-Attention and Relational Cross-Attention to encode subject–attribute dependencies. Companion evaluation resources live under
LumosX/benchmark/.
Quick links
- Model weights: Hugging Face · LumosX
- Documentation: LumosX/README.md — installation, checkpoints, inference, and benchmark evaluation
✦ NeurIPS 2025 ✦
UniLumos: Fast and Unified Image and Video Relighting with Physics-Plausible Feedback
Unified image & video relighting · physics-plausible feedback
 |
 |
 |
 |
➜ Assets live under UniLumos/assets/ (same as UniLumos/README.md) · add the GIFs locally if the folder is empty
- Venue: NeurIPS 2025
- Summary: We propose UniLumos, a unified relighting framework for images and videos. Supervision uses depth and normal maps from model outputs to align lighting with scene geometry; path consistency learning keeps this effective under few-step training. Companion evaluation is provided by LumosBench (see
UniLumos/LumosBench/).
Quick links
- Model weights: Hugging Face · UniLumos
- Documentation: UniLumos/README.md — installation, checkpoints, inference, and LumosBench evaluation
Lumos-Custom/
├── README.md # This file: umbrella overview
├── LumosX/ # ICLR 2026 · personalized multi-subject video generation
│ └── README.md
├── UniLumos/ # NeurIPS 2025 · unified relighting + LumosBench/
│ ├── README.md
│ └── LumosBench/
└── Lumos-Nexus/ # arXiv preprint · reasoning-driven unified video generation + VR-Bench
├── README.md
└── vr_bench_eval/
git clone https://github.com/alibaba-damo-academy/Lumos-Custom.git
cd Lumos-Custom
# LumosX
cd LumosX
# Follow LumosX/README.md
# or UniLumos
cd ../UniLumos
# Follow UniLumos/README.md
# or Lumos-Nexus
cd ../Lumos-Nexus
# Follow Lumos-Nexus/README.mdIf you use any project, please cite the corresponding paper. BibTeX entries are in the Citation section of each subproject README.md.
@inproceedings{UniLumos,
title={UniLumos: Fast and Unified Image and Video Relighting with Physics-Plausible Feedback},
author={Liu, Pengwei and Yuan, Hangjie and Dong, Bo and Xing, Jiazheng and Wang, Jinwang and Zhao, Rui and Chen, Weihua and Wang, Fan},
booktitle={The Thirty-ninth Annual Conference on Neural Information Processing Systems}
}
@inproceedings{LumosX,
title={LumosX: Relate Any Identities with Their Attributes for Personalized Video Generation},
author={Xing, Jiazheng and Du, Fei and Yuan, Hangjie and Liu, Pengwei and Xu, Hongbin and Ci, Hai and Niu, Ruigang and Chen, Weihua and Wang, Fan and Liu, Yong},
booktitle={The Fourteenth International Conference on Learning Representations}
}
@misc{LumosNexus,
title={Lumos-Nexus: Efficient Frequency Bridging with Homogeneous Latent Space for Video Unified Models},
author={Jiazheng Xing and Hangjie Yuan and Lingling Cai and Xinyu Liu and Yujie Wei and Fei Du and Hai Ci and Tao Feng and Jiasheng Tang and Weihua Chen and Fan Wang and Yong Liu},
year={2026},
eprint={2605.31603},
archivePrefix={arXiv},
primaryClass={cs.CV}
}- Foundational video generation: Lumos-1.