Search papers, labs, and topics across Lattice.
Lumos-Nexus is introduced as a two-stage framework for training-efficient unified video generation, first aligning a lightweight generator with the understanding block, then using Unified Progressive Frequency Bridging (UPFB) to hand off generation to a high-capacity pretrained generator during inference. UPFB operates in a shared latent space, enabling coarse-to-fine refinement and high-fidelity video generation without sacrificing reasoning quality. The introduction of VR-Bench helps evaluate reasoning-driven video generation, and experiments show Lumos-Nexus achieves significant improvements in visual realism and temporal coherence on VBench and strong reasoning-based generative performance on VR-Bench.
Achieve high-fidelity video generation without compromising reasoning by progressively handing off generation from a lightweight, reasoning-aligned generator to a high-capacity pretrained generator in a shared latent space.
Connector-based video unified models have demonstrated strong capability in instruction-grounded video synthesis, but integrating a large high-fidelity generator into the unified training loop is computationally prohibitive, limiting achievable visual quality. We therefore propose Lumos-Nexus, a training-efficient unified video generation framework that facilitates the development of strong reasoning-driven generation capabilities while significantly enhancing visual fidelity. Lumos-Nexus adopts a two-stage design: 1) During training, only a lightweight generator is aligned with the understanding block to learn to take in reasoning-driven semantic control. 2) During inference, we introduce Unified Progressive Frequency Bridging (UPFB) to progressively hand off generation to a high-capacity pretrained generator in the shared latent space, enabling coarse-to-fine refinement and producing high-fidelity videos without compromising reasoning quality. To fill the gap in reasoning-driven video generation benchmarks, we introduce VR-Bench, which assesses a model's capability to translate inferred intent into coherent and semantically aligned video content. Extensive experiments demonstrate that Lumos-Nexus achieves substantial gains in visual realism and temporal coherence on VBench, while exhibiting strong reasoning-based generative performance on VR-Bench. Code and models are available at https://jiazheng-xing.github.io/nexus-lumos-home/.
