Zirui Tao

Team Seedance, De Chen, Liyang Chen et al. · gatech

Seedance 2.0 is a new native multi-modal audio-video generation model, officially released in China in early February 2026. Compared with its predecessors, Seedance 1.0 and 1.5 Pro, Seedance 2.0 adopts a unified, highly efficient, and large-scale architecture for multi-modal audio-video joint generation. This allows it to support four input modalities: text, image, audio, and video, by integrating one of the most comprehensive suites of multi-modal content reference and editing capabilities available in the industry to date. It delivers substantial, well-rounded improvements across all key sub-dimensions of video and audio generation. In both expert evaluations and public user tests, the model has demonstrated performance on par with the leading levels in the field. Seedance 2.0 supports direct generation of audio-video content with durations ranging from 4 to 15 seconds, with native output resolutions of 480p and 720p. For multi-modal inputs as reference, its current open platform supports up to 3 video clips, 9 images, and 3 audio clips. In addition, we provide Seedance 2.0 Fast version, an accelerated variant of Seedance 2.0 designed to boost generation speed for low-latency scenarios. Seedance 2.0 has delivered significant improvements to its foundational generation capabilities and multi-modal generation performance, bringing an enhanced creative experience for end users.

Team Seedance, Heyi Chen, Siyan Chen et al.

Recent strides in video generation have paved the way for unified audio-visual generation. In this work, we present Seedance 1.5 pro, a foundational model engineered specifically for native, joint audio-video generation. Leveraging a dual-branch Diffusion Transformer architecture, the model integrates a cross-modal joint module with a specialized multi-stage data pipeline, achieving exceptional audio-visual synchronization and superior generation quality. To ensure practical utility, we implement meticulous post-training optimizations, including Supervised Fine-Tuning (SFT) on high-quality datasets and Reinforcement Learning from Human Feedback (RLHF) with multi-dimensional reward models. Furthermore, we introduce an acceleration framework that boosts inference speed by over 10X. Seedance 1.5 pro distinguishes itself through precise multilingual and dialect lip-syncing, dynamic cinematic camera control, and enhanced narrative coherence, positioning it as a robust engine for professional-grade content creation. Seedance 1.5 pro is now accessible on Volcano Engine at https://console.volcengine.com/ark/region:ark+cn-beijing/experience/vision?type=GenVideo.

Seong Jae Hwang, Zirui Tao, Won Hwa Kim et al.

Generative models using neural network have opened a door to large-scale studies for various application domains, especially for studies that suffer from lack of real samples to obtain statistically robust inference. Typically, these generative models would train on existing data to learn the underlying distribution of the measurements (e.g., images) in latent spaces conditioned on covariates (e.g., image labels), and generate independent samples that are identically distributed in the latent space. Such models may work for cross-sectional studies, however, they are not suitable to generate data for longitudinal studies that focus on "progressive" behavior in a sequence of data. In practice, this is a quite common case in various neuroimaging studies whose goal is to characterize a trajectory of pathologies of a specific disease even from early stages. This may be too ambitious especially when the sample size is small (e.g., up to a few hundreds). Motivated from the setup above, we seek to develop a conditional generative model for longitudinal data generation by designing an invertable neural network. Inspired by recurrent nature of longitudinal data, we propose a novel neural network that incorporates recurrent subnetwork and context gating to include smooth transition in a sequence of generated data. Our model is validated on a video sequence dataset and a longitudinal AD dataset with various experimental settings for qualitative and quantitative evaluations of the generated samples. The results with the AD dataset captures AD specific group differences with sufficiently generated longitudinal samples that are consistent with existing literature, which implies a great potential to be applicable to other disease studies.