Señorita-2M: A High-Quality Instruction-based Dataset for General Video Editing by Video Specialists

Recent advancements in video generation have spurred the development of video editing techniques, which can be divided into inversion-based and end-to-end methods. However, current video editing methods still suffer from several challenges. Inversion-based methods, though training-free and flexible, are time-consuming during inference, struggle with fine-grained editing instructions, and produce artifacts and jitter. On the other hand, end-to-end methods, which rely on edited video pairs for training, offer faster inference speeds but often produce poor editing results due to a lack of high-quality training video pairs. In this paper, to close the gap in end-to-end methods, we introduce Se\~norita-2M, a high-quality video editing dataset. Se\~norita-2M consists of approximately 2 millions of video editing pairs. It is built by crafting four high-quality, specialized video editing models, each crafted and trained by our team to achieve state-of-the-art editing results. We also propose a filtering pipeline to eliminate poorly edited video pairs. Furthermore, we explore common video editing architectures to identify the most effective structure based on current pre-trained generative model. Extensive experiments show that our dataset can help to yield remarkably high-quality video editing results. More details are available at https://senorita.github.io.

Enhancing Motion in Text-to-Video Generation with Decomposed Encoding and Conditioning

Despite advancements in Text-to-Video (T2V) generation, producing videos with realistic motion remains challenging. Current models often yield static or minimally dynamic outputs, failing to capture complex motions described by text. This issue stems from the internal biases in text encoding, which overlooks motions, and inadequate conditioning mechanisms in T2V generation models. To address this, we propose a novel framework called DEcomposed MOtion (DEMO), which enhances motion synthesis in T2V generation by decomposing both text encoding and conditioning into content and motion components. Our method includes a content encoder for static elements and a motion encoder for temporal dynamics, alongside separate content and motion conditioning mechanisms. Crucially, we introduce text-motion and video-motion supervision to improve the model's understanding and generation of motion. Evaluations on benchmarks such as MSR-VTT, UCF-101, WebVid-10M, EvalCrafter, and VBench demonstrate DEMO's superior ability to produce videos with enhanced motion dynamics while maintaining high visual quality. Our approach significantly advances T2V generation by integrating comprehensive motion understanding directly from textual descriptions. Project page: https://PR-Ryan.github.io/DEMO-project/

Client-wise Modality Selection for Balanced Multi-modal Federated Learning

Selecting proper clients to participate in the iterative federated learning (FL) rounds is critical to effectively harness a broad range of distributed datasets. Existing client selection methods simply consider the variability among FL clients with uni-modal data, however, have yet to consider clients with multi-modalities. We reveal that traditional client selection scheme in MFL may suffer from a severe modality-level bias, which impedes the collaborative exploitation of multi-modal data, leading to insufficient local data exploration and global aggregation. To tackle this challenge, we propose a Client-wise Modality Selection scheme for MFL (CMSFed) that can comprehensively utilize information from each modality via avoiding such client selection bias caused by modality imbalance. Specifically, in each MFL round, the local data from different modalities are selectively employed to participate in local training and aggregation to mitigate potential modality imbalance of the global model. To approximate the fully aggregated model update in a balanced way, we introduce a novel local training loss function to enhance the weak modality and align the divergent feature spaces caused by inconsistent modality adoption strategies for different clients simultaneously. Then, a modality-level gradient decoupling method is designed to derive respective submodular functions to maintain the gradient diversity during the selection progress and balance MFL according to local modality imbalance in each iteration. Our extensive experiments showcase the superiority of CMSFed over baselines and its effectiveness in multi-modal data exploitation.

MGAS: Multi-Granularity Architecture Search for Effective and Efficient Neural Networks

Differentiable architecture search (DAS) revolutionizes neural architecture search (NAS) with time-efficient automation, transitioning from discrete candidate sampling and evaluation to differentiable super-net optimization and discretization. However, existing DAS methods either only conduct coarse-grained operation-level search or manually define the remaining ratios for fine-grained kernel-level and weight-level units, which fail to simultaneously optimize model size and model performance. Furthermore, these methods compromise search quality to reduce memory consumption. To tackle these issues, we introduce multi-granularity architecture search (MGAS), a unified framework which aims to comprehensively and memory-efficiently explore the multi-granularity search space to discover both effective and efficient neural networks. Specifically, we learn discretization functions specific to each granularity level to adaptively determine the remaining ratios according to the evolving architecture. This ensures an optimal balance among units of different granularity levels for different target model sizes. Considering the memory demands, we break down the super-net optimization and discretization into multiple sub-net stages. Nevertheless, the greedy nature of this approach may introduce bias in the early stages. To compensate for the bias, we propose progressive re-evaluation to allow for re-pruning and regrowing of previous units during subsequent stages. Extensive experiments on CIFAR-10, CIFAR-100 and ImageNet demonstrate that MGAS outperforms other state-of-the-art methods in achieving a better trade-off between model performance and model size.