VMAS: Video-to-Music Generation via Semantic Alignment in Web Music Videos

We present a framework for learning to generate background music from video inputs. Unlike existing works that rely on symbolic musical annotations, which are limited in quantity and diversity, our method leverages large-scale web videos accompanied by background music. This enables our model to learn to generate realistic and diverse music. To accomplish this goal, we develop a generative video-music Transformer with a novel semantic video-music alignment scheme. Our model uses a joint autoregressive and contrastive learning objective, which encourages the generation of music aligned with high-level video content. We also introduce a novel video-beat alignment scheme to match the generated music beats with the low-level motions in the video. Lastly, to capture fine-grained visual cues in a video needed for realistic background music generation, we introduce a new temporal video encoder architecture, allowing us to efficiently process videos consisting of many densely sampled frames. We train our framework on our newly curated DISCO-MV dataset, consisting of 2.2M video-music samples, which is orders of magnitude larger than any prior datasets used for video music generation. Our method outperforms existing approaches on the DISCO-MV and MusicCaps datasets according to various music generation evaluation metrics, including human evaluation. Results are available at https://genjib.github.io/project_page/VMAs/index.html

Future Does Matter: Boosting 3D Object Detection with Temporal Motion Estimation in Point Cloud Sequences

Accurate and robust LiDAR 3D object detection is essential for comprehensive scene understanding in autonomous driving. Despite its importance, LiDAR detection performance is limited by inherent constraints of point cloud data, particularly under conditions of extended distances and occlusions. Recently, temporal aggregation has been proven to significantly enhance detection accuracy by fusing multi-frame viewpoint information and enriching the spatial representation of objects. In this work, we introduce a novel LiDAR 3D object detection framework, namely LiSTM, to facilitate spatial-temporal feature learning with cross-frame motion forecasting information. We aim to improve the spatial-temporal interpretation capabilities of the LiDAR detector by incorporating a dynamic prior, generated from a non-learnable motion estimation model. Specifically, Motion-Guided Feature Aggregation (MGFA) is proposed to utilize the object trajectory from previous and future motion states to model spatial-temporal correlations into gaussian heatmap over a driving sequence. This motion-based heatmap then guides the temporal feature fusion, enriching the proposed object features. Moreover, we design a Dual Correlation Weighting Module (DCWM) that effectively facilitates the interaction between past and prospective frames through scene- and channel-wise feature abstraction. In the end, a cascade cross-attention-based decoder is employed to refine the 3D prediction. We have conducted experiments on the Waymo and nuScenes datasets to demonstrate that the proposed framework achieves superior 3D detection performance with effective spatial-temporal feature learning.

Multimodal Causal Reasoning Benchmark: Challenging Vision Large Language Models to Infer Causal Links Between Siamese Images

Large Language Models (LLMs) have showcased exceptional ability in causal reasoning from textual information. However, will these causalities remain straightforward for Vision Large Language Models (VLLMs) when only visual hints are provided? Motivated by this, we propose a novel Multimodal Causal Reasoning benchmark, namely MuCR, to challenge VLLMs to infer semantic cause-and-effect relationship when solely relying on visual cues such as action, appearance, clothing, and environment. Specifically, we introduce a prompt-driven image synthesis approach to create siamese images with embedded semantic causality and visual cues, which can effectively evaluate VLLMs' causal reasoning capabilities. Additionally, we develop tailored metrics from multiple perspectives, including image-level match, phrase-level understanding, and sentence-level explanation, to comprehensively assess VLLMs' comprehension abilities. Our extensive experiments reveal that the current state-of-the-art VLLMs are not as skilled at multimodal causal reasoning as we might have hoped. Furthermore, we perform a comprehensive analysis to understand these models' shortcomings from different views and suggest directions for future research. We hope MuCR can serve as a valuable resource and foundational benchmark in multimodal causal reasoning research. The project is available at: https://github.com/Zhiyuan-Li-John/MuCR

Can LLM Graph Reasoning Generalize beyond Pattern Memorization?

Large language models (LLMs) demonstrate great potential for problems with implicit graphical structures, while recent works seek to enhance the graph reasoning capabilities of LLMs through specialized instruction tuning. The resulting 'graph LLMs' are evaluated with in-distribution settings only, thus it remains underexplored whether LLMs are learning generalizable graph reasoning skills or merely memorizing patterns in the synthetic training data. To this end, we propose the NLGift benchmark, an evaluation suite of LLM graph reasoning generalization: whether LLMs could go beyond semantic, numeric, structural, reasoning patterns in the synthetic training data and improve utility on real-world graph-based tasks. Extensive experiments with two LLMs across four graph reasoning tasks demonstrate that while generalization on simple patterns (semantic, numeric) is somewhat satisfactory, LLMs struggle to generalize across reasoning and real-world patterns, casting doubt on the benefit of synthetic graph tuning for real-world tasks with underlying network structures. We explore three strategies to improve LLM graph reasoning generalization, and we find that while post-training alignment is most promising for real-world tasks, empowering LLM graph reasoning to go beyond pattern memorization remains an open research question.

Autoregressive Pretraining with Mamba in Vision

The vision community has started to build with the recently developed state space model, Mamba, as the new backbone for a range of tasks. This paper shows that Mamba's visual capability can be significantly enhanced through autoregressive pretraining, a direction not previously explored. Efficiency-wise, the autoregressive nature can well capitalize on the Mamba's unidirectional recurrent structure, enabling faster overall training speed compared to other training strategies like mask modeling. Performance-wise, autoregressive pretraining equips the Mamba architecture with markedly higher accuracy over its supervised-trained counterparts and, more importantly, successfully unlocks its scaling potential to large and even huge model sizes. For example, with autoregressive pretraining, a base-size Mamba attains 83.2\% ImageNet accuracy, outperforming its supervised counterpart by 2.0\%; our huge-size Mamba, the largest Vision Mamba to date, attains 85.0\% ImageNet accuracy (85.5\% when finetuned with $384\times384$ inputs), notably surpassing all other Mamba variants in vision. The code is available at \url{https://github.com/OliverRensu/ARM}.

Dance Any Beat: Blending Beats with Visuals in Dance Video Generation

The task of generating dance from music is crucial, yet current methods, which mainly produce joint sequences, lead to outputs that lack intuitiveness and complicate data collection due to the necessity for precise joint annotations. We introduce a Dance Any Beat Diffusion model, namely DabFusion, that employs music as a conditional input to directly create dance videos from still images, utilizing conditional image-to-video generation principles. This approach pioneers the use of music as a conditioning factor in image-to-video synthesis. Our method unfolds in two stages: training an auto-encoder to predict latent optical flow between reference and driving frames, eliminating the need for joint annotation, and training a U-Net-based diffusion model to produce these latent optical flows guided by music rhythm encoded by CLAP. Although capable of producing high-quality dance videos, the baseline model struggles with rhythm alignment. We enhance the model by adding beat information, improving synchronization. We introduce a 2D motion-music alignment score (2D-MM Align) for quantitative assessment. Evaluated on the AIST++ dataset, our enhanced model shows marked improvements in 2D-MM Align score and established metrics. Video results can be found on our project page: https://DabFusion.github.io.

Boosting 3D Neuron Segmentation with 2D Vision Transformer Pre-trained on Natural Images

Neuron reconstruction, one of the fundamental tasks in neuroscience, rebuilds neuronal morphology from 3D light microscope imaging data. It plays a critical role in analyzing the structure-function relationship of neurons in the nervous system. However, due to the scarcity of neuron datasets and high-quality SWC annotations, it is still challenging to develop robust segmentation methods for single neuron reconstruction. To address this limitation, we aim to distill the consensus knowledge from massive natural image data to aid the segmentation model in learning the complex neuron structures. Specifically, in this work, we propose a novel training paradigm that leverages a 2D Vision Transformer model pre-trained on large-scale natural images to initialize our Transformer-based 3D neuron segmentation model with a tailored 2D-to-3D weight transferring strategy. Our method builds a knowledge sharing connection between the abundant natural and the scarce neuron image domains to improve the 3D neuron segmentation ability in a data-efficiency manner. Evaluated on a popular benchmark, BigNeuron, our method enhances neuron segmentation performance by 8.71% over the model trained from scratch with the same amount of training samples.

HQ-Edit: A High-Quality Dataset for Instruction-based Image Editing

This study introduces HQ-Edit, a high-quality instruction-based image editing dataset with around 200,000 edits. Unlike prior approaches relying on attribute guidance or human feedback on building datasets, we devise a scalable data collection pipeline leveraging advanced foundation models, namely GPT-4V and DALL-E 3. To ensure its high quality, diverse examples are first collected online, expanded, and then used to create high-quality diptychs featuring input and output images with detailed text prompts, followed by precise alignment ensured through post-processing. In addition, we propose two evaluation metrics, Alignment and Coherence, to quantitatively assess the quality of image edit pairs using GPT-4V. HQ-Edits high-resolution images, rich in detail and accompanied by comprehensive editing prompts, substantially enhance the capabilities of existing image editing models. For example, an HQ-Edit finetuned InstructPix2Pix can attain state-of-the-art image editing performance, even surpassing those models fine-tuned with human-annotated data. The project page is https://thefllood.github.io/HQEdit_web.

Digital Twin Channel for 6G: Concepts, Architectures and Potential Applications

Digital twin channel (DTC) is the real-time mapping of a wireless channel from the physical world to the digital world, which is expected to provide significant performance enhancements for the sixth-generation (6G) air-interface design. In this work, we first define five evolution levels of channel twins with the progression of wireless communication. The fifth level, autonomous DTC, is elaborated with multi-dimensional factors such as methodology, characterization precision, and data category. Then, we provide detailed insights into the requirements and architecture of a complete DTC for 6G. Subsequently, a sensing-enhanced real-time channel prediction platform and experimental validations are exhibited. Finally, drawing from the vision of the 6G network, we explore the potential applications and the open issues in future DTC research.

MMoE: Robust Spoiler Detection with Multi-modal Information and Domain-aware Mixture-of-Experts

Online movie review websites are valuable for information and discussion about movies. However, the massive spoiler reviews detract from the movie-watching experience, making spoiler detection an important task. Previous methods simply focus on reviews' text content, ignoring the heterogeneity of information in the platform. For instance, the metadata and the corresponding user's information of a review could be helpful. Besides, the spoiler language of movie reviews tends to be genre-specific, thus posing a domain generalization challenge for existing methods. To this end, we propose MMoE, a multi-modal network that utilizes information from multiple modalities to facilitate robust spoiler detection and adopts Mixture-of-Experts to enhance domain generalization. MMoE first extracts graph, text, and meta feature from the user-movie network, the review's textual content, and the review's metadata respectively. To handle genre-specific spoilers, we then adopt Mixture-of-Experts architecture to process information in three modalities to promote robustness. Finally, we use an expert fusion layer to integrate the features from different perspectives and make predictions based on the fused embedding. Experiments demonstrate that MMoE achieves state-of-the-art performance on two widely-used spoiler detection datasets, surpassing previous SOTA methods by 2.56% and 8.41% in terms of accuracy and F1-score. Further experiments also demonstrate MMoE's superiority in robustness and generalization.