Search

This work aims to predict the popularity of short videos using the videos themselves and their related features. Popularity is measured by four key engagement metrics: view count, like count, comment count, and share count. This study employs video classification models with different architectures and training methods as backbone networks to extract video modality features. Meanwhile, the cleaned video captions are incorporated into a carefully designed prompt framework, along with the video, as input for video-to-text generation models, which generate detailed text-based video content understanding. These texts are then encoded into vectors using a pre-trained BERT model. Based on the six sets of vectors mentioned above, a neural network is trained for each of the four prediction metrics. Moreover, the study conducts data mining and feature engineering based on the video and tabular data, constructing practical features such as the total frequency of hashtag appearances, the total frequency of mention appearances, video duration, frame count, frame rate, and total time online. Multiple machine learning models are trained, and the most stable model, XGBoost, is selected. Finally, the predictions from the neural network and XGBoost models are averaged to obtain the final result., Comment: INFORMS 2024 Data Challenge Competition

Neural Audio Codecs, initially designed as a compression technique, have gained more attention recently for speech generation. Codec models represent each audio frame as a sequence of tokens, i.e., discrete embeddings. The discrete and low-frequency nature of neural codecs introduced a new way to generate speech with token-based models. As these tokens encode information at various levels of granularity, from coarse to fine, most existing works focus on how to better generate the coarse tokens. In this paper, we focus on an equally important but often overlooked question: How can we better resynthesize the waveform from coarse tokens? We point out that both the choice of learning target and resynthesis approach have a dramatic impact on the generated audio quality. Specifically, we study two different strategies based on token prediction and regression, and introduce a new method based on Schr\"odinger Bridge. We examine how different design choices affect machine and human perception., Comment: NeurIPS 2024 Audio Imagination workshop paper; demo page at https://alexander-h-liu.github.io/codec-resyn.github.io/

As interactions between humans and AI become more prevalent, it is critical to have better predictors of human behavior in these interactions. We investigated how changes in the AI's adaptive algorithm impact behavior predictions in two-player continuous games. In our experiments, the AI adapted its actions using a gradient descent algorithm under different adaptation rates while human participants were provided cost feedback. The cost feedback was provided by one of two types of visual displays: (a) cost at the current joint action vector, or (b) cost in a local neighborhood of the current joint action vector. Our results demonstrate that AI adaptation rate can significantly affect human behavior, having the ability to shift the outcome between two game theoretic equilibrium. We observed that slow adaptation rates shift the outcome towards the Nash equilibrium, while fast rates shift the outcome towards the human-led Stackelberg equilibrium. The addition of localized cost information had the effect of shifting outcomes towards Nash, compared to the outcomes from cost information at only the current joint action vector. Future work will investigate other effects that influence the convergence of gradient descent games.

Extracting the speech of participants in a conversation amidst interfering speakers and noise presents a challenging problem. In this paper, we introduce the novel task of target conversation extraction, where the goal is to extract the audio of a target conversation based on the speaker embedding of one of its participants. To accomplish this, we propose leveraging temporal patterns inherent in human conversations, particularly turn-taking dynamics, which uniquely characterize speakers engaged in conversation and distinguish them from interfering speakers and noise. Using neural networks, we show the feasibility of our approach on English and Mandarin conversation datasets. In the presence of interfering speakers, our results show an 8.19 dB improvement in signal-to-noise ratio for 2-speaker conversations and a 7.92 dB improvement for 2-4-speaker conversations. Code, dataset available at https://github.com/chentuochao/Target-Conversation-Extraction., Comment: Accepted by Interspeech 2024

Speech restoration aims at restoring high quality speech in the presence of a diverse set of distortions. Although several deep learning paradigms have been studied for this task, the power of the recently emerging language models has not been fully explored. In this paper, we propose MaskSR, a masked language model capable of restoring full-band 44.1 kHz speech jointly considering noise, reverb, clipping, and low bandwidth. MaskSR works with discrete acoustic tokens extracted using a pre-trained neural codec. During training, MaskSR is optimized to predict randomly masked tokens extracted from the high quality target speech, conditioned on the corrupted speech with various distortions. During inference, MaskSR reconstructs the target speech tokens with efficient iterative sampling. Extensive experiments show that MaskSR obtains competitive results on both the full-band speech restoration task and also on sub-tasks compared with a wide range of models., Comment: Accepted by INTERSPEECH 2024. Demo page: https://masksr.github.io/MaskSR/

Multivariate spatial data plays an important role in computational science and engineering simulations. The potential features and hidden relationships in multivariate data can assist scientists to gain an in-depth understanding of a scientific process, verify a hypothesis and further discover a new physical or chemical law. In this paper, we present a comprehensive survey of the state-of-the-art techniques for multivariate spatial data visualization. We first introduce the basic concept and characteristics of multivariate spatial data, and describe three main tasks in multivariate data visualization: feature classification, fusion visualization, and correlation analysis. Finally, we prospect potential research topics for multivariate data visualization according to the current research., Comment: 16 pages, 5 figures. Corresponding author: Yubo Tao

In complex multivariate data sets, different features usually include diverse associations with different variables, and different variables are associated within different regions. Therefore, exploring the associations between variables and voxels locally becomes necessary to better understand the underlying phenomena. In this paper, we propose a co-analysis framework based on biclusters, which are two subsets of variables and voxels with close scalar-value relationships, to guide the process of visually exploring multivariate data. We first automatically extract all meaningful biclusters, each of which only contains voxels with a similar scalar-value pattern over a subset of variables. These biclusters are organized according to their variable sets, and biclusters in each variable set are further grouped by a similarity metric to reduce redundancy and support diversity during visual exploration. Biclusters are visually represented in coordinated views to facilitate interactive exploration of multivariate data based on the similarity between biclusters and the correlation of scalar values with different variables. Experiments on several representative multivariate scientific data sets demonstrate the effectiveness of our framework in exploring local relationships among variables, biclusters and scalar values in the data., Comment: 31 pages, 7 figures