With the rapid advancement of video generation models such as Sora, video quality assessment (VQA) is becoming increasingly crucial for selecting high-quality videos from large-scale datasets used in pre-training. Traditional VQA methods, typically producing single numerical scores, often lack comprehensiveness and interpretability. To address these challenges, we introduce MVQA-68K, a novel multi-dimensional VQA dataset comprising over 68,000 carefully annotated videos, covering seven essential quality dimensions: overall aesthetics, camera movement, dynamic degree, texture detail, composition, visual quality, and factual consistency. Each annotation includes detailed chain-of-thought reasoning to facilitate interpretability and comprehensive understanding. Extensive experiments demonstrate that MVQA-68K significantly enhances the performance of various multimodal large language models (MLLMs) on the VQA task, achieving state-of-the-art results not only on our internal test set (Fig.1) but also on public benchmarks including LSVQ-test, LSVQ-1080p, and LIVE-VQC. Meantime, incorporating explicit reasoning process during VQA training substantially boosts the zero-shot generalization. Code and dataset will be available at github: https://github.com/Controller01-ai/MVQA-68K.

Modeling and predicting the radio spectrum is vital for spectrum management, such as spectrum sharing and anomaly detection. Nevertheless, the precise spectrum prediction is challenging due to the interference from both intra-spectrum and external factors. To tackle these complex internal and external correlations, we develop a model named TF$^2$AN, consisting of three components: 1) a robust signal detection algorithm based on image processing, 2) an attention-based Long Short-term Memory network to capture the temporal-frequency correlations, 3) a generalized fusion module to take the heterogeneous external factors into account. This structure shows prominent effectiveness for spectrum prediction on a single monitoring station with sufficient data. However, when the data derived from a single station is insufficient, the performance of the deep learning model will decline a lot. Considering that more than one monitoring station is deployed in practice, the new challenge becomes how to enhance our model by leveraging the data from multiple stations or frequency bands. Therefore, we further propose T-TF$^2$AN, a transfer learning-based framework for data augmentation and knowledge sharing in spectrum prediction. Compared to TF$^2$AN, better performance is achieved. Besides, the model interpretability and training efficiency are also discussed with two case studies, respectively.

As the IoT technology becomes well established and sharing economy expands worldwide, the bike-sharing system (BSS) has spread fast in the last decade. When introducing the BSS in a new city, the operator often faces many challenges: e.g., optimizing station siting (physical or electric), constructing bike lanes, and making strategies for the initial distribution and rebalancing of bikes. These challenges require an accurate interblock flow prediction before deploying the BSS. In this article, we derive blocks from the urban road network and extract blocks’ features based on the distribution and types of POI. Then, the interblock flow can be predicted based on the features of both start/end block and neighbor blocks. We propose a unified architecture with a generalized attention mechanism named the relation graph network (RGN) to extract block features and make predictions across cities. The evaluations on real-world data sets show that RGN outperforms other popular graph neural networks in multiple metrics. We further simulate the applications of station site selection and bike lane planning based on the prediction from RGN, and the result provides a noticeable increase in the BSS’s deployment/operation efficiency.

As the sharing economy develops and bike-sharing apps emerge, the dockless bike-sharing system (DLBS) has become a competitive alternative to the docked bike-sharing system because of its convenience of finding and parking without physical docks. Meanwhile, new demands are rapidly increasing as DLBS expands, e.g., crowd-sourced re-balancing and pre-ordering during rush hours. A more fine-grained destination prediction is required to tackle these issues. In this paper, we propose a probabilistic-trip-based destination prediction method named P3M. To overcome the uncertainty due to docks’ absence, we introduce the virtual docks derived from POIs and convert a single trip recorded in GPS into several probabilistic trips among POIs using an innovative user behavior model Walking-Riding-Walking Probabilistic Trip. To deal with sparsity, P3M adapts a trip-wised parameter share strategy together with a statistical-based history-feature extractor for better performance without overfitting. Compared with the baseline method, P3M reduces the mean absolute errors measured with distance by 31.55% (from 1.1036 km to 0.7554 km) and is less sensitive to the sparsity of user’s records. Further, we analyze the application of P3M in different types of DLBS and use two simulations to prove its efficiency under insufficient bike supply circumstances.

Radio map, serving as an efficient indicator of wireless environments, has been widely used in smart-city applications, including network monitoring/planning, anomaly signal detection, and indoor/outdoor localization. It is hard to maintain an update-to-date fine-grained radio map within a large area, since the radio map changes rapidly due to the internal and external factors. Previous studies usually relied on time-consuming site surveys at densely predefined reference points, leading to either coarse-grained or out-of-date radio maps. In this paper, we propose a fine-grained radio map reconstruction framework, called Supreme, based on crowd-sourced data in an image super-resolution manner. Specifically, Supreme explores spatial-temporal relationships in historical coarse-grained radio maps and builds a real-time fine-grained radio map using deep spatial-temporal reconstruction networks. Furthermore, a heterogeneous data fusion module is devised to make full use of external information. To evaluate the performance of Supreme, we conduct extensive experiments and ablation studies on a large-scale dataset with a total of six-month data collected from two university campuses. Besides, we investigate the transferability of Supreme in different locations and service networks, showing that the fine-tuned model can largely reduce the training time and achieve better performance. Experimental results demonstrate that our model outperforms state-of-the-art baselines and a case study on the localization is enhanced with marginal improvements on accuracy.

Modeling and predicting radio spectrum are significant for better understanding the behavior of spectrum, managing their usage as well as optimizing the performance of dynamic spectrum access. Most of the existing works concentrate on predicting the occupation status of the spectrum via threshold-based binary time series, ignoring abundant frequency correlations. In fact, precisely predicting the energy level of the radio spectrum can provide richer information for applications such as characterizing the spectrum trending for earlier anomaly detection and estimating the channel quality for efficient spectrum sharing. However, the precise prediction is challenging due to the interference from both intra-spectrum and external factors. In this paper, we propose a temporal-frequency fusion attention network to model the complex internal and external correlations for precise prediction. More specifically, our framework consists of three major components: 1) an image processing based robust signal detection algorithm to locate the signal as model input. 2) an attention-based Long Short-term Memory network to model the temporal-frequency correlation of the spectrum. 3) a generalized fusion module to take in the external factors from heterogeneous domains. Extensive experiments are conducted on real-world datasets collected by our spectrum monitoring station deployed in the city of Hangzhou, China, which shows that the proposed signal detection algorithm is robust for frequency bands with different signal to noise ratios. Furthermore, experimental results demonstrate that our method outperforms seven baseline methods in terms of prediction accuracy. The sensitivities of hyper-parameters are analyzed and the interpretability is also well discussed to prove the effectiveness of our method.