Hearing loss, affecting individuals of all ages, can impair education, social function, and quality of life. Current treatments, such as hearing aids and implants, aim to mitigate these effects but often fall short in addressing the critical issue of accurately pinpointing sound sources. We report an intelligent hearing system inspired by the human auditory system: asymmetric well-aligned piezoelectric nanofibers combined with neural networks to mimic natural auditory processes. Piezoelectric nanofibers with spirally varying lengths and directions transmit and convert acoustic sound into mechanoelectrical signals, mimicking the complex cochlear dynamics. These signals are then encoded by digital neural networks, enabling accurate sound direction recognition. This intelligent hearing system surpasses human directional hearing, accurately recognizing sound directions horizontally and vertically. The advancement represents a substantial stride toward next-generation artificial hearing, harmonizing transduction and perception with a nature-inspired design. It promises for applications in hearing aids, wearable devices, and implants, offering enhanced auditory experiences for those with hearing impairments.

The long-tail effect is a common issue that limits the performance of deep learning models on real-world datasets. Character image datasets are also affected by such unbalanced data distribution due to differences in character usage frequency. Thus, current character recognition methods are limited when applied in the real world, especially for the categories in the tail that lack training samples, e.g., uncommon characters. In this paper, we propose a zero-shot character recognition framework via radical-based reasoning, called RZCR, to improve the recognition performance of few-sample character categories in the tail. Specifically, we exploit radicals, the graphical units of characters, by decomposing and reconstructing characters according to orthography. RZCR consists of a visual semantic fusion-based radical information extractor (RIE) and a knowledge graph character reasoner (KGR). RIE aims to recognize candidate radicals and their possible structural relations from character images in parallel. The results are then fed into KGR to recognize the target character by reasoning with a knowledge graph. We validate our method on multiple datasets, and RZCR shows promising experimental results, especially on few-sample character datasets.

Knowledge graphs (KGs) have been proven to be effective for improving the performance of recommender systems. KGs can store rich side information and relieve the data sparsity problem. There are many linked attributes between entity pairs (e.g., items and users) in KGs, which can be called multiple-step relation paths. Existing methods do not sufficiently exploit the information encoded in KGs. In this paper, we propose MRP2Rec to explore various semantic relations in multiple-step relation paths to improve recommendation performance. The knowledge representation learning approach is used in our method to learn and represent multiple-step relation paths, and they are further utilized to generate prediction lists by inner products in top-K recommendations. Experiments on two real-world datasets demonstrate that our model achieves higher performance compared with many state-of-the-art baselines.

Optical character recognition (OCR) methods have been applied to diverse tasks, e.g., street view text recognition and document analysis. Recently, zero-shot OCR has piqued the interest of the research community because it considers a practical OCR scenario with unbalanced data distribution. However, there is a lack of benchmarks for evaluating such zero-shot methods that apply a divide-and-conquer recognition strategy by decomposing characters into radicals. Meanwhile, radical recognition, as another important OCR task, also lacks radical-level annotation for model training. In this paper, we construct an ancient Chinese character image dataset that contains both radical-level and character-level annotations to satisfy the requirements of the above-mentioned methods, namely, ACCID, where radical-level annotations include radical categories, radical locations, and structural relations. To increase the adaptability of ACCID, we propose a splicing-based synthetic character algorithm to augment the training samples and apply an image denoising method to improve the image quality. By introducing character decomposition and recombination, we propose a baseline method for zero-shot OCR. The experimental results demonstrate the validity of ACCID and the baseline model quantitatively and qualitatively.

Traffic forecasting is the foundation of intelligent transportation systems (ITS). In recent, graph neural networks (GNNs) have successfully captured spatial-temporal dependencies to forecast traffic conditions by transforming traffic data in the graph domain. Nevertheless, the existing methods focus only on learning informative graph representations and fail to model informative graph structures, which hinders the capture of dynamic spatial-temporal dependencies caused by dynamic factors such as weather, accidents, and special events. In this paper, we propose a novel task-oriented Spatial Graph Structure Learning (SGSL) method, which aims to capture dynamic dependencies by jointly learning graph structures and graph representations. Compared to methods that use spectral graph representations, we exploit a learnable spatial graph to effectively model dynamic dependencies in traffic data. Moreover, we directly define graph convolutions on spatial relations to specify different edge weights when aggregating the information of spatial neighbours. Thus, the graph structure alterations, i.e., the relation changes, and the time-varying weights of relations can be encapsulated, thereby effectively representing dynamic dependencies. The gradient descent strategy is introduced to periodically learn a spatial graph through joint optimization with a newly designed deep graph learning model named GAT-nLSTM. In this manner, the intrinsic behaviours of nodes are learned to capture correlations across periods. Notably, the optimization process is performed under the traffic forecasting constraint to ensure that the learned spatial graph is specific to this task. Compared with those of state-of-the-art baselines, the experimental results obtained on real-world traffic datasets show significant improvement, which verifies the superiority of the proposed SGSL.