Large Language Models (LLMs) are equipped with profound semantic knowledge, making them a natural choice for injecting semantic generalization into personalized search systems. However, in practice we find that directly fine-tuning LLMs on industrial personalized tasks (e.g. next item prediction) often yields suboptimal results. We attribute this bottleneck to a critical Knowledge--Action Gap: the inherent conflict between preserving pre-trained semantic knowledge and aligning with specific personalized actions by discriminative objectives. Empirically, action-only training objectives induce Semantic Collapse, such as attention ''sinks''. This degradation severely cripples the LLM's generalization, failing to bring improvements to personalized search systems.

Meta-learning has recently emerged as a promising technique to address the challenge of few-shot learning. However, standard meta-learning methods mainly focus on visual tasks, which makes it hard for them to deal with diverse text data directly. In this paper, we introduce a novel framework for few-shot text classification, which is named as MEta-learning with Data Augmentation (MEDA). MEDA is composed of two modules, a ball generator and a meta-learner, which are learned jointly. The ball generator is to increase the number of shots per class by generating more samples, so that meta-learner can be trained with both original and augmented samples. It is worth noting that ball generator is agnostic to the choice of the meta-learning methods. Experiment results show that on both datasets, MEDA outperforms existing state-of-the-art methods and significantly improves the performance of meta-learning on few-shot text classification.

Social recommendation is effective in improving the recommendation performance by leveraging social relations from online social networking platforms. Social relations among users provide friends’ information for modeling users’ interest in candidate items and help items expose to potential consumers (i.e., item attraction). However, there are two issues haven’t been well-studied: Firstly, for the user interests, existing methods typically aggregate friends’ information contextualized on the candidate item only, and this shallow context-aware aggregation makes them suffer from the limited friends’ information. Secondly, for the item attraction, if the item’s past consumers are the friends of or have a similar consumption habit to the targeted user, the item may be more attractive to the targeted user, but most existing methods neglect the relation enhanced context-aware item attraction.

Although deep neural networks have made significant progress in semantic segmentation, speed and computational cost still can't meet the strict requirements of real-world applications. In this paper, we present an enhanced asymmetric convolution network (EACNet) to seek a balance between accuracy and speed. Specifically, we design a pair of enhancing asymmetric convolution modules constructed by depth-wise asymmetric convolution and dilated convolution to extract short-range and long-range features, which are efficient and powerful. Additionally, we apply a bilateral structure in which the detail branch preserves low-level spatial details while the semantic branch captures high-level context information. The two branches are merged at different stages of the network to strengthen information propagation between different levels. The experiments on the Cityscapes dataset show that our method achieves high accuracy and speed with relatively small parameters. Compared with other real-time semantic segmentation methods, our network attains a good trade-off among parameters, speed, and accuracy.

The authors propose a new single image dehazing method. Different from image restoration and image enhancement method, their method is based on the idea of image fusion. Image dehazing is to remove the influence of the haze between the scene and the camera. First, combined with the depth information, the haze layer is subtracted in the hazy image to improve the colour saturation, which produces the first input image. Then, the gamma correction is used on the grey image. Second, the details of the gamma correction image are enhanced to produce the second input image. Finally, the two input images are fused by local linear model to obtain the final restored image. Experimental results show that the restored image has high contrast, rich details, and without colour distortion in the sky area.

In most applications, component extraction is important when components of non-stationary multi-component signal are key features to be monitored and analyzed. Existing methods are either sensitive to noise or forced to select a proper time-frequency representation for the considered signal. In this paper, we present a novel component extraction method for non-stationary multi-component signal. The proposed method combines parameterized de-chirping and band-pass filter to obtain components of multi-component signal, which avoids dealing with time-frequency representation of the signal and works well under heavy noise. In addition, it is able to analyze the multi-component signal whose components have intersected instantaneous frequency trajectories. Simulation results show that the proposed method is promising in analyzing complicated multi-component signals. Moreover, it works effective in a high noise environment in terms of improving the output signal-to-noise rate for the interested component.