Text data augmentation is an effective strategy for overcoming the challenge of limited sample sizes in many natural language processing (NLP) tasks. This challenge is especially prominent in the few-shot learning (FSL) scenario, where the data in the target domain is generally much scarcer and of lowered quality. A natural and widely used strategy to mitigate such challenges is to perform data augmentation to better capture data invariance and increase the sample size. However, current text data augmentation methods either can’t ensure the correct labeling of the generated data (lacking faithfulness), or can’t ensure sufficient diversity in the generated data (lacking compactness), or both. Inspired by the recent success of large language models (LLM), especially the development of ChatGPT, we propose a text data augmentation approach based on ChatGPT (named ”AugGPT”). AugGPT rephrases each sentence in the training samples into multiple conceptually similar but semantically different samples. The augmented samples can then be used in downstream model training. Experiment results on multiple few-shot learning text classification tasks show the superior performance of the proposed AugGPT approach over state-of-the-art text data augmentation methods in terms of testing accuracy and distribution of the augmented samples.

Posting images on social network platforms is happening everywhere and every single second. Thus, the communication channels offered by various social networks have a great potential for covert communication. However, images transmitted through such channels will usually be JPEG compressed, which fails most of the existing steganographic schemes. In this paper, we propose a novel image steganography framework that is robust for such channels. In particular, we first obtain the channel compressed version (i.e., the channel output) of the original image. Secret data is embedded into the channel compressed original image by using any of the existing JPEG steganographic schemes, which produces the stego-image after the channel transmission. To generate the corresponding image before the channel transmission (termed the intermediate image), we propose a coefficient adjustment scheme to slightly modify the original image based on the stego-image. The adjustment is done such that the channel compressed version of the intermediate image is exactly the same as the stego-image. Therefore, after the channel transmission, secret data can be extracted from the stego-image with 100% accuracy. Various experiments are conducted to show the effectiveness of the proposed framework for image steganography robust to JPEG compression.

Graph is ubiquitous in many real world applications ranging from social network analysis to biology. How to correctly and effectively learn and extract information from graph is essential for a large number of machine learning tasks. Graph embedding is a way to transform and encode data structure in high dimensional and Non-Euclidean feature space to a low dimensional and structural space. We have witnessed a huge surge of such embedding methods, from statistical approaches to recent deep learning methods such as graph convolutional networks (GCN). However, most of existing GCN methods can only perform convolution operations with node features, while ignoring the handy information in edge features. To address this problem, we present CensNet, Convolution with Edge-Node Switching graph neural network for learning tasks in graph-structured data with both node and edge features. By using line graph of the original undirected graph, the role of nodes and edges are switched, and two novel graph convolution operations are proposed for feature propagation. Experimental results on real-world academic citation networks and quantum chemistry graphs show that our approach achieves or matches the state-of-the-art performance, in four graph learning tasks including semi-supervised node classification, multi-task graph classification, graph regression and link prediction.

In this paper, we aim at learning robust and discriminative subspaces from noisy data. Subspace learning is widely used in extracting discriminative features for classification. However, when data are contaminated with severe noise, the performance of most existing subspace learning methods would be limited. Recent advances in low-rank modeling provide effective solutions for removing noise or outliers contained in sample sets, which motivates us to take advantage of low-rank constraints in order to exploit robust and discriminative subspace for classification. In particular, we present a discriminative subspace learning method called the supervised regularization-based robust subspace (SRRS) approach, by incorporating the low-rank constraint. SRRS seeks low-rank representations from the noisy data, and learns a discriminative subspace from the recovered clean data jointly. A supervised regularization function is designed to make use of the class label information, and therefore to enhance the discriminability of subspace. Our approach is formulated as a constrained rank-minimization problem. We design an inexact augmented Lagrange multiplier optimization algorithm to solve it. Unlike the existing sparse representation and low-rank learning methods, our approach learns a low-dimensional subspace from recovered data, and explicitly incorporates the supervised information. Our approach and some baselines are evaluated on the COIL-100, ALOI, Extended YaleB, FERET, AR, and KinFace databases. The experimental results demonstrate the effectiveness of our approach, especially when the data contain considerable noise or variations.

Visual data such as images and videos are easily accessible nowadays, and they play critical roles in many real-world applications like surveillance. This raises a series of technological demands for automatic visual understanding and content summarization, which has guided the research community to move towards a better achievement of such capabilities. Meanwhile, it presents the big challenge of semantic understanding of video content and automatically translating them into human language. When developing such automatic translation systems, one critical issue is how to bridge the gap between low level features and high level semantic information. Furthermore, as a large amount of videos are captured under unconstrained conditions by nonprofessional users, this issue becomes even more serious. Therefore, brand new sets of technologies are required to address these difficulties and narrow the semantic gap effectively. These thoughts drive us to survey the complete state-of-the-art techniques in the visual to text topic. Existing methods, popular datasets, technical difficulties, and promising future directions are discussed systematically. In particular, we classify existing methods by their mechanism to link visual information (including both images and videos) and text descriptions, and emphasize the latest advances on deep learning based approaches. The quantitative evaluations of representative approaches on benchmark dataset are also presented and discussed. Finally, we provide with the promising research directions on this topic.

Remote sensing imagery is vital for environmental and land use monitoring, necessitating efficient segmentation methods. While deep learning models have become standard for segmentation, they require extensive per-pixel annotations, which are often scarce. High intra-dataset variance also limits the transferability of these models. Although models like SAM offer zero-shot instance-level segmentation, adapting them for semantic segmentation remains challenging. We introduce Text2Seg, a method for remote sensing semantic segmentation that reduces dependency on extensive annotations by automatically generating prompts using visual foundation models (VFMs). This approach enhances generalization across datasets. Experiments on four remote sensing datasets show Text2Seg significantly outperforms the vanilla SAM model, with improvements ranging from 31% to 225%. Our code is available at https://github.com/Douglas2Code/Text2Seg.