Abstract Sarcasm detection is a nuanced challenge in natural language processing, requiring deep understanding of textual and contextual cues. We present Sarcasm-GPT, a large language model-based model that integrates four key components: prompt template generation, retrieval-augmented generation, chain-of-thought generation, and a context fusion module. Together, these modules enrich contextual modeling, enable systematic reasoning, and seamlessly incorporate multimodal information, significantly boosting sarcasm detection accuracy. Extensive experiments, including ablation studies, confirm the complementary contributions of each module and demonstrate substantial performance gains over baselines. Additionally, our findings highlight the importance of context length in balancing interpretive accuracy and computational efficiency. The code is available at Sarcasm-GPT.

Abstract Although multimodal face data have obvious advantages in describing live and spoofed features, single‐modality face antispoofing technologies are still widely used when it is difficult to obtain multimodal face images or inconvenient to integrate and deploy multimodal sensors. Since the live/spoofed representations in visible light facial images include considerable face identity information interference, existing deep learning‐based face antispoofing models achieve poor performance when only the visible light modality is used. To address the above problems, the authors design a dual‐channel network structure and a constrained representation learning method for face antispoofing. First, they design a dual‐channel attention mechanism‐based grouped convolutional neural network (CNN) to learn important deceptive cues in live and spoofed faces. Second, they design inner contrastive estimation‐based representation constraints for both live and spoofed samples to minimise the sample similarity loss to prevent the CNN from learning more facial appearance information. This increases the distance between live and spoofed faces and enhances the network's ability to identify deceptive cues. The evaluation results indicate that the framework we designed achieves an average classification error rate (ACER) of 2.37% on the visible light modality subset of the CASIA‐SURF dataset and an ACER of 2.4% on the CASIA‐SURF CeFA dataset, outperforming existing methods. The proposed method achieves low ACER scores in cross‐dataset testing, demonstrating its advantage in domain generalisation.

Summary Face antispoofing (FAS) is attracting increasing attention from researchers because of its important role in preventing facial recognition systems from face spoofing attacks. With the advancement of various new convolutional neural network structures and the construction of various face antispoofing databases, the deep learning‐based face antispoofing algorithm has become the main method in the FAS field. However, the generalization performance of the current multimodal face antispoofing algorithm is poor, and the recognition performance of the model on different datasets is quite different. Therefore, we design a multimodal face antispoofing framework based on a multifeature transformer (MFViT) and multirank fusion (MRF). First, we use a vision transformer structure, MFViT, for multimodal face antispoofing and a combination of modalities to capture the distinguishing characteristics in each modality. Second, we design a multidimensional multimodal fusion module, MRF, according to the various modal fusion characteristics obtained by the MFViT to fuse modal information in different dimensions more effectively. Evaluation results indicate that framework we designed achieves an average classification error rate (ACER) of 1.61% on the CASIA‐SURF dataset and an ACER of 6.5% on the CASIA‐SURF CeFA dataset.

The dimensionality reduction techniques are often used to reduce data dimensionality for computational efficiency or other purposes in existing low-rank representation (LRR)-based methods. However, the two steps of dimensionality reduction and learning low-rank representation coefficients are implemented in an independent way; thus, the adaptability of representation coefficients to the original data space may not be guaranteed. This article proposes a novel model, i.e., low-rank representation with adaptive dimensionality reduction (LRRARD) via manifold optimization for clustering, where dimensionality reduction and learning low-rank representation coefficients are integrated into a unified framework. This model introduces a low-dimensional projection matrix to find the projection that best fits the original data space. And the low-dimensional projection matrix and the low-rank representation coefficients interact with each other to simultaneously obtain the best projection matrix and representation coefficients. In addition, a manifold optimization method is employed to obtain the optimal projection matrix, which is an unconstrained optimization method in a constrained search space. The experimental results on several real datasets demonstrate the superiority of our proposed method.

Aspect sentiment triplet extraction (ASTE) is a challenging subtask in aspect-based sentiment analysis.It aims to explore the triplets of aspects, opinions and sentiments with complex correspondence from the context.The bidirectional machine reading comprehension (BMRC) can effectively deal with ASTE task, but several problems remains, such as query conflict and probability unilateral decrease.Therefore, this paper presents a robustly optimized BMRC method by incorporating four improvements.The word segmentation is applied to facilitate the semantic learning.Exclusive classifiers are designed to avoid the interference between different queries.A span matching rule is proposed to select the aspects and opinions that better represent the expectations of the model.The probability generation strategy is also introduced to obtain the predicted probability for aspects, opinions and aspect-opinion pairs.We have conducted extensive experiments on multiple benchmark datasets, where our model achieves the stateof-the-art performance.1