Speech translation (ST) aims to translate speech from a source language into text in the target language. Naturally, speech signals contain paralinguistic cues beyond linguistic content, which could influence or even alter the interpretation of a lexically identical sentence, thereby yielding distinct translations. However, existing ST models lack direct and sufficient modeling of paralinguistic information, which limits their ability to perceive paralinguistic cues and understand speech comprehensively, leading to degraded translation performance. In response, we propose Paralinguistic-aware Speech Translation (PLaST), a novel dual-branch framework which directly leverages paralinguistic cues beyond the linguistic content. Specifically, PLaST employs a speech encoder and a style extractor to independently generate linguistic and paralinguistic representations, respectively. To obtain a purified linguistic representation aligned with the text representation, a hierarchical Optimal Transport (OT) is applied on the layer-wise outputs from an LLM decoder. Then, the paralinguistic information is retrieved and refined with an Attention-based Retrieval (AR) module, with the linguistic representation serving as queries to enable joint guidance for semantic understanding and translation generation. PLaST outperforms the strong baseline with an average of 5.0 directional and 4.5 global contrastive likelihood scores on the paralinguistic-sensitive benchmark ContraProST, demonstrating its superior capability in paralinguistic perception. Further experiments on the standard speech translation benchmark CoVoST-2 show that PLaST generalizes well to typical ST scenarios.

Sign Language Translation (SLT) bridges the communication gap between deaf and hearing individuals by converting sign language videos into spoken language texts. While most SLT research has focused on bilingual translation models, the recent surge in interest has led to the exploration of Multilingual Sign Language Translation (MSLT). However, MSLT presents unique challenges due to the diversity of sign languages across nations. This diversity can lead to cross-linguistic conflicts and hinder translation accuracy. To use the similarity of actions and semantics between sign languages to alleviate conflict, we propose a novel approach that leverages sign language families to improve MSLT performance. Sign languages were clustered into families automatically based on their Language distribution in the MSLT network. We compare the results of our proposed family clustering method with the analysis conducted by sign language linguists and then train dedicated translation models for each family in the many-to-one translation scenario. Our experiments on the SP-10 dataset demonstrate that our approach can achieve a balance between translation accuracy and computational cost by regulating the number of language families.

Sign language recognition (SLR) serves as a critical technological bridge facilitating seamless communication between the hearing-impaired population and the general public. While deep learning-based SLR systems have demonstrated remarkable progress, they remain constrained by three fundamental challenges: (i) the necessity for large-scale annotated datasets, (ii) heightened privacy concerns surrounding biometric data, and (iii) inherent heterogeneity in sign language data distributions. To address these limitations, we present FedSign, a novel model-agnostic federated learning (FL) framework designed for universal applicability across diverse SLR architectures. Our framework simultaneously preserves data privacy through decentralized training while enhancing model robustness in non-independent and identically distributed (Non-IID) environments via an innovative entropy-based pseudo-label selection mechanism. This adaptive approach dynamically optimizes knowledge distillation by selectively leveraging either global or local model outputs based on uncertainty quantification. Comprehensive evaluations on three benchmark datasets (CSL-Daily, Phoenix-2014, and Phoenix-2014T) demonstrate that FedSign significantly outperforms both centralized and federated baselines, particularly in scenarios with extreme data heterogeneity. Notably, our framework establishes new state-of-the-art (SOTA) performance metrics on two challenging German sign language corpora.

The widespread adoption of large language models (LLMs) has increased the need for reliable AI-text detection.While current detectors perform well on benchmark datasets, we highlight a critical vulnerability: increasing the temperature parameter during inference significantly reduces detection accuracy.Based on this weakness, we propose Temp-Paraphraser, a simple yet effective paraphrasing framework that simulates high-temperature sampling effects through multiple normaltemperature generations, effectively evading detection.Experiments show that TempParaphraser reduces detector accuracy by an average of 82.5% while preserving high text quality.We also demonstrate that training on TempParaphraser-augmented data improves detector robustness.

Traditional non-simultaneous Sign Language Translation (SLT) methods, while effective for pre-recorded videos, face challenges in real-time scenarios due to inherent inference delays. The emerging field of simultaneous SLT aims to address this issue by progressively translating incrementally received sign video. However, the sole existing work in simultaneous SLT adopts a fixed gloss-based policy, which suffer from limitations in boundary prediction and contextual comprehension. In this paper, we delve deeper into this area and propose an adaptive policy for simultaneous SLT. Our approach introduces the concept of “confident translation length”, denoting maximum accurate translation achievable from current input. An estimator measures this length for streaming sign video, enabling the model to make informed decisions on whether to wait for more input or proceed with translation. To train the estimator, we construct a training data of confident translation length based on the longest common prefix between translations of partial and complete inputs. Furthermore, we incorporate adaptive training, utilizing pseudo prefix pairs, to refine the offline translation model for optimal performance in simultaneous scenarios. Experimental results on PHOENIX2014T and CSL-Daily demonstrate the superiority of our adaptive policy over existing methods, particularly excelling in situations requiring extremely low latency.

In multivariate chaotic time series prediction, correlation analysis is important for reducing input dimensions and improving prediction performance. Grey relational analysis (GRA) has proved to be an effective method for data correlation analysis, especially for inexact data and incomplete data. In GRA, points are usually regarded as objects, and the distance between points or the concave and convex degree are mostly used to measure the correlations. However, with discrete variables, correlation analysis results always tend to have some deviations when using prior GRA methods. Furthermore, GRA methods cannot directly use vector datasets. Therefore, in this paper, an improved GRA method is proposed based on vector projections. The input and output variables are expressed as vectors by linking two adjacent points. The vectors, instants of the points, are regarded as the objects, and the projection length of input variables to output variables is used to measure the correlations. The smaller the difference between the projection length and the input variables, the higher the correlation. Then, a hybrid variable selection and prediction model is proposed based on the improved GRA method for multivariate chaotic time series predictions, in order to overcome the negative effects of irrelevant and redundant variables caused by phase-space reconstruction. The experimental results based on the gas furnace dataset and San Francisco river runoff dataset demonstrate that the improved GRA method is effective for data correlation analysis, and the prediction accuracy is better than prior GRA-based methods.

We propose an algorithm for sensorless reorientation of 3D convex polyhedral parts through a sequence of step devices. Part is assumed to arrive in an arbitrary orientation and is being translated forward (e.g., on a conveyor belt) at slow speed. After precomputing, our algorithm produces a sequence of O(n) distinct steps, where n is the number of faces in the polyhedral part. As the part passes (drops) through the steps, it successively changes orientation. At the output end, the part will be oriented (for most initial orientations) to a pose such that its center-of-mass is the lowest possible.