Xiang Yin

SwanTale: Unified Multi-Speaker Speech and Audio Generation for Instruct and Zero-Shot Tasks

Yu Zhang, Ruiqi Li, Changhao Pan, Lei Ke et autres

Speech and audio generation is often needed in animation dubbing, audio drama, movies, advertising, games, podcasts, and short-video production. In these scenarios, creators may need to design voices without reference recordings, control speaker styles with natural language, support acoustic scenes with environments …

cn (code pays fourni par la source)

Room‐Temperature 2D Topological Ferrimagnetic Semiconductors With Multiple Electronic Phase Transitions

Li Deng, Fang Wang, Xiang Yin, Junwei Tong et autres

ABSTRACT Integrating ferrimagnetic order and electronic topology in one two‐dimensional material has been highly expected to revolutionize the next‐generation high‐speed and ultralow‐loss information technology, but remains challenging. Here, via a half‐substitution strategy, the FeNb X 2 ( X = Se/Te) monolayer is …

cn, de (code pays fourni par la source)

First-principles prediction of spin filter effect with reversible spin polarization in a ferroelectric/antiferromagnetic/ferroelectric heterostructure

Li Deng, Fei Wang, Xiang Yin, Junwei Tong et autres

Abstract Spin filter effect with reversible spin polarization has long been expected to revolutionize next-generation spintronic devices with low-energy dissipation and nonvolatility. Here, a two-dimensional ferroelectric/antiferromagnetic (AFM)/ferroelectric heterostructure is theoretically proposed to achieve a tunable spin filter effect, originating from spin and …

cn, de (code pays fourni par la source)

InfinityHuman: Towards Long-Term Audio-Driven Human

Xiaodi Li, Pan Xie, Yi Ren, Qijun Gan et autres

Audio-driven human animation has attracted wide attention thanks to its practical applications. However, critical challenges remain in generating high-resolution, long-duration videos with consistent appearance and natural hand motions. Existing methods extend videos using overlapping motion frames but suffer from error accumulation, leading …

Contestability in Quantitative Argumentation

Xiang Yin, Nico Potyka, Antonio Rago, Timotheus Kampik et autres

Contestable AI requires that AI-driven decisions align with human preferences. While various forms of argumentation have been shown to support contestability, Edge-Weighted Quantitative Bipolar Argumentation Frameworks (EW-QBAFs) have received little attention. In this work, we show how EW-QBAFs can be deployed for …

Phased array feed beamforming method based on MLFFNN

Xiang Yin, Chengxiang Zhang

针对现有相控阵馈源（Phased Array Feed，PAF）自适应波束合成方法存在计算复杂高的问题。本文提出一种基于前馈神经网络（Multilayer Feedforward Neural Network，MLFFNN）的相控阵馈源波束合成方法，首先，搭建相控阵馈源几何模型，根据每个阵元的分布特征，建立相控阵馈源信号模型；其次，利用信号模型计算不同方位和角度下的焦平面，根据焦平面构建数据集，用于样本训练；接着，通过MLFFNN模型对数据集进行训练，获取波束合成的自适应参数；最后，利用自适应参数实现相控阵馈源波束合成。对比现有自适应波束合成方法，本文提出方法计算复杂度低、波束指向精度高。仿真实验验证了提出方法的有效性。

cn (code pays fourni par la source)

MegaTTS 3: Sparse Alignment Enhanced Latent Diffusion Transformer for Zero-Shot Speech Synthesis

Ziyue Karen Jiang, Yi Ren, Ruiqi Li, Shengpeng Ji et autres

While recent zero-shot text-to-speech (TTS) models have significantly improved speech quality and expressiveness, mainstream systems still suffer from issues related to speech-text alignment modeling: 1) models without explicit speech-text alignment modeling exhibit less robustness, especially for hard sentences in practical applications; 2) …

HumanDiT: Pose-Guided Diffusion Transformer for Long-form Human Motion Video Generation

Qijun Gan, Yi Ren, Chen Guang Zhang, Zhenhui Ye et autres

Human motion video generation has advanced significantly, while existing methods still struggle with accurately rendering detailed body parts like hands and faces, especially in long sequences and intricate motions. Current approaches also rely on fixed resolution and struggle to maintain visual consistency. …

MetaFruit meets foundation models: Leveraging a comprehensive multi-fruit dataset for advancing agricultural foundation models

Jiajia Li, Kyle Lammers, Xunyuan Yin, Xiang Yin et autres

us, cn (code pays fourni par la source)

Contestable AI Needs Computational Argumentation

Francesco Leofante, Hamed Ayoobi, Adam Dejl, Gabriel P. Freedman et autres

AI has become pervasive in recent years, but state-of-the-art approaches predominantly neglect the need for AI systems to be contestable. Instead, contestability is advocated by AI guidelines (e.g. by the OECD) and regulation of automated decision-making (e.g. GDPR). In this position paper …

gb (code pays fourni par la source)

CE-QArg: Counterfactual Explanations for Quantitative Bipolar Argumentation Frameworks

Xiang Yin, Nico Potyka, Francesca Toni

There is a growing interest in understanding arguments' strength in Quantitative Bipolar Argumentation Frameworks (QBAFs). Most existing studies focus on attribution-based methods that explain an argument's strength by assigning importance scores to other arguments but fail to explain how to change the …

gb (code pays fourni par la source)

SPARC: Prediction-Based Safe Control for Coupled Controllable and Uncontrollable Agents with Conformal Predictions

Shuqi Wang, Shaoyuan Li, Xiang Yin

We investigate the problem of safe control synthesis for systems operating in environments with uncontrollable agents whose dynamics are unknown but coupled with those of the controlled system. This scenario naturally arises in various applications, such as autonomous driving and human-robot collaboration, …