arXiv subjects

Yupei Li

Publications and source records attributed to Yupei Li.

33 records · Page 2Linked to original sources

arXiv2026

M6: Multi-generator, Multi-domain, Multi-lingual and cultural, Multi-genres, Multi-instrument Machine-Generated Music Detection Databases

Machine-generated music (MGM) has emerged as a powerful tool with applications in music therapy, personalised editing, and creative inspiration for the music community. However, its unregulated use threatens the entertainment, education, and arts sectors by diminishing the value of high-quality human compositions. Detecting machine-generated music (MGMD) is, therefore, critical to safeguarding these domains, yet the field lacks comprehensive datasets to support meaningful progress. To address this gap, we introduce \textbf{M6}, a large-scale benchmark dataset tailored for MGMD research. M6 is distinguished by its diversity, encompassing multiple generators, domains, languages, cultural contexts, genres, and instruments. We outline our methodology for data selection and collection, accompanied by detailed data analysis, providing all WAV form of music. Additionally, we provide baseline performance scores using foundational binary classification models, illustrating the complexity of MGMD and the significant room for improvement. By offering a robust and multifaceted resource, we aim to empower future research to develop more effective detection methods for MGM. We believe M6 will serve as a critical step toward addressing this societal challenge. The dataset and code will be freely available to support open collaboration and innovation in this field.

cs.SD↗

arXiv2026

Affect and Effect: Limitations of regularisation-based continual learning in EEG-based emotion classification

Generalisation to unseen subjects in EEG-based emotion classification remains a challenge due to high inter-and intra-subject variability. Continual learning (CL) poses a promising solution by learning from a sequence of tasks while mitigating catastrophic forgetting. Regularisation-based CL approaches, such as Elastic Weight Consolidation (EWC), Synaptic Intelligence (SI), and Memory Aware Synapses (MAS), are commonly used as baselines in EEG-based CL studies, yet their suitability for this problem remains underexplored. This study theoretically and empirically finds that regularisation-based CL methods show limited performance for EEG-based emotion classification on the DREAMER and SEED datasets. We identify a fundamental misalignment in the stability-plasticity trade-off, where regularisation-based methods prioritise mitigating catastrophic forgetting (backward transfer) over adapting to new subjects (forward transfer). We investigate this limitation under subject-incremental sequences and observe that: (1) the heuristics for estimating parameter importance become less reliable under noisy data and covariate shift, (2) gradients on parameters deemed important by these heuristics often interfere with gradient updates required for new subjects, moving optimisation away from the minimum, (3) importance values accumulated across tasks over-constrain the model, and (4) performance is sensitive to subject order. Forward transfer showed no statistically significant improvement over sequential fine-tuning (p > 0.05 across approaches and datasets). The high variability of EEG signals means past subjects provide limited value to future subjects. Regularisation-based continual learning approaches are therefore limited for robust generalisation to unseen subjects in EEG-based emotion classification.

cs.LG↗

arXiv2025

DFALLM: Achieving Generalizable Multitask Deepfake Detection by Optimizing Audio LLM Components

Audio deepfake detection has recently garnered public concern due to its implications for security and reliability. Traditional deep learning methods have been widely applied to this task but often lack generalisability when confronted with newly emerging spoofing techniques and more tasks such as spoof attribution recognition rather than simple binary classification. In principle, Large Language Models (LLMs) are considered to possess the needed generalisation capabilities. However, previous research on Audio LLMs (ALLMs) indicates a generalization bottleneck in audio deepfake detection performance, even when sufficient data is available. Consequently, this study investigates the model architecture and examines the effects of the primary components of ALLMs, namely the audio encoder and the text-based LLM. Our experiments demonstrate that the careful selection and combination of audio encoders and text-based LLMs are crucial for unlocking the deepfake detection potential of ALLMs. We further propose an ALLM structure capable of generalizing deepfake detection abilities to out-of-domain spoofing tests and other deepfake tasks, such as spoof positioning and spoof attribution recognition. Our proposed model architecture achieves state-of-the-art (SOTA) performance across multiple datasets, including ASVSpoof2019, InTheWild, and Demopage, with accuracy reaching up to 95.76% on average, and exhibits competitive capabilities in other deepfake detection tasks such as attribution, and localisation compared to SOTA audio understanding models. Data and codes are provided in supplementary materials.

cs.SD↗

arXiv2025

Ore's Theorem for rainbow Hamiltonian-connected graphs

Let $G = (G_1, G_2, \ldots, G_m)$ be a collection of $m$ graphs on a common vertex set $V$. For a graph $H$ with vertices in $V$, we say that $G$ contains a rainbow $H$ if there is an injection $c: E(H) \to [m]$ such that for every edge $e \in E(H)$, we have $e \in E(G_{c(e)})$. In this paper, we show that if $G = (G_1, \ldots, G_n)$ is a collection of graphs on $n$ vertices such that for every $i \in [n]$, $d_{G_i}(u) + d_{G_i}(v) \geq n$ whenever $uv \notin E(G_i)$, then either $G$ contains rainbow Hamiltonian paths between every pair of vertices, or $G$ contains a rainbow Hamiltonian cycle. Moreover, we prove a stronger version in which we may also embed prescribed rainbow linear forests into the Hamiltonian paths.

math.CO↗

arXiv2025

Discourse Features Enhance Detection of Document-Level Machine-Generated Content

The availability of high-quality APIs for Large Language Models (LLMs) has facilitated the widespread creation of Machine-Generated Content (MGC), posing challenges such as academic plagiarism and the spread of misinformation. Existing MGC detectors often focus solely on surface-level information, overlooking implicit and structural features. This makes them susceptible to deception by surface-level sentence patterns, particularly for longer texts and in texts that have been subsequently paraphrased. To overcome these challenges, we introduce novel methodologies and datasets. Besides the publicly available dataset Plagbench, we developed the paraphrased Long-Form Question and Answer (paraLFQA) and paraphrased Writing Prompts (paraWP) datasets using GPT and DIPPER, a discourse paraphrasing tool, by extending artifacts from their original versions. To better capture the structure of longer texts at document level, we propose DTransformer, a model that integrates discourse analysis through PDTB preprocessing to encode structural features. It results in substantial performance gains across both datasets - 15.5% absolute improvement on paraLFQA, 4% absolute improvement on paraWP, and 1.5% absolute improvemene on M4 compared to SOTA approaches. The data and code are available at: https://github.com/myxp-lyp/Discourse-Features-Enhance-Detection-of-Document-Level-Machine-Generated-Content.git.

cs.CL↗

arXiv2025

Artificial Emotion: A Survey of Theories and Debates on Realising Emotion in Artificial Intelligence

Affective Computing (AC) has enabled Artificial Intelligence (AI) systems to recognise, interpret, and respond to human emotions - a capability also known as Artificial Emotional Intelligence (AEI). It is increasingly seen as an important component of Artificial General Intelligence (AGI). We discuss whether in order to peruse this goal, AI benefits from moving beyond emotion recognition and synthesis to develop internal emotion-like states, which we term as Artificial Emotion (AE). This shift potentially allows AI to benefit from the paradigm of `inner emotions' in ways we - as humans - do. Although recent research shows early signs that AI systems may exhibit AE-like behaviours, a clear framework for how emotions can be realised in AI remains underexplored. In this paper, we discuss potential advantages of AE in AI, review current manifestations of AE in machine learning systems, examine emotion-modulated architectures, and summarise mechanisms for modelling and integrating AE into future AI. We also explore the ethical implications and safety risks associated with `emotional' AGI, while concluding with our opinion on how AE could be beneficial in the future.

cs.HC↗

arXiv2025

Automatic Contouring of Spinal Vertebrae on X-Ray using a Novel Sandwich U-Net Architecture

In spinal vertebral mobility disease, accurately extracting and contouring vertebrae is essential for assessing mobility impairments and monitoring variations during flexion-extension movements. Precise vertebral contouring plays a crucial role in surgical planning; however, this process is traditionally performed manually by radiologists or surgeons, making it labour-intensive, time-consuming, and prone to human error. In particular, mobility disease analysis requires the individual contouring of each vertebra, which is both tedious and susceptible to inconsistencies. Automated methods provide a more efficient alternative, enabling vertebra identification, segmentation, and contouring with greater accuracy and reduced time consumption. In this study, we propose a novel U-Net variation designed to accurately segment thoracic vertebrae from anteroposterior view on X-Ray images. Our proposed approach, incorporating a ``sandwich" U-Net structure with dual activation functions, achieves a 4.1\% improvement in Dice score compared to the baseline U-Net model, enhancing segmentation accuracy while ensuring reliable vertebral contour extraction.

eess.IV↗

arXiv2025

Large Language Models for Depression Recognition in Spoken Language Integrating Psychological Knowledge

Depression is a growing concern gaining attention in both public discourse and AI research. While deep neural networks (DNNs) have been used for recognition, they still lack real-world effectiveness. Large language models (LLMs) show strong potential but require domain-specific fine-tuning and struggle with non-textual cues. Since depression is often expressed through vocal tone and behaviour rather than explicit text, relying on language alone is insufficient. Diagnostic accuracy also suffers without incorporating psychological expertise. To address these limitations, we present, to the best of our knowledge, the first application of LLMs to multimodal depression detection using the DAIC-WOZ dataset. We extract the audio features using the pre-trained model Wav2Vec, and mapped it to text-based LLMs for further processing. We also propose a novel strategy for incorporating psychological knowledge into LLMs to enhance diagnostic performance, specifically using a question and answer set to grant authorised knowledge to LLMs. Our approach yields a notable improvement in both Mean Absolute Error (MAE) and Root Mean Square Error (RMSE) compared to a base score proposed by the related original paper. The codes are available at https://github.com/myxp-lyp/Depression-detection.git

cs.HC↗

arXiv2025

An Ore-type theorem for $[3]$-graphs

Ore's Theorem states that if $G$ is an $n$-vertex graph and every pair of non-adjacent vertices has degree sum at least $n$, then $G$ is Hamiltonian. A $[3]$-graph is a hypergraph in which every edge contains at most $3$ vertices. In this paper, we prove an Ore-type result on the existence of Hamiltonian Berge cycles in $[3]$-graph $\cH$, based on the degree sum of every pair of non-adjacent vertices in the $2$-shadow graph $\partial \cH$ of $\cH$. Namely, we prove that there exists a constant $d_0$ such that for all $n \geq 6$, if a $[3]$-graph $\cH$ on $n$ vertices satisfies that every pair $u,v \in V(\cH)$ of non-adjacent vertices has degree sum $d_{\partial \cH}(u) + d_{\partial \cH}(v) \geq n+d_0$, then $\cH$ contains a Hamiltonian Berge cycle. Moreover, we conjecture that $d_0=1$ suffices.

math.CO↗

arXiv2025

Pancyclicity in hypergraphs with large uniformity

A Berge cycle of length $\ell$ in a hypergraph $\mathcal{H}$ is a sequence of alternating vertices and edges $v_0e_0v_1e_1...v_\ell e_\ell v_0$ such that $\{v_i,v_{i+1}\}\subseteq e_i$ for all $i$, with indices taken modulo $\ell$. For $n$ sufficiently large and $r\geq \lfloor\frac{n-1}{2}\rfloor-1$ we prove exact minimum degree conditions for an $n$-vertex, $r$-uniform hypergraph to contain Berge cycles of every length between $2$ and $n$. In conjunction with previous work, this provides sharp Dirac-type conditions for pancyclicity in $r$-uniform hypergraphs for all $3\leq r\leq n$ when $n$ is sufficiently large.

math.CO↗

arXiv2025

Neuroplasticity in Artificial Intelligence -- An Overview and Inspirations on Drop In & Out Learning

Artificial Intelligence (AI) has achieved new levels of performance and spread in public usage with the rise of deep neural networks (DNNs). Initially inspired by human neurons and their connections, NNs have become the foundation of AI models for many advanced architectures. However, some of the most integral processes in the human brain, particularly neurogenesis and neuroplasticity in addition to the more spread neuroapoptosis have largely been ignored in DNN architecture design. Instead, contemporary AI development predominantly focuses on constructing advanced frameworks, such as large language models, which retain a static structure of neural connections during training and inference. In this light, we explore how neurogenesis, neuroapoptosis, and neuroplasticity can inspire future AI advances. Specifically, we examine analogous activities in artificial NNs, introducing the concepts of ``dropin'' for neurogenesis and revisiting ``dropout'' and structural pruning for neuroapoptosis. We additionally suggest neuroplasticity combining the two for future large NNs in ``life-long learning'' settings following the biological inspiration. We conclude by advocating for greater research efforts in this interdisciplinary domain and identifying promising directions for future exploration.

cs.AI↗

arXiv2025

GatedxLSTM: A Multimodal Affective Computing Approach for Emotion Recognition in Conversations

Affective Computing (AC) is essential for advancing Artificial General Intelligence (AGI), with emotion recognition serving as a key component. However, human emotions are inherently dynamic, influenced not only by an individual's expressions but also by interactions with others, and single-modality approaches often fail to capture their full dynamics. Multimodal Emotion Recognition (MER) leverages multiple signals but traditionally relies on utterance-level analysis, overlooking the dynamic nature of emotions in conversations. Emotion Recognition in Conversation (ERC) addresses this limitation, yet existing methods struggle to align multimodal features and explain why emotions evolve within dialogues. To bridge this gap, we propose GatedxLSTM, a novel speech-text multimodal ERC model that explicitly considers voice and transcripts of both the speaker and their conversational partner(s) to identify the most influential sentences driving emotional shifts. By integrating Contrastive Language-Audio Pretraining (CLAP) for improved cross-modal alignment and employing a gating mechanism to emphasise emotionally impactful utterances, GatedxLSTM enhances both interpretability and performance. Additionally, the Dialogical Emotion Decoder (DED) refines emotion predictions by modelling contextual dependencies. Experiments on the IEMOCAP dataset demonstrate that GatedxLSTM achieves state-of-the-art (SOTA) performance among open-source methods in four-class emotion classification. These results validate its effectiveness for ERC applications and provide an interpretability analysis from a psychological perspective.

cs.CL↗

arXiv2024

Towards Friendly AI: A Comprehensive Review and New Perspectives on Human-AI Alignment

As Artificial Intelligence (AI) continues to advance rapidly, Friendly AI (FAI) has been proposed to advocate for more equitable and fair development of AI. Despite its importance, there is a lack of comprehensive reviews examining FAI from an ethical perspective, as well as limited discussion on its potential applications and future directions. This paper addresses these gaps by providing a thorough review of FAI, focusing on theoretical perspectives both for and against its development, and presenting a formal definition in a clear and accessible format. Key applications are discussed from the perspectives of eXplainable AI (XAI), privacy, fairness and affective computing (AC). Additionally, the paper identifies challenges in current technological advancements and explores future research avenues. The findings emphasise the significance of developing FAI and advocate for its continued advancement to ensure ethical and beneficial AI development.

cs.AI↗

arXiv2024

Ricci Curvature Formula: Applications to Bonnet-Myers Sharp Irregular Graphs

In this paper, we establish a simple formula for computing the Lin-Lu-Yau Ricci curvature on graphs. For any edge $xy$ in a simple locally finite graph $G$, the curvature $κ(x,y)$ can be expressed as a cost function of an optimal bijection between two blow-up sets of the neighbors of $x$ and $y$. Utilizing this approach, we derive several results including a structural theorem for the Bonnet-Myers sharp irregular graphs of diameter $3$ and a theorem on $C_3$-free Bonnet-Myers sharp graphs.

math.CO↗

arXiv2024

Berge Pancyclic hypergraphs

A Berge cycle of length $\ell$ in a hypergraph is an alternating sequence of $\ell$ distinct vertices and $\ell$ distinct edges $v_1,e_1,v_2, \ldots, v_\ell, e_{\ell}$ such that $\{v_i, v_{i+1}\} \subseteq e_i$ for all $i$, with indices taken modulo $\ell$. We call an $n$-vertex hypergraph pancyclic if it contains Berge cycles of every length from $3$ to $n$. We prove a sharp Dirac-type result guaranteeing pancyclicity in uniform hypergraphs: for $n \geq 70$, $3 \leq r \leq \lfloor (n-1)/2\rfloor - 2$, if $\cH$ is an $n$-vertex, $r$-uniform hypergraph with minimum degree at least ${\lfloor (n-1)/2 \rfloor \choose r-1} + 1$, then $\cH$ is pancyclic.

math.CO↗