arXiv subjects

Chang Lei

Publications and source records attributed to Chang Lei.

8 recordsLinked to original sources

arXiv2025

Towards Paradigm-General Suicide Risk Detection via Speech LLM

Suicide risk among adolescents remains a critical public health concern, and speech provides a non-invasive and scalable approach for its detection. Speech-based suicide risk assessment commonly relies on carefully designed speech elicitation paradigms (\textit{e.g.,} verbal fluency, reading, or question answering) to probe cognitive and affective states. Existing approaches, however, typically focus on one single paradigm at a time. This paper, for the first time, investigates cross-paradigm approaches that unify diverse speech elicitation paradigms within a single model. Specifically, we use a speech LLM as backbone with a mixture of DoRA experts (MoDE) to capture complementary cues across assessments dynamically, tested on 1,223 participants across ten speech elicitation paradigms. Results show that MoDE outperforms both paradigm-specific and conventional joint-learning models. Moreover, it can generalise to unseen paradigms and provide better confidence calibration.

eess.AS↗

arXiv2025

Speaker Anonymisation for Speech-based Suicide Risk Detection

Adolescent suicide is a critical global health issue, and speech provides a cost-effective modality for automatic suicide risk detection. Given the vulnerable population, protecting speaker identity is particularly important, as speech itself can reveal personally identifiable information if the data is leaked or maliciously exploited. This work presents the first systematic study of speaker anonymisation for speech-based suicide risk detection. A broad range of anonymisation methods are investigated, including techniques based on traditional signal processing, neural voice conversion, and speech synthesis. A comprehensive evaluation framework is built to assess the trade-off between protecting speaker identity and preserving information essential for suicide risk detection. Results show that combining anonymisation methods that retain complementary information yields detection performance comparable to that of original speech, while achieving protection of speaker identity for vulnerable populations.

eess.AS↗

arXiv2025

M2-brane indices on Higgs vacua and black holes

As an exact count of protected states, the superconformal index provides a powerful probe into holography and quantum aspects of gravity, reproducing the Bekenstein--Hawking entropy of supersymmetric AdS black holes in the large-$N$ limit. As a step toward understanding quantum black hole microstates, we study the finite-$N$ index of the 3d ADHM quiver gauge theory, a UV description of the 3d $\mathcal N=8$ SCFT dual to M-theory on AdS$_4 \times S^7$. In this note, we analyze both microcanonical and canonical features of the superconformal index. By computing the index to sufficiently high orders using the factorization formula, we identify signatures of quantum black hole states in the finite-$N$ spectrum of the ADHM quiver, which align with the leading large-$N$ contribution reflecting the holographic dual black hole entropy. Furthermore, we introduce the complex-$\beta$ phase diagram of the index, which exhibits distinct peaks potentially associated with different gravitational saddles. We also examine the Hilbert series limit of the factorized index. Our results demonstrate that the finite-$N$ index encodes rich information about black hole microstates and their quantum gravitational interpretation.

hep-th↗

arXiv2025

The 1st SpeechWellness Challenge: Detecting Suicide Risk Among Adolescents

The 1st SpeechWellness Challenge (SW1) aims to advance methods for detecting current suicide risk in adolescents using speech analysis techniques. Suicide among adolescents is a critical public health issue globally. Early detection of suicidal tendencies can lead to timely intervention and potentially save lives. Traditional methods of assessment often rely on self-reporting or clinical interviews, which may not always be accessible. The SW1 challenge addresses this gap by exploring speech as a non-invasive and readily available indicator of mental health. We release the SW1 dataset which contains speech recordings from 600 adolescents aged 10-18 years. By focusing on speech generated from natural tasks, the challenge seeks to uncover patterns and markers that correlate with current suicide risk.

eess.AS↗

arXiv2024

AlphaDou: High-Performance End-to-End Doudizhu AI Integrating Bidding

Artificial intelligence for card games has long been a popular topic in AI research. In recent years, complex card games like Mahjong and Texas Hold'em have been solved, with corresponding AI programs reaching the level of human experts. However, the game of Doudizhu presents significant challenges due to its vast state/action space and unique characteristics involving reasoning about competition and cooperation, making the game extremely difficult to solve.The RL model Douzero, trained using the Deep Monte Carlo algorithm framework, has shown excellent performance in Doudizhu. However, there are differences between its simplified game environment and the actual Doudizhu environment, and its performance is still a considerable distance from that of human experts. This paper modifies the Deep Monte Carlo algorithm framework by using reinforcement learning to obtain a neural network that simultaneously estimates win rates and expectations. The action space is pruned using expectations, and strategies are generated based on win rates. The modified algorithm enables the AI to perform the full range of tasks in the Doudizhu game, including bidding and cardplay. The model was trained in a actual Doudizhu environment and achieved state-of-the-art performance among publicly available models. We hope that this new framework will provide valuable insights for AI development in other bidding-based games.

cs.AI↗

arXiv2024

Spontaneous Speech-Based Suicide Risk Detection Using Whisper and Large Language Models

The early detection of suicide risk is important since it enables the intervention to prevent potential suicide attempts. This paper studies the automatic detection of suicide risk based on spontaneous speech from adolescents, and collects a Mandarin dataset with 15 hours of suicide speech from more than a thousand adolescents aged from ten to eighteen for our experiments. To leverage the diverse acoustic and linguistic features embedded in spontaneous speech, both the Whisper speech model and textual large language models (LLMs) are used for suicide risk detection. Both all-parameter finetuning and parameter-efficient finetuning approaches are used to adapt the pre-trained models for suicide risk detection, and multiple audio-text fusion approaches are evaluated to combine the representations of Whisper and the LLM. The proposed system achieves a detection accuracy of 0.807 and an F1-score of 0.846 on the test set with 119 subjects, indicating promising potential for real suicide risk detection applications.

cs.CL↗

arXiv2024

Non-Gaussianity from Schwinger-Keldysh Effective Field Theory

We present a systematic treatment of non-Gaussianity in stochastic systems using the Schwinger-Keldysh effective field theory framework, in which the non-Gaussianity is realized as nonlinear terms in the fluctuation field. We establish two stochastic formulations of the Schwinger-Keldysh effective field theory, with those nonlinear terms manifested as multiple non-Gaussian noises in the Langevin equation and as higher order diffusive terms in the Fokker-Planck equation. The equivalence of the stochastic formulations with the original Schwinger-Keldysh effective field theory is demonstrated with non-trivial examples for arbitrary non-Gaussian parameters. The stochastic formulations will be more flexible and effective in studying non-equilibrium dynamics. We also reveal an ambiguity when coarse-graining time scale and non-Gaussian parameters vanish simultaneously, which may be responsible for the unphysical divergence found in perturbative analysis.

hep-th↗

arXiv2023

Phases of Quench Dynamics in the Presence of Fluctuation

We study the effect of thermal fluctuations on a sourced quench in a system with $Z_2$ symmetry. By ignoring the fluctuation of finite momentum modes and tracing the dynamics of zero momentum mode driven by a spatially homogeneous source near the critical point, we map out a phase diagram for the quench dynamics. The phase diagram consists of three different phases. Phase I occurs for large fluctuations with the relaxation time set by fluctuation induced inverse effective mass square. Phase II occurs for small fluctuation and slow quench rate. The dynamics is characterized by a modified Kibble-Zurek scaling. Phase III corresponds to small fluctuations and rapid quench rate. The relaxation time tends to a finite value in the rapid quench limit. We also estimate the fluctuations of finite momentum modes, finding significant enhancement of the effective mass square. We speculate the qualitative features of the phase diagram may remain the same, but enhanced fluctuations can lead to shrinkage of phase II and III.

nucl-th↗