arXiv subjects

Yukun Liu

Publications and source records attributed to Yukun Liu.

At least 37 records · Page 2Linked to original sources

arXiv2025

Instability of inverse probability weighting methods and a remedy for non-ignorable missing data

Inverse probability weighting (IPW) methods are commonly used to analyze non-ignorable missing data under the assumption of a logistic model for the missingness probability. However, solving IPW equations numerically may involve non-convergence problems when the sample size is moderate and the missingness probability is high. Moreover, those equations often have multiple roots, and identifying the best root is challenging. Therefore, IPW methods may have low efficiency or even produce biased results. We identify the pitfall in these methods pathologically: they involve the estimation of a moment-generating function, and such functions are notoriously unstable in general. As a remedy, we model the outcome distribution given the covariates of the completely observed individuals semiparametrically. After forming an induced logistic regression model for the missingness status of the outcome and covariate, we develop a maximum conditional likelihood method to estimate the underlying parameters. The proposed method circumvents the estimation of a moment-generating function and hence overcomes the instability of IPW methods. Our theoretical and simulation results show that the proposed method outperforms existing competitors greatly. Two real data examples are analyzed to illustrate the advantages of our method. We conclude that if only a parametric logistic regression is assumed but the outcome regression model is left arbitrary, then one has to be cautious in using any of the existing statistical methods in problems involving non-ignorable missing data.

stat.ME↗

arXiv2025

Maximum likelihood abundance estimation from capture-recapture data when covariates are missing at random

In capture-recapture experiments, individual covariates may be subject to missing, especially when the number of times of being captured is small. When the covariate information is missing at random, the inverse probability weighting method and multiple imputation method are widely used to obtain the point estimators of the abundance. These point estimators are then used to construct the Wald-type confidence intervals for the abundance. However, such intervals may have severely inaccurate coverage probabilities and their lower limits can be even less than the number of individuals ever captured. In this paper, we proposed a maximum empirical likelihood estimation approach for the abundance in presence of missing covariates. We show that the maximum empirical likelihood estimator is asymptotically normal, and that the empirical likelihood ratio statistic for abundance has a chisquare limiting distribution with one degree of freedom. Simulations indicate that the proposed estimator has smaller mean square error than the existing estimators, and the proposed empirical likelihood ratio confidence interval usually has more accurate coverage probabilities than the existing Wald-type confidence intervals. We illustrate the proposed method by analyzing the bird species yellow-bellied prinia collected in Hong Kong.

stat.ME↗

arXiv2025

Empirical likelihood meta analysis with publication bias correction under Copas-like selection model

Meta analysis is commonly-used to synthesize multiple results from individual studies. However, its validation is usually threatened by publication bias and between-study heterogeneity, which can be captured by the Copas selection model. Existing inference methods under this model are all based on conditional likelihood and may not be fully efficient. In this paper, we propose a full likelihood approach to meta analysis by integrating the conditional likelihood and a marginal semi-parametric empirical likelihood under a Copas-like selection model. We show that the maximum likelihood estimators (MLE) of all the underlying parameters have a jointly normal limiting distribution, and the full likelihood ratio follows an asymptotic central chisquare distribution. Our simulation results indicate that compared with the conditional likelihood method, the proposed MLEs have smaller mean squared errors and the full likelihood ratio confidence intervals have more accurate coverage probabilities. A real data example is analyzed to show the advantages of the full likelihood method over the conditional likelihood method.

stat.ME↗

arXiv2025

On consistency of the MLE under finite mixtures of location-scale distributions with a structural parameter

We provide a general and rigorous proof for the strong consistency of maximum likelihood estimators of the cumulative distribution function of the mixing distribution and structural parameter under finite mixtures of location-scale distributions with a structural parameter. The consistency results do not require the parameter space of location and scale to be compact. We illustrate the results by applying them to finite mixtures of location-scale distributions with the component density function being one of the commonly used density functions: normal, logistic, extreme-value, or $t$. An extension of the strong consistency results to finite mixtures of multivariate elliptical distributions is also discussed.

math.ST↗

arXiv2025

Composite empirical likelihood for multisample clustered data

In many applications, data cluster. Failing to take the cluster structure into consideration generally leads to underestimated variances of point estimators and inflated type I errors in hypothesis tests. Many circumstance-dependent approaches have been developed to handle clustered data. A working covariance matrix may be used in generalized estimating equations. One may throw out the cluster structure and use only the cluster means, or explicitly model the cluster structure. Our interest is the case where multiple samples of clustered data are collected, and the population quantiles are particularly important. We develop a composite empirical likelihood for clustered data under a density ratio model. This approach avoids parametric assumptions on the population distributions or the cluster structure. It efficiently utilizes the common features of the multiple populations and the exchangeability of the cluster members. We also develop a cluster-based bootstrap method to provide valid variance estimation and to control the type I errors. We examine the performance of the proposed method through simulation experiments and illustrate its usage via a real-world example.

stat.ME↗

arXiv2025

Hypothesis testing for quantitative trait locus effects in both location and scale in genetic backcross studies

Testing the existence of a quantitative trait locus (QTL) effect is an important task in QTL mapping studies. Most studies concentrate on the case where the phenotype distributions of different QTL groups follow normal distributions with the same unknown variance. In this paper we make a more general assumption that the phenotype distributions come from a location-scale distribution family. We derive the limiting distribution of the LRT for the existence of the QTL effect in both location and scale in genetic backcross studies. We further identify an explicit representation for this limiting distribution. As a complement, we study the limiting distribution of the LRT and its explicit representation for the existence of the QTL effect in the location only. The asymptotic properties of the LRTs under a local alternative are also investigated. Simulation studies are used to evaluate the asymptotic results, and a real-data example is included for illustration.

stat.ME↗

arXiv2025

Semiparametric empirical likelihood inference for abundance from one-inflated capture-recapture data

Abundance estimation from capture-recapture data is of great importance in many disciplines. Analysis of capture-recapture data is often complicated by the existence of one-inflation and heterogeneity problems. Simultaneously taking these issues into account, existing abundance estimation methods are usually constructed on the basis of conditional likelihood (CL) under one-inflated zero-truncated count models. However, the resulting Horvitz-Thompson-type estimators may be unstable, and the resulting Wald-type confidence intervals may exhibit severe undercoverage. In this paper, we propose a semiparametric empirical likelihood (EL) approach to abundance estimation under one-inflated binomial and Poisson regression models. We show that the maximum EL estimator for the abundance follows an asymptotically normal distribution and that the EL ratio statistic of abundance follows a limiting chi-square distribution with one degree of freedom. To facilitate computation of the EL method, we develop an expectation-maximization (EM) algorithm, and establish its appealing convergence property. We also propose a new score test for the existence of one-inflation and prove its asymptotic normality. Our simulation studies indicate that compared with CL-based methods, the maximum EL estimator has a smaller mean square error, the EL ratio confidence interval has a remarkable gain in coverage probability, and the proposed score test is more powerful. The advantages of the proposed approaches are further demonstrated by analyses of prinia data from Hong Kong and drug user data from Bangkok.

stat.ME↗

arXiv2025

Two-step semiparametric empirical likelihood inference from capture-recapture data with missing covariates

Missing covariates are not uncommon in capture-recapture studies. When covariate information is missing at random in capture-recapture data, an empirical full likelihood method has been demonstrated to outperform conditional-likelihood-based methods in abundance estimation. However, the fully observed covariates must be discrete, and the method is not directly applicable to continuous-time capture-recapture data. Based on the Binomial and Poisson regression models, we propose a two-step semiparametric empirical likelihood approach for abundance estimation in the presence of missing covariates, regardless of whether the fully observed covariates are discrete or continuous. We show that the maximum semiparametric empirical likelihood estimators for the underlying parameters and the abundance are asymptotically normal, and more efficient than the counterpart for a completely known non-missingness probability. After scaling, the empirical likelihood ratio test statistic for abundance follows a limiting chi-square distribution with one degree of freedom. The proposed approach is further extended to one-inflated count regression models, and a score-like test is constructed to assess whether one-inflation exists among the number of captures. Our simulation shows that, compared with the previous method, the proposed method not only performs better in correcting bias, but also has a more accurate coverage in the presence of fully observed continuous covariates, although there may be a slight efficiency loss when the fully observed covariates are only discrete. The performance of the new method is illustrated by an analysis of the Hong Kong prinia data.

stat.ME↗

arXiv2025

Debunk and Infer: Multimodal Fake News Detection via Diffusion-Generated Evidence and LLM Reasoning

The rapid spread of fake news across multimedia platforms presents serious challenges to information credibility. In this paper, we propose a Debunk-and-Infer framework for Fake News Detection(DIFND) that leverages debunking knowledge to enhance both the performance and interpretability of fake news detection. DIFND integrates the generative strength of conditional diffusion models with the collaborative reasoning capabilities of multimodal large language models (MLLMs). Specifically, debunk diffusion is employed to generate refuting or authenticating evidence based on the multimodal content of news videos, enriching the evaluation process with diverse yet semantically aligned synthetic samples. To improve inference, we propose a chain-of-debunk strategy where a multi-agent MLLM system produces logic-grounded, multimodal-aware reasoning content and final veracity judgment. By jointly modeling multimodal features, generative debunking cues, and reasoning-rich verification within a unified architecture, DIFND achieves notable improvements in detection accuracy. Extensive experiments on the FakeSV and FVC datasets show that DIFND not only outperforms existing approaches but also delivers trustworthy decisions.

cs.CL↗

arXiv2025

Exploring Modality Disruption in Multimodal Fake News Detection

The rapid growth of social media has led to the widespread dissemination of fake news across multiple content forms, including text, images, audio, and video. Compared to unimodal fake news detection, multimodal fake news detection benefits from the increased availability of information across multiple modalities. However, in the context of social media, certain modalities in multimodal fake news detection tasks may contain disruptive or over-expressive information. These elements often include exaggerated or embellished content. We define this phenomenon as modality disruption and explore its impact on detection models through experiments. To address the issue of modality disruption in a targeted manner, we propose a multimodal fake news detection framework, FND-MoE. Additionally, we design a two-pass feature selection mechanism to further mitigate the impact of modality disruption. Extensive experiments on the FakeSV and FVC-2018 datasets demonstrate that FND-MoE significantly outperforms state-of-the-art methods, with accuracy improvements of 3.45% and 3.71% on the respective datasets compared to baseline models.

cs.MM↗

arXiv2025

Deconfounded Reasoning for Multimodal Fake News Detection via Causal Intervention

The rapid growth of social media has led to the widespread dissemination of fake news across multiple content forms, including text, images, audio, and video. Traditional unimodal detection methods fall short in addressing complex cross-modal manipulations; as a result, multimodal fake news detection has emerged as a more effective solution. However, existing multimodal approaches, especially in the context of fake news detection on social media, often overlook the confounders hidden within complex cross-modal interactions, leading models to rely on spurious statistical correlations rather than genuine causal mechanisms. In this paper, we propose the Causal Intervention-based Multimodal Deconfounded Detection (CIMDD) framework, which systematically models three types of confounders via a unified Structural Causal Model (SCM): (1) Lexical Semantic Confounder (LSC); (2) Latent Visual Confounder (LVC); (3) Dynamic Cross-Modal Coupling Confounder (DCCC). To mitigate the influence of these confounders, we specifically design three causal modules based on backdoor adjustment, frontdoor adjustment, and cross-modal joint intervention to block spurious correlations from different perspectives and achieve causal disentanglement of representations for deconfounded reasoning. Experimental results on the FakeSV and FVC datasets demonstrate that CIMDD significantly improves detection accuracy, outperforming state-of-the-art methods by 4.27% and 4.80%, respectively. Furthermore, extensive experimental results indicate that CIMDD exhibits strong generalization and robustness across diverse multimodal scenarios.

cs.MM↗

arXiv2025

Revamping Conformal Selection With Optimal Power: A Neyman--Pearson Perspective

This paper introduces a novel conformal selection procedure, inspired by the Neyman--Pearson paradigm, to maximize the power of selecting qualified units while maintaining false discovery rate (FDR) control. Existing conformal selection methods may yield suboptimal power due to their reliance on conformal p-values, which are derived by substituting unobserved future outcomes with thresholds set by the null hypothesis. This substitution invalidates the exchangeability between imputed nonconformity scores for test data and those derived from calibration data, resulting in reduced power. In contrast, our approach circumvents the need for conformal p-values by constructing a likelihood-ratio-based decision rule that directly utilizes observed covariates from both calibration and test samples. The asymptotic optimality and FDR control of the proposed method are established under a correctly specified model, and modified selection procedures are introduced to improve power under model misspecification. The proposed methods are computationally efficient and can be readily extended to handle covariate shifts, making them well-suited for real-world applications. Simulation results show that these methods consistently achieve comparable or higher power than existing conformal p-value-based selection rules, particularly when the underlying distribution deviates from location-shift models, while effectively maintaining FDR control.

stat.ME↗

arXiv2025

Unsupervised optimal deep transfer learning for classification under general conditional shift

Classifiers trained solely on labeled source data may yield misleading results when applied to unlabeled target data drawn from a different distribution. Transfer learning can rectify this by transferring knowledge from source to target data, but its effectiveness frequently relies on stringent assumptions, such as label shift. In this paper, we introduce a novel General Conditional Shift (GCS) assumption, which encompasses label shift as a special scenario. Under GCS, we demonstrate that both the target distribution and the shift function are identifiable. To estimate the conditional probabilities ${\bmη}_P$ for source data, we propose leveraging deep neural networks (DNNs). Subsequent to transferring the DNN estimator, we estimate the target label distribution ${\bmπ}_Q$ utilizing a pseudo-maximum likelihood approach. Ultimately, by incorporating these estimates and circumventing the need to estimate the shift function, we construct our proposed Bayes classifier. We establish concentration bounds for our estimators of both ${\bmη}_P$ and ${\bmπ}_Q$ in terms of the intrinsic dimension of ${\bmη}_P$ . Notably, our DNN-based classifier achieves the optimal minimax rate, up to a logarithmic factor. A key advantage of our method is its capacity to effectively combat the curse of dimensionality when ${\bmη}_P$ exhibits a low-dimensional structure. Numerical simulations, along with an analysis of an Alzheimer's disease dataset, underscore its exceptional performance.

math.ST↗

arXiv2025

MTPareto: A MultiModal Targeted Pareto Framework for Fake News Detection

Multimodal fake news detection is essential for maintaining the authenticity of Internet multimedia information. Significant differences in form and content of multimodal information lead to intensified optimization conflicts, hindering effective model training as well as reducing the effectiveness of existing fusion methods for bimodal. To address this problem, we propose the MTPareto framework to optimize multimodal fusion, using a Targeted Pareto(TPareto) optimization algorithm for fusion-level-specific objective learning with a certain focus. Based on the designed hierarchical fusion network, the algorithm defines three fusion levels with corresponding losses and implements all-modal-oriented Pareto gradient integration for each. This approach accomplishes superior multimodal fusion by utilizing the information obtained from intermediate fusion to provide positive effects to the entire process. Experiment results on FakeSV and FVC datasets show that the proposed framework outperforms baselines and the TPareto optimization algorithm achieves 2.40% and 1.89% accuracy improvement respectively.

cs.LG↗

arXiv2024

The Codecfake Dataset and Countermeasures for the Universally Detection of Deepfake Audio

With the proliferation of Audio Language Model (ALM) based deepfake audio, there is an urgent need for generalized detection methods. ALM-based deepfake audio currently exhibits widespread, high deception, and type versatility, posing a significant challenge to current audio deepfake detection (ADD) models trained solely on vocoded data. To effectively detect ALM-based deepfake audio, we focus on the mechanism of the ALM-based audio generation method, the conversion from neural codec to waveform. We initially constructed the Codecfake dataset, an open-source, large-scale collection comprising over 1 million audio samples in both English and Chinese, focus on ALM-based audio detection. As countermeasure, to achieve universal detection of deepfake audio and tackle domain ascent bias issue of original sharpness aware minimization (SAM), we propose the CSAM strategy to learn a domain balanced and generalized minima. In our experiments, we first demonstrate that ADD model training with the Codecfake dataset can effectively detects ALM-based audio. Furthermore, our proposed generalization countermeasure yields the lowest average equal error rate (EER) of 0.616% across all test conditions compared to baseline models. The dataset and associated code are available online.

cs.SD↗

arXiv2024

Mel-Refine: A Plug-and-Play Approach to Refine Mel-Spectrogram in Audio Generation

Text-to-audio (TTA) model is capable of generating diverse audio from textual prompts. However, most mainstream TTA models, which predominantly rely on Mel-spectrograms, still face challenges in producing audio with rich content. The intricate details and texture required in Mel-spectrograms for such audio often surpass the models' capacity, leading to outputs that are blurred or lack coherence. In this paper, we begin by investigating the critical role of U-Net in Mel-spectrogram generation. Our analysis shows that in U-Net structure, high-frequency components in skip-connections and the backbone influence texture and detail, while low-frequency components in the backbone are critical for the diffusion denoising process. We further propose ``Mel-Refine'', a plug-and-play approach that enhances Mel-spectrogram texture and detail by adjusting different component weights during inference. Our method requires no additional training or fine-tuning and is fully compatible with any diffusion-based TTA architecture. Experimental results show that our approach boosts performance metrics of the latest TTA model Tango2 by 25\%, demonstrating its effectiveness.

cs.SD↗

arXiv2024

DPI-TTS: Directional Patch Interaction for Fast-Converging and Style Temporal Modeling in Text-to-Speech

In recent years, speech diffusion models have advanced rapidly. Alongside the widely used U-Net architecture, transformer-based models such as the Diffusion Transformer (DiT) have also gained attention. However, current DiT speech models treat Mel spectrograms as general images, which overlooks the specific acoustic properties of speech. To address these limitations, we propose a method called Directional Patch Interaction for Text-to-Speech (DPI-TTS), which builds on DiT and achieves fast training without compromising accuracy. Notably, DPI-TTS employs a low-to-high frequency, frame-by-frame progressive inference approach that aligns more closely with acoustic properties, enhancing the naturalness of the generated speech. Additionally, we introduce a fine-grained style temporal modeling method that further improves speaker style similarity. Experimental results demonstrate that our method increases the training speed by nearly 2 times and significantly outperforms the baseline models.

cs.SD↗

arXiv2024

Mixture of Experts Fusion for Fake Audio Detection Using Frozen wav2vec 2.0

Speech synthesis technology has posed a serious threat to speaker verification systems. Currently, the most effective fake audio detection methods utilize pretrained models, and integrating features from various layers of pretrained model further enhances detection performance. However, most of the previously proposed fusion methods require fine-tuning the pretrained models, resulting in excessively long training times and hindering model iteration when facing new speech synthesis technology. To address this issue, this paper proposes a feature fusion method based on the Mixture of Experts, which extracts and integrates features relevant to fake audio detection from layer features, guided by a gating network based on the last layer feature, while freezing the pretrained model. Experiments conducted on the ASVspoof2019 and ASVspoof2021 datasets demonstrate that the proposed method achieves competitive performance compared to those requiring fine-tuning.

cs.SD↗