arXiv subjects

Fan Li

Publications and source records attributed to Fan Li.

At least 109 records · Page 6Linked to original sources

arXiv2026

Generalized win fraction regression for composite survival endpoints

We propose a generalized win fraction regression framework for prioritized composite survival outcomes. The framework models the conditional win fraction through a chosen link function (including identity, logit, or probit), thereby accommodating multi-component time-to-event endpoints within a unified regression structure. To handle right censoring, we construct inverse-probability-of-censoring-weighted estimating equations that target the win fraction as if censoring were absent. Under the identity link, regression parameters characterize covariate associations on the natural win fraction scale. Under the logit link, they characterize the log odds of winning -- a new and complementary effect measure that treats ties as failures to win, imposing a more conservative standard than the win ratio or win odds. When there are no ties, the logit win fraction model reduces to proportional win fraction regression; moreover, the unweighted version of our estimating equations numerically coincides with the proportional win fraction point estimator regardless of ties. We establish large-sample properties of the proposed estimators and derive a consistent sandwich variance estimator that accounts for uncertainty from the estimated censoring weights. Extensive simulations examine finite-sample performance across link functions and censoring rates, and our method is illustrated through a reanalysis of the HF-ACTION clinical trial.

stat.ME↗

arXiv2026

Enhancing Foundation VLM Robustness to Missing Modality: Scalable Diffusion for Bi-directional Feature Restoration

Vision Language Model (VLM) typically assume complete modality input during inference. However, their effectiveness drops sharply when certain modalities are unavailable or incomplete. Current research on missing modality primarily faces two dilemmas: Prompt-based methods struggle to restore missing yet indispensable features and degrade the generalizability of VLM. Imputation-based approaches, lacking effective guidance, are prone to generating semantically irrelevant noise. Restoring precise semantics while sustaining VLM's generalization remains challenging. Therefore, we propose a general missing modality restoration strategy in this paper. We introduce an enhanced diffusion model as a pluggable mid-stage training module to effectively restore missing features. Our strategy introduces two key innovations: (I) Dynamic Modality Gating, which adaptively leverages conditional features to guide the generation of semantically consistent features; (II) Cross-Modal Mutual Learning mechanism, which bridges the semantic spaces of the dual models to achieve bi-directional alignment. Notably, our strategy maintains the original integrity of the pre-trained VLM, requiring no fine-tuning of the backbone models while significantly boosting resilience to information loss. Zero-shot evaluations across benchmark datasets demonstrate that our approach consistently outperforms existing baselines, establishing it as a robust and scalable extension that ensures VLM reliability across diverse missing rates and conditions. Our code and models will be publicly available.

cs.AI↗

arXiv2026

Two-stage Least Squares with Clustered Data under the Local Average Treatment Effect Framework

To estimate the causal effect of an endogenous treatment using clustered data, the canonical two-stage least squares (2sls) estimates a linear regression of the outcome on treatment status using an instrumental variable (IV) and conducts inference with cluster-robust standard errors. When both the treatment and the IV vary within clusters, an alternative two-stage least squares with fixed effects (2sfe) additionally includes cluster indicators in the regression, thereby incorporating cluster information into point estimation as well. This paper studies the trade-off between these approaches within the local average treatment effect (LATE) framework. When clusters are homogeneous, we show that both approaches yield valid large-sample inference for the LATE, and that 2sfe is more efficient than canonical 2sls only when the variation in cluster-specific effects dominates idiosyncratic variation and the IV has sufficient within-cluster variation. When clusters are heterogeneous, we show that 2sfe identifies a weighted average of cluster-specific LATEs, whereas the canonical 2sls generally does not. We further propose a test for detecting cluster heterogeneity.

stat.ME↗

arXiv2026

AeroTherm-GPT: A Verification-Centered LLM Framework for Thermal Protection System Engineering Workflows

Integrating Large Language Models (LLMs) into hypersonic thermal protection system (TPS) design is bottlenecked by cascading constraint violations when generating executable simulation artifacts. General-purpose LLMs, treating generation as single-pass text completion, fail to satisfy the sequential, multi-gate constraints inherent in safety-critical engineering workflows. To address this, we propose AeroTherm-GPT, the first TPS-specialized LLM Agent, instantiated through a Constraint-Closed-Loop Generation (CCLG) framework. CCLG organizes TPS artifact generation as an iterative workflow comprising generation, validation, CDG-guided repair, execution, and audit. The Constraint Dependency Graph (CDG) encodes empirical co-resolution structure among constraint categories, directing repair toward upstream fault candidates based on lifecycle ordering priors and empirical co-resolution probabilities. This upstream-priority mechanism resolves multiple downstream violations per action, achieving a Root-Cause Fix Efficiency of 4.16 versus 1.76 for flat-checklist repair. Evaluated on HyTPS-Bench and validated against external benchmarks, AeroTherm-GPT achieves 88.7% End-to-End Success Rate (95% CI: 87.5-89.9), a gain of +12.5 pp over the matched non-CDG ablation baseline, without catastrophic forgetting on scientific reasoning and code generation tasks.

cs.AI↗

arXiv2026

ColorFLUX: A Structure-Color Decoupling Framework for Old Photo Colorization

Old photos preserve invaluable historical memories, making their restoration and colorization highly desirable. While existing restoration models can address some degradation issues like denoising and scratch removal, they often struggle with accurate colorization. This limitation arises from the unique degradation inherent in old photos, such as faded brightness and altered color hues, which are different from modern photo distributions, creating a substantial domain gap during colorization. In this paper, we propose a novel old photo colorization framework based on the generative diffusion model FLUX. Our approach introduces a structure-color decoupling strategy that separates structure preservation from color restoration, enabling accurate colorization of old photos while maintaining structural consistency. We further enhance the model with a progressive Direct Preference Optimization (Pro-DPO) strategy, which allows the model to learn subtle color preferences through coarse-to-fine transitions in color augmentation. Additionally, we address the limitations of text-based prompts by introducing visual semantic prompts, which extract fine-grained semantic information directly from old photos, helping to eliminate the color bias inherent in old photos. Experimental results on both synthetic and real datasets demonstrate that our approach outperforms existing state-of-the-art colorization methods, including closed-source commercial models, producing high-quality and vivid colorization.

cs.CV↗

arXiv2026

CREval: An Automated Interpretable Evaluation for Creative Image Manipulation under Complex Instructions

Instruction-based multimodal image manipulation has recently made rapid progress. However, existing evaluation methods lack a systematic and human-aligned framework for assessing model performance on complex and creative editing tasks. To address this gap, we propose CREval, a fully automated question-answer (QA)-based evaluation pipeline that overcomes the incompleteness and poor interpretability of opaque Multimodal Large Language Models (MLLMs) scoring. Simultaneously, we introduce CREval-Bench, a comprehensive benchmark specifically designed for creative image manipulation under complex instructions. CREval-Bench covers three categories and nine creative dimensions, comprising over 800 editing samples and 13K evaluation queries. Leveraging this pipeline and benchmark, we systematically evaluate a diverse set of state-of-the-art open and closed-source models. The results reveal that while closed-source models generally outperform open-source ones on complex and creative tasks, all models still struggle to complete such edits effectively. In addition, user studies demonstrate strong consistency between CREval's automated metrics and human judgments. Therefore, CREval provides a reliable foundation for evaluating image editing models on complex and creative image manipulation tasks, and highlights key challenges and opportunities for future research.

cs.CV↗

arXiv2026

Identification and estimation of causal mechanisms in cluster-randomized trials with post-treatment confounding using Bayesian nonparametrics

Causal mediation analysis in cluster-randomized trials (CRTs) is essential for explaining how cluster-level interventions affect individual outcomes, yet it is complicated by interference, post-treatment confounding, and hierarchical covariate adjustment. We develop a Bayesian nonparametric framework that simultaneously accommodates interference and a post-treatment confounder that precedes the mediator. Identification is achieved through a multivariate Gaussian copula that replaces cross-world independence with a single dependence parameter, yielding a built-in sensitivity analysis to residual post-treatment confounding. For estimation, we introduce a nested common atoms enriched Dirichlet process (CA-EDP) prior that integrates the Common Atoms Model (CAM) to share information across clusters while capturing between- and within-cluster heterogeneity, and an Enriched Dirichlet Process (EDP) structure delivering robust covariate adjustment without impacting the outcome model. We provide formal theoretical support for our prior by deriving the model's key distributional properties, including its partially exchangeable partition structure, and by establishing convergence guarantees for the practical truncation-based posterior inference strategy. We demonstrate the performance of the proposed methods in simulations and provide further illustration through a reanalysis of a completed CRT.

stat.ME↗

arXiv2026

Optimal Sample Size Calculation in Cost-Effectiveness Longitudinal Cluster Randomized Trials

Longitudinal cluster randomized trials (L-CRTs) are increasingly used to evaluate the cost-effectiveness of healthcare interventions across multiple assessment periods, yet design methods for powering these trials remain underdeveloped. Existing methods for cost-effectiveness analyses in cluster settings are limited to simple parallel-arm cluster randomized trials with a single follow-up assessment period. These methods cannot accommodate the complex correlation structures in L-CRTs conducted over multiple periods, which require differentiation between within-period and between-period correlations for both clinical and cost outcomes, as well as between-outcome correlations. Moreover, while substantial methodological advances have been made for the design of L-CRTs with univariate outcomes, none specifically address cost-effectiveness objectives where clinical and cost outcomes must be jointly modeled. We provide a design-stage framework for powering cost-effectiveness L-CRTs across three design variants: parallel-arm, crossover, and stepped wedge designs. We derive closed-form variance expressions for the generalized least squares estimator of the average incremental net monetary benefit under a bivariate linear mixed model. We propose a standardized ceiling ratio that adjusts willingness-to-pay for relative outcome variability to inform optimal design. We then develop local optimal designs that maximize statistical power under known correlation parameters and MaxiMin designs that ensure robust performance across parameter uncertainty for all three design variants. Through a real stepped wedge trial data example, we demonstrate the sample size calculation for testing intervention cost-effectiveness under local optimal and MaxiMin designs.

stat.ME↗

arXiv2026

Principal stratification with recurrent events truncated by a terminal event: A nested Bayesian nonparametric approach

Recurrent events often serve as key endpoints in clinical studies but may be prematurely truncated by terminal events such as death, creating selection bias and complicating causal inference. To address this challenge, we develop a Bayesian nonparametric framework to address potential selection bias due to truncation by death within the continuous-time principal stratification framework. We introduce causal estimands for recurrent events in the presence of a terminal event and derive a partial identification result for the estimand under a dual-frailty framework, enabling transparent sensitivity analysis for non-identifiable parameters. We then propose a flexible Bayesian nonparametric prior, the enriched dependent Dirichlet process, specifically designed for joint modeling of recurrent and terminal events, addressing a limitation where standard Dirichlet process priors create random partitions dominated by recurrent events, yielding poor predictive performance for terminal events. Simulations are carried out to show that our method has superior performance compared to existing methods. We apply the proposed new Bayesian nonparametric methods to infer the causal effect of a structured exercise program on rehospitalizations, which are subject to truncation by death.

stat.ME↗

arXiv2026

BCMI-Driven Motion Control Detection: EEG-Based Machine Learning and Interaction Entropy for High-Order Brain Networks

This study investigates the cognitive motor control detection and the underlying neuroregulatory mechanisms during music-assisted simulated driving. Using a dynamic higher-order network model constructed with EEG-based cross-information entropy, we quantify the dynamic coordination within brain networks activated during both music listening and driving. This approach, which contrasts with previous static network analyses, provides novel insights into how musical stimuli modulate the complex interplay of brain regions during demanding tasks. Results demonstrated enhanced third-order connectivity and elevated higher-order information entropy in music-stimulated driving compared to baseline driving, as evidenced by increasing Phi values of higher-order network indices. Supervised machine learning, including support vector machines, revealed a strong correlation between model accuracy and ROC-AUC values and the hierarchy of brain network features. This underscores the importance of higher-order features in decoding brain motor-control states during music-simulated driving. These findings deepen our understanding of the interplay between music cognition and motor control, offering valuable insights for the development of novel brain-computer-music interfaces (BCMI) and adaptive human-machine systems to enhance performance in demanding tasks like driving.

q-bio.NC↗

arXiv2026

CFD-HAR: User-controllable Privacy through Conditional Feature Disentanglement

Modern wearable and mobile devices are equipped with inertial measurement units (IMUs). Human Activity Recognition (HAR) applications running on such devices use machine-learning-based, data-driven techniques that leverage such sensor data. However, sensor-data-driven HAR deployments face two critical challenges: protecting sensitive user information embedded in sensor data in accordance with users' privacy preferences and maintaining high recognition performance with limited labeled samples. This paper proposes a technique for user-controllable privacy through feature disentanglement-based representation learning at the granular level for dynamic privacy filtering. We also compare the efficacy of our technique against few-shot HAR using autoencoder-based representation learning. We analyze their architectural designs, learning objectives, privacy guarantees, data efficiency, and suitability for edge Internet of Things (IoT) deployment. Our study shows that CFD-based HAR provides explicit, tunable privacy protection controls by separating activity and sensitive attributes in the latent space, whereas autoencoder-based few-shot HAR offers superior label efficiency and lightweight adaptability but lacks inherent privacy safeguards. We further examine the security implications of both approaches in continual IoT settings, highlighting differences in susceptibility to representation leakage and embedding-level attacks. The analysis reveals that neither paradigm alone fully satisfies the emerging requirements of next-generation IoT HAR systems. We conclude by outlining research directions toward unified frameworks that jointly optimize privacy preservation, few-shot adaptability, and robustness for trustworthy IoT intelligence.

cs.LG↗

arXiv2026

Forecasting Causal Effects of Future Interventions: Confounding and Transportability Issues

Recent developments in causal inference allow us to transport a causal effect of a time-fixed treatment from a randomized trial to a target population across space but within the same time frame. In contrast to transportability across space, transporting causal effects across time or forecasting causal effects of future interventions is more challenging due to time-varying confounders and time-varying effect modifiers. In this article, we seek to formally clarify the causal estimands for forecasting causal effects over time and the structural assumptions required to identify these estimands. Specifically, we develop a set of novel nonparametric identification formulas--g-computation formulas--for these causal estimands, and lay out the conditions required to accurately forecast causal effects from a past observed sample to a future population in a future time window. Our overarching objective is to leverage the modern causal inference theory to provide a theoretical framework for investigating whether the effects seen in a past sample would carry over to a new future population. Throughout the article, a working example addressing the effect of public policies or social events on COVID-related deaths is considered to contextualize the developments of analytical results.

stat.ME↗

arXiv2026

From Tokenizer Bias to Backbone Capability: A Controlled Study of LLMs for Time Series Forecasting

Using pre-trained large language models (LLMs) as a backbone for time series prediction has recently attracted growing research interest. Existing approaches typically split time series into patches, map them to the token space of LLMs via a Tokenizer, process the tokens through a frozen or fine-tuned LLM backbone, and then reconstruct numerical forecasts using a Detokenizer. However, the actual effectiveness of LLMs for time series forecasting remains under debate. We observe that when trained and evaluated on small datasets, these Tokenizer-Detokenizer pairs often overfit to the specific data distribution, thereby masking the intrinsic predictive capability of the LLM backbone. To investigate the inherent potential of LLMs in this context, we design three models with identical architectures but distinct pre-training strategies. By leveraging large-scale pre-training, we obtain more unbiased Tokenizer-Detokenizer pairs that are seamlessly integrated with the LLM backbone. Through controlled experiments, we evaluate the zero-shot and few-shot forecasting performance of the LLM, offering insights into its true capabilities. Our extensive experiments reveal that, although the LLM backbone shows some promise, its performance remains limited and does not consistently surpass that of models specifically trained on large-scale time series data. Our source code is publicly available in the repository: https://github.com/SiriZhang45/LLM4TS.

cs.LG↗

arXiv2026

Steering and Rectifying Latent Representation Manifolds in Frozen Multi-modal LLMs for Video Anomaly Detection

Video anomaly detection (VAD) aims to identify abnormal events in videos. Traditional VAD methods generally suffer from the high costs of labeled data and full training, thus some recent works have explored leveraging frozen multi-modal large language models (MLLMs) in a tuning-free manner to perform VAD. However, their performance is limited as they directly inherit pre-training biases and cannot adapt internal representations to specific video contexts, leading to difficulties in handling subtle or ambiguous anomalies. To address these limitations, we propose a novel intervention framework, termed SteerVAD, which advances MLLM-based VAD by shifting from passively reading to actively steering and rectifying internal representations. Our approach first leverages the gradient-free representational separability analysis (RSA) to identify top attention heads as latent anomaly experts (LAEs) which are most discriminative for VAD. Then a hierarchical meta-controller (HMC) generates dynamic rectification signals by jointly conditioning on global context and these LAE outputs. The signals execute targeted, anisotropic scaling directly upon the LAE representation manifolds, amplifying anomaly-relevant dimensions while suppressing inherent biases. Extensive experiments on mainstream benchmarks demonstrate our method achieves state-of-the-art performance among tuning-free approaches requiring only 1% of training data, establishing it as a powerful new direction for video anomaly detection. The code will be released upon the publication.

cs.CV↗

arXiv2026

C$^{2}$TC: A Training-Free Framework for Efficient Tabular Data Condensation

Tabular data is the primary data format in industrial relational databases, underpinning modern data analytics and decision-making. However, the increasing scale of tabular data poses significant computational and storage challenges to learning-based analytical systems. This highlights the need for data-efficient learning, which enables effective model training and generalization using substantially fewer samples. Dataset condensation (DC) has emerged as a promising data-centric paradigm that synthesizes small yet informative datasets to preserve data utility while reducing storage and training costs. However, existing DC methods are computationally intensive due to reliance on complex gradient-based optimization. Moreover, they often overlook key characteristics of tabular data, such as heterogeneous features and class imbalance. To address these limitations, we introduce C$^{2}$TC (Class-Adaptive Clustering for Tabular Condensation), the first training-free tabular dataset condensation framework that jointly optimizes class allocation and feature representation, enabling efficient and scalable condensation. Specifically, we reformulate the dataset condensation objective into a novel class-adaptive cluster allocation problem (CCAP), which eliminates costly training and integrates adaptive label allocation to handle class imbalance. To solve the NP-hard CCAP, we develop HFILS, a heuristic local search that alternates between soft allocation and class-wise clustering to efficiently obtain high-quality solutions. Moreover, a hybrid categorical feature encoding (HCFE) is proposed for semantics-preserving clustering of heterogeneous discrete attributes. Extensive experiments on 10 real-world datasets demonstrate that C$^{2}$TC improves efficiency by at least 2 orders of magnitude over state-of-the-art baselines, while achieving superior downstream performance.

cs.LG↗

arXiv2026

A sensitivity analysis approach to principal stratification with a continuous longitudinal intermediate outcome: Applications to a cohort stepped wedge trial

Causal inference in the presence of intermediate variables is a challenging problem in many applications. Principal stratification (PS) provides a framework to estimate principal causal effects (PCE) in such settings. However, existing PS methods primarily focus on settings with binary intermediate variables. We propose a novel approach to estimate PCE with continuous intermediate variables in the context of stepped wedge cluster randomized trials (SW-CRTs). Our method leverages the time-varying treatment assignment in SW-CRTs to calibrate sensitivity parameters and identify the PCE under realistic assumptions. We demonstrate the application of our approach using data from a cohort SW-CRT evaluating the effect of a crowdsourcing intervention on HIV testing uptake among men who have sex with men in China, with social norms as a continuous intermediate variable. The proposed methodology expands the scope of PS to accommodate continuous variables and provides a practical tool for causal inference in SW-CRTs.

stat.ME↗

arXiv2026

Who's Winning? Clarifying Estimands Based on Win Statistics in Cluster Randomized Trials

Treatment effect estimands based on win statistics, including the win ratio, win odds, and win difference are increasingly popular targets for summarizing endpoints in clinical trials. Such win estimands offer an intuitive approach for prioritizing outcomes by clinical importance. The implementation and interpretation of win estimands is complicated in cluster randomized trials (CRTs), where researchers can target fundamentally different estimands on the individual-level or cluster-level. We numerically demonstrate that individual-pair and cluster-pair win estimands can substantially differ when cluster size is informative: where outcomes and/or treatment effects depend on cluster size. With such informative cluster sizes, individual-pair and cluster-pair win estimands can even yield opposite conclusions regarding treatment benefit. We describe consistent estimators for individual-pair and cluster-pair win estimands and propose a leave-one-cluster-out jackknife variance estimator for inference. Despite being consistent, our simulations highlight that some caution is needed when implementing individual-pair win estimators due to finite-sample bias. In contrast, cluster-pair win estimators are unbiased for their respective targets. Altogether, careful specification of the target estimand is essential when applying win estimators in CRTs. Failure to clearly define whether individual-pair or cluster-pair win estimands are of primary interest may result in answering a dramatically different question than intended.

stat.ME↗

arXiv2026

Moving toward best practice when using propensity score weighting in survey observational studies

Propensity score weighting is a common method for estimating treatment effects with survey data. The method is applied to minimize confounding using measured covariates that are often different between individuals in treatment and control. However, existing literature does not reach a consensus on the optimal use of survey weights for population-level inference in the propensity score weighting analysis. Under the balancing weights framework, we provided a unified solution for incorporating survey weights in both the propensity score of estimation and the outcome regression model. We derived estimators for different target populations, including the combined, treated, controlled, and overlap populations. We provide a unified expression of the sandwich variance estimator and demonstrate that the survey-weighted estimator is asymptotically normal, as established through the theory of M-estimators. Through an extensive series of simulation studies, we examined the performance of our derived estimators and compared the results to those of alternative methods. We further carried out two case studies to illustrate the application of the different methods of propensity score analysis with complex survey data. We concluded with a discussion of our findings and provided practical guidelines for propensity score weighting analysis of observational data from complex surveys.

stat.ME↗