arXiv subjects

Yun Lu

Publications and source records attributed to Yun Lu.

16 recordsLinked to original sources

arXiv2026

SmoothVLA: Aligning Vision-Language-Action Models with Physical Constraints via Intrinsic Smoothness Optimization

Vision-Language-Action (VLA) models have emerged as a powerful paradigm for robotic manipulation. However, existing post-training methods face a dilemma between stability and exploration: Supervised Fine-Tuning (SFT) is constrained by demonstration quality and lacks generalization, whereas Reinforcement Learning (RL) improves exploration but often induces erratic, jittery trajectories that violate physical constraints. To bridge this gap, we propose SmoothVLA, a novel reinforcement learning fine-tuning framework that synergistically optimizes task performance and motion smoothness. The technical core is a physics-informed hybrid reward function that integrates binary sparse task rewards with a continuous dense term derived from trajectory jerk. Crucially, this reward is intrinsic, that computing directly from policy rollouts, without requiring extrinsic environment feedback or laborious reward engineering. Leveraging the Group Relative Policy Optimization (GRPO), SmoothVLA establishes trajectory smoothness as an explicit optimization prior, guiding the model toward physically feasible and stable control. Extensive experiments on the LIBERO benchmark demonstrate that SmoothVLA outperforms standard RL by 13.8\% in smoothness and significantly surpasses SFT in generalization across diverse tasks. Our work offers a scalable approach to aligning VLA models with physical-world constraints through intrinsic reward optimization.

cs.RO↗

arXiv2026

Fairness Begins with State: Purifying Latent Preferences for Hierarchical Reinforcement Learning in Interactive Recommendation

Interactive recommender systems (IRS) are increasingly optimized with Reinforcement Learning (RL) to capture the sequential nature of user-system dynamics. However, existing fairness-aware methods often suffer from a fundamental oversight: they assume the observed user state is a faithful representation of true preferences. In reality, implicit feedback is contaminated by popularity-driven noise and exposure bias, creating a distorted state that misleads the RL agent. We argue that the persistent conflict between accuracy and fairness is not merely a reward-shaping issue, but a state estimation failure. In this work, we propose \textbf{DSRM-HRL}, a framework that reformulates fairness-aware recommendation as a latent state purification problem followed by decoupled hierarchical decision-making. We introduce a Denoising State Representation Module (DSRM) based on diffusion models to recover the low-entropy latent preference manifold from high-entropy, noisy interaction histories. Built upon this purified state, a Hierarchical Reinforcement Learning (HRL) agent is employed to decouple conflicting objectives: a high-level policy regulates long-term fairness trajectories, while a low-level policy optimizes short-term engagement under these dynamic constraints. Extensive experiments on high-fidelity simulators (KuaiRec, KuaiRand) demonstrate that DSRM-HRL effectively breaks the "rich-get-richer" feedback loop, achieving a superior Pareto frontier between recommendation utility and exposure equity.

cs.LG↗

arXiv2025

PDA in Action: Ten Principles for High-Quality Multi-Site Clinical Evidence Generation

Background: Distributed Research Networks (DRNs) offer significant opportunities for collaborative multi-site research and have significantly advanced healthcare research based on clinical observational data. However, generating high-quality real-world evidence using fit-for-use data from multi-site studies faces important challenges, including biases associated with various types of heterogeneity within and across sites and data sharing difficulties. Over the last ten years, Privacy-Preserving Distributed Algorithms (PDA) have been developed and utilized in numerous national and international real-world studies spanning diverse domains, from comparative effectiveness research, target trial emulation, to healthcare delivery, policy evaluation, and system performance assessment. Despite these advances, there remains a lack of comprehensive and clear guiding principles for generating high-quality real-world evidence through collaborative studies leveraging the methods under PDA. Objective: The paper aims to establish ten principles of best practice for conducting high-quality multi-site studies using PDA. These principles cover all phases of research, including study preparation, protocol development, analysis, and final reporting. Discussion: The ten principles for conducting a PDA study outline a principled, efficient, and transparent framework for employing distributed learning algorithms within DRNs to generate reliable and reproducible real-world evidence.

cs.CY↗

arXiv2025

Revisiting Fairness-aware Interactive Recommendation: Item Lifecycle as a Control Knob

This paper revisits fairness-aware interactive recommendation (e.g., TikTok, KuaiShou) by introducing a novel control knob, i.e., the lifecycle of items. We make threefold contributions. First, we conduct a comprehensive empirical analysis and uncover that item lifecycles in short-video platforms follow a compressed three-phase pattern, i.e., rapid growth, transient stability, and sharp decay, which significantly deviates from the classical four-stage model (introduction, growth, maturity, decline). Second, we introduce LHRL, a lifecycle-aware hierarchical reinforcement learning framework that dynamically harmonizes fairness and accuracy by leveraging phase-specific exposure dynamics. LHRL consists of two key components: (1) PhaseFormer, a lightweight encoder combining STL decomposition and attention mechanisms for robust phase detection; (2) a two-level HRL agent, where the high-level policy imposes phase-aware fairness constraints, and the low-level policy optimizes immediate user engagement. This decoupled optimization allows for effective reconciliation between long-term equity and short-term utility. Third, experiments on multiple real-world interactive recommendation datasets demonstrate that LHRL significantly improves both fairness and user engagement. Furthermore, the integration of lifecycle-aware rewards into existing RL-based models consistently yields performance gains, highlighting the generalizability and practical value of our approach.

cs.AI↗

arXiv2025

Empowering Clinical Trial Design through AI: A Randomized Evaluation of PowerGPT

Sample size calculations for power analysis are critical for clinical research and trial design, yet their complexity and reliance on statistical expertise create barriers for many researchers. We introduce PowerGPT, an AI-powered system integrating large language models (LLMs) with statistical engines to automate test selection and sample size estimation in trial design. In a randomized trial to evaluate its effectiveness, PowerGPT significantly improved task completion rates (99.3% vs. 88.9% for test selection, 99.3% vs. 77.8% for sample size calculation) and accuracy (94.1% vs. 55.4% in sample size estimation, p < 0.001), while reducing average completion time (4.0 vs. 9.3 minutes, p < 0.001). These gains were consistent across various statistical tests and benefited both statisticians and non-statisticians as well as bridging expertise gaps. Already under deployment across multiple institutions, PowerGPT represents a scalable AI-driven approach that enhances accessibility, efficiency, and accuracy in statistical power analysis for clinical research.

cs.AI↗

arXiv2025

Is Artificial Intelligence Reshaping the Landscape of the International Academic Community of Geosciences?

Through bibliometric analysis and topic modeling, we find that artificial intelligence (AI) is positively transforming geosciences research, with a notable increase in AI-related scientific output in recent years. We are encouraged to observe that earth scientists from developing countries have gained better visibility in the recent AI for Science (AI4S) paradigm and that AI is also improving the landscape of international collaboration in geoscience-related research.

cs.DL↗

arXiv2025

General-Purpose $f$-DP Estimation and Auditing in a Black-Box Setting

In this paper we propose new methods to statistically assess $f$-Differential Privacy ($f$-DP), a recent refinement of differential privacy (DP) that remedies certain weaknesses of standard DP (including tightness under algorithmic composition). A challenge when deploying differentially private mechanisms is that DP is hard to validate, especially in the black-box setting. This has led to numerous empirical methods for auditing standard DP, while $f$-DP remains less explored. We introduce new black-box methods for $f$-DP that, unlike existing approaches for this privacy notion, do not require prior knowledge of the investigated algorithm. Our procedure yields a complete estimate of the $f$-DP trade-off curve, with theoretical guarantees of convergence. Additionally, we propose an efficient auditing method that empirically detects $f$-DP violations with statistical certainty, merging techniques from non-parametric estimation and optimal classification theory. Through experiments on a range of DP mechanisms, we demonstrate the effectiveness of our estimation and auditing procedures.

cs.CR↗

arXiv2024

A Surrogate Endpoint Based Provisional Approval Causal Roadmap

For many rare diseases with no approved preventive interventions, promising interventions exist, yet it has been difficult to conduct a pivotal phase 3 trial that could provide direct evidence demonstrating a beneficial effect on the target disease outcome. When a promising putative surrogate endpoint(s) for the target outcome is available, surrogate-based provisional approval of an intervention may be pursued. We apply the Causal Roadmap rubric to define a surrogate endpoint based provisional approval causal roadmap, which combines observational study data that estimates the relationship between the putative surrogate and the target outcome, with a phase 3 surrogate endpoint study that collects the same data but is very under-powered to assess the treatment effect (TE) on the target outcome. The objective is conservative estimation/inference for the TE with an estimated lower uncertainty bound that allows (through two bias functions) for an imperfect surrogate and imperfect transport of the conditional target outcome risk in the untreated between the observational and phase 3 studies. Two estimators of TE (plug-in, nonparametric efficient one-step) with corresponding inference procedures are developed. Finite-sample performance of the plug-in estimator is evaluated in two simulation studies, with R code provided. The roadmap is illustrated with contemporary Group B Streptococcus vaccine development.

stat.ME↗

arXiv2024

Why Studying Cut-ins? Comparing Cut-ins and Other Lane Changes Based on Naturalistic Driving Data

Extensive research has been conducted to explore vehicle lane changes, while the study on cut-ins has not received sufficient attention. The existing studies have not addressed the fundamental question of why studying cut-ins is crucial, despite the extensive investigation into lane changes. To tackle this issue, it is important to demonstrate how cut-ins, as a special type of lane change, differ from other lane changes. In this paper, we explore to compare driving characteristics of cut-ins and other lane changes based on naturalistic driving data. The highD dataset is employed to conduct the comparison. We extract all lane-change events from the dataset and exclude events that are not suitable for our comparison. Lane-change events are then categorized into the cut-in events and other lane-change events based on various gap-based rules. Several performance metrics are designed to measure the driving characteristics of the two types of events. We prove the significant differences between the cut-in behavior and other lane-change behavior by using the Wilcoxon rank-sum test. The results suggest the necessity of conducting specialized studies on cut-ins, offering valuable insights for future research in this field.

eess.SY↗

arXiv2024

Flexible Non-interactive Short-term Implicit Certificate Generation for VANETs

A leading industry standard for secure and trusted communication in vehicular ad-hoc networks (VANETs) is the Security Credential Management System (SCMS). It uses anonymous certificates, functioning as pseudonyms, to preserve the privacy of vehicles. With the rapid development of advanced applications in VANETs, such as crowdsensing and federated learning, vehicles need to communicate with each other or infrastructures more frequently, leading to a higher demand for pseudonyms. However, the current approach of certificate provisioning in SCMS is not able to fully support pseudonyms, due to storage limitation, cost of connectivity establishment, and communication overhead of certificate downloading. To tackle this challenge, we propose a non-interactive approach for SCMS, allowing vehicles themselves to generate short-term key pairs and anonymous implicit certificates. Our evaluation and comparison with previous work show that our solution not only effectively reduces the communication cost, but also grants vehicles greater flexibility in certificate generation and use. On the technical side, to the best of our knowledge, this is the first work which (1) applies sanitizable signature for non-interactive anonymous certificate generation, and (2) is specifically designed for SCMS, which opens up possibilities for extensions and applications in industry.

cs.CR↗

arXiv2023

The Normal Distributions Indistinguishability Spectrum and its Application to Privacy-Preserving Machine Learning

We investigate the privacy of {\em any} algorithm whose outputs have Gaussian distribution. This work is motivated by the prevalence of such algorithms in several useful (ML) applications, and the comparatively little research that focuses on privacy-preserving learning outside of adding Gaussian noise to the data (such as DP-SGD). {\em What is the DP of any algorithm with multivariate Gaussian output?} We answer the above research question with a general lemma which we call {\em Normal Distributions Indistinguishability Spectrum} (NDIS), a closed-form analytic computation of the hockey-stick divergence $\delta$ between an arbitrary pair of multivariate Gaussians, parameterized by privacy parameter $\epsilon$. To show its practical implications, we prove several properties of our NDIS lemma. These properties form a {\em toolbox} of results which lead to potentially {\em easier} privacy proofs for any Gaussian-output algorithm. As an example application of our toolbox, we prove a tighter parametrisation of the privacy of {\em random projection (RP)}, and obtaining from it a more noise-frugal DP mechanism. Beyond random projection, NDIS can be used to lift {\em any} Gaussian-output algorithm with a `sensitivity' (which we define) to a Gaussian-output DP mechanism. The mechanism boosts the existing randomness in the algorithm, so that one can describe the mechanism's privacy as the IS between a single pair of Gaussians, which can then be analyzed via NDIS. Lastly, we leverage the connections between NDIS and the CDF of the generalized $\chi^2$ distribution (which have efficient empirical estimators) to present a tool for white-box auditing of Gaussian-output algorithms.

cs.CR↗

arXiv2022

Distributed Observers-based Cooperative Platooning Tracking Control and Intermittent Optimization for Connected Automated Vehicles with Unknown Jerk Dynamics

The unknown sharp changes of vehicle acceleration rates, also called the unknown jerk dynamics, may significantly affect the driving performance of the leader vehicle in a platoon, resulting in more drastic car-following movements in platooning tracking control, which could cause safety and traffic capacity concerns. To address these issues, in this paper, we investigate cooperative platooning tracking control and intermittent optimization problems for connected automated vehicles (CAVs) with a nonlinear car-following model. We assume that the external inputs of the leader CAV contain unknown but bounded jerk parameters, and the acceleration signals of the leader CAV are known only to a few neighboring follower CAVs in a free-design but directed communication network. To solve these problems, a distributed observer law is developed to provide a reference signal expressed as an estimated unknown jerk dynamic of the leader CAV and implemented by each follower CAV. Then, a novel distributed platooning tracking control protocol is proposed to construct the cooperative tracking controllers under identical inter-vehicle constraints, which can ensure a desired safety distance among the CAVs and allow each follower CAV to track their leader CAV by using only local information interaction. We also present a novel intermittent sampling condition and a robust intermittent optimization design that can ensure optimally scheduled feedback gains for the cooperative platooning tracking controllers to minimize the control cost under nonidentical inter-vehicle constraints and unknown jerk dynamics. Simulation case studies are carried out to illustrate the effectiveness of the proposed approaches

eess.SY↗

arXiv2021

How Private Are Commonly-Used Voting Rules?

Differential privacy has been widely applied to provide privacy guarantees by adding random noise to the function output. However, it inevitably fails in many high-stakes voting scenarios, where voting rules are required to be deterministic. In this work, we present the first framework for answering the question: "How private are commonly-used voting rules?" Our answers are two-fold. First, we show that deterministic voting rules provide sufficient privacy in the sense of distributional differential privacy (DDP). We show that assuming the adversarial observer has uncertainty about individual votes, even publishing the histogram of votes achieves good DDP. Second, we introduce the notion of exact privacy to compare the privacy preserved in various commonly-studied voting rules, and obtain dichotomy theorems of exact DDP within a large subset of voting rules called generalized scoring rules.

cs.CR↗

arXiv2020

A Novel and Efficient Tumor Detection Framework for Pancreatic Cancer via CT Images

As Deep Convolutional Neural Networks (DCNNs) have shown robust performance and results in medical image analysis, a number of deep-learning-based tumor detection methods were developed in recent years. Nowadays, the automatic detection of pancreatic tumors using contrast-enhanced Computed Tomography (CT) is widely applied for the diagnosis and staging of pancreatic cancer. Traditional hand-crafted methods only extract low-level features. Normal convolutional neural networks, however, fail to make full use of effective context information, which causes inferior detection results. In this paper, a novel and efficient pancreatic tumor detection framework aiming at fully exploiting the context information at multiple scales is designed. More specifically, the contribution of the proposed method mainly consists of three components: Augmented Feature Pyramid networks, Self-adaptive Feature Fusion and a Dependencies Computation (DC) Module. A bottom-up path augmentation to fully extract and propagate low-level accurate localization information is established firstly. Then, the Self-adaptive Feature Fusion can encode much richer context information at multiple scales based on the proposed regions. Finally, the DC Module is specifically designed to capture the interaction information between proposals and surrounding tissues. Experimental results achieve competitive performance in detection with the AUC of 0.9455, which outperforms other state-of-the-art methods to our best of knowledge, demonstrating the proposed framework can detect the tumor of pancreatic cancer efficiently and accurately.

cs.CV↗

arXiv2011

Reducts of the random bipartite graph

Let $Γ$ be the random bipartite graph, a countable graph with two infinite sides, edges randomly distributed between the sides, but no edges within a side. In this paper, we investigate the reducts of $Γ$ that preserve sides. We classify the closed permutation subgroups containing the group $Aut(Γ)^*$, where $Aut(Γ)^*$ is the group of all isomorphisms and anti-isomorphisms of $Γ$ preserving the two sides. Our results rely on a combinatorial theorem of Nešetřil-Rödl and a strong finite submodel property for $Γ$.

math.LO↗

arXiv2011

Reducts of the Generalized Random Bipartite Graph

Let Γbe the generalized random bipartite graph that has two sides Rl and Rr with edges for every pair of vertices between R1 and Rr but no edges within each side, where all the edges are randomly colored by three colors P1; P2; P3. In this paper, we investigate the reducts of Γthat preserve Rl and Rr, and classify the closed permutation subgroups in Sym(Rl)\timesSym(Rr) containing the group Aut(Γ). Our results rely on a combinatorial theorem of Nesetril-Rodl and the strong finite submodel property of the random bipartite graph.

math.LO↗