arXiv subjects

Tian Liu

Publications and source records attributed to Tian Liu.

At least 37 records · Page 2Linked to original sources

arXiv2025

EANS: Reducing Energy Consumption for UAV with an Environmental Adaptive Navigation Strategy

Unmanned Aerial Vehicles (UAVS) are limited by the onboard energy. Refinement of the navigation strategy directly affects both the flight velocity and the trajectory based on the adjustment of key parameters in the UAVS pipeline, thus reducing energy consumption. However, existing techniques tend to adopt static and conservative strategies in dynamic scenarios, leading to inefficient energy reduction. Dynamically adjusting the navigation strategy requires overcoming the challenges including the task pipeline interdependencies, the environmental-strategy correlations, and the selecting parameters. To solve the aforementioned problems, this paper proposes a method to dynamically adjust the navigation strategy of the UAVS by analyzing its dynamic characteristics and the temporal characteristics of the autonomous navigation pipeline, thereby reducing UAVS energy consumption in response to environmental changes. We compare our method with the baseline through hardware-in-the-loop (HIL) simulation and real-world experiments, showing our method 3.2X and 2.6X improvements in mission time, 2.4X and 1.6X improvements in energy, respectively.

cs.RO↗

arXiv2025

Hunyuan3D 2.5: Towards High-Fidelity 3D Assets Generation with Ultimate Details

In this report, we present Hunyuan3D 2.5, a robust suite of 3D diffusion models aimed at generating high-fidelity and detailed textured 3D assets. Hunyuan3D 2.5 follows two-stages pipeline of its previous version Hunyuan3D 2.0, while demonstrating substantial advancements in both shape and texture generation. In terms of shape generation, we introduce a new shape foundation model -- LATTICE, which is trained with scaled high-quality datasets, model-size, and compute. Our largest model reaches 10B parameters and generates sharp and detailed 3D shape with precise image-3D following while keeping mesh surface clean and smooth, significantly closing the gap between generated and handcrafted 3D shapes. In terms of texture generation, it is upgraded with phyiscal-based rendering (PBR) via a novel multi-view architecture extended from Hunyuan3D 2.0 Paint model. Our extensive evaluation shows that Hunyuan3D 2.5 significantly outperforms previous methods in both shape and end-to-end texture generation.

cs.CV↗

arXiv2025

Hunyuan3D 2.1: From Images to High-Fidelity 3D Assets with Production-Ready PBR Material

3D AI-generated content (AIGC) is a passionate field that has significantly accelerated the creation of 3D models in gaming, film, and design. Despite the development of several groundbreaking models that have revolutionized 3D generation, the field remains largely accessible only to researchers, developers, and designers due to the complexities involved in collecting, processing, and training 3D models. To address these challenges, we introduce Hunyuan3D 2.1 as a case study in this tutorial. This tutorial offers a comprehensive, step-by-step guide on processing 3D data, training a 3D generative model, and evaluating its performance using Hunyuan3D 2.1, an advanced system for producing high-resolution, textured 3D assets. The system comprises two core components: the Hunyuan3D-DiT for shape generation and the Hunyuan3D-Paint for texture synthesis. We will explore the entire workflow, including data preparation, model architecture, training strategies, evaluation metrics, and deployment. By the conclusion of this tutorial, you will have the knowledge to finetune or develop a robust 3D generative model suitable for applications in gaming, virtual reality, and industrial design.

cs.CV↗

arXiv2025

Impact of normal lung volume choices on radiation pneumonitis risk prediction in locally advanced NSCLC radiotherapy

This study is to evaluate the impact of lung volume choices on predicting radiation pneumonitis (RP) risk in patients with locally advanced NSCLC undergoing radiotherapy. Dosimetric variables V20, V5, and mean lung dose (MLD) were extracted from the treatment plans of 442 patients enrolled in the NRG Oncology RTOG 0617 trial. Three lung volumes were defined: total lung excluding gross-tumor-target (TL-GTV), total lung excluding clinical-target-volume (TL-CTV), and total lung excluding planning-target-volume (TL-PTV). Patients were grouped as no-RP2 (N = 377, grade <= 1 RP) and RP2 (N = 65, grade >= 2 RP). Statistical analyses were performed to assess the effect on lung volume definition on RP2 prediction. Three supervised machine learning (ML) models: logistic regression (LR), k-Nearest Neighbor (kNN), and eXtreme Gradient Boosting (XGB), were used to evaluate predictive performance. Model performance was quantified using the area under the receiver operating characteristic curve (AUC), and statistical significance was tested via a bootstrap analysis. Shapley Additive Explanations (SHAP) were applied to interpret feature contributions to model predictions. Statistical analyses showed that V20 and MLD were significantly associated with RP2, while differences among volume definitions were not statistically significant. Both kNN and XGB classifiers consistently yielded higher AUC values for the TL-PTV definition compared to the other definitions, a finding supported by bootstrap analysis. SHAP analysis further indicated that V20 and MLD were the most influential predictors of RP2. Both statistical analysis and SHAP confirmed that V20 and MLD were associated with RP2. The ML models indicated that defining normal lung volume as total lung excluding PTV yielded the highest predictive performance for RP2 risk. Further validation using external datasets is warranted to confirm these findings.

physics.med-ph↗

arXiv2025

Confidence in Large Language Model Evaluation: A Bayesian Approach to Limited-Sample Challenges

Large language models (LLMs) exhibit probabilistic output characteristics, yet conventional evaluation frameworks rely on deterministic scalar metrics. This study introduces a Bayesian approach for LLM capability assessment that integrates prior knowledge through probabilistic inference, addressing limitations under limited-sample regimes. By treating model capabilities as latent variables and leveraging a curated query set to induce discriminative responses, we formalize model ranking as a Bayesian hypothesis testing problem over mutually exclusive capability intervals. Experimental evaluations with GPT-series models demonstrate that the proposed method achieves superior discrimination compared to conventional evaluation methods. Results indicate that even with reduced sample sizes, the approach maintains statistical robustness while providing actionable insights, such as probabilistic statements about a model's likelihood of surpassing specific baselines. This work advances LLM evaluation methodologies by bridging Bayesian inference with practical constraints in real-world deployment scenarios.

cs.CL↗

arXiv2025

Profile-Preserving Phase-Field Model for Surfactant Transport and Adsorption-Desorption in Two-Phase Flow Systems

The diffuse-interface model for two-phase flows with soluble surfactants has garnered considerable attention due to its ability to circumvent the need for Robin boundary condition in the bulk surfactant transport equation. However, the coupling between surfactant concentration and the phase field within this framework underscores the importance of accurately resolving interfacial equilibrium profiles. To address this limitation, we have developed a profile-preserving phase-field model for simulating surfactant transport and adsorption-desorption in two-phase flow systems. This approach iteratively refines interfacial profiles and delta functions, removing concentration singularities and improving mass conservation. The effectiveness of the model is demonstrated through two benchmark simulations: surfactant transport in a vortex-deformed droplet, which quantitatively reveals reduced mass error over time, and adsorption-desorption dynamics on a stationary spherical interface, showing strong agreement with one-dimensional analytical solutions for surfactant concentration distributions. We further highlight the model's capability by simulating the settling behavior of a surfactant-laden droplet, underscoring the critical role of adsorption-desorption kinetics in governing droplet dynamics.

physics.flu-dyn↗

arXiv2025

Few-Shot Recognition via Stage-Wise Retrieval-Augmented Finetuning

Few-shot recognition (FSR) aims to train a classification model with only a few labeled examples of each concept concerned by a downstream task, where data annotation cost can be prohibitively high. We develop methods to solve FSR by leveraging a pretrained Vision-Language Model (VLM). We particularly explore retrieval-augmented learning (RAL), which retrieves open data, e.g., the VLM's pretraining dataset, to learn models for better serving downstream tasks. RAL has been studied in zero-shot recognition but remains under-explored in FSR. Although applying RAL to FSR may seem straightforward, we observe interesting and novel challenges and opportunities. First, somewhat surprisingly, finetuning a VLM on a large amount of retrieved data underperforms state-of-the-art zero-shot methods. This is due to the imbalanced distribution of retrieved data and its domain gaps with the few-shot examples in the downstream task. Second, more surprisingly, we find that simply finetuning a VLM solely on few-shot examples significantly outperforms previous FSR methods, and finetuning on the mix of retrieved and few-shot data yields even better results. Third, to mitigate the imbalanced distribution and domain gap issues, we propose Stage-Wise retrieval-Augmented fineTuning (SWAT), which involves end-to-end finetuning on mixed data in the first stage and retraining the classifier on the few-shot data in the second stage. Extensive experiments on nine popular benchmarks demonstrate that SWAT significantly outperforms previous methods by >6% accuracy.

cs.CV↗

arXiv2025

Weakly Supervised Spatial Implicit Neural Representation Learning for 3D MRI-Ultrasound Deformable Image Registration in HDR Prostate Brachytherapy

Purpose: Accurate 3D MRI-ultrasound (US) deformable registration is critical for real-time guidance in high-dose-rate (HDR) prostate brachytherapy. We present a weakly supervised spatial implicit neural representation (SINR) method to address modality differences and pelvic anatomy challenges. Methods: The framework uses sparse surface supervision from MRI/US segmentations instead of dense intensity matching. SINR models deformations as continuous spatial functions, with patient-specific surface priors guiding a stationary velocity field for biologically plausible deformations. Validation included 20 public Prostate-MRI-US-Biopsy cases and 10 institutional HDR cases, evaluated via Dice similarity coefficient (DSC), mean surface distance (MSD), and 95% Hausdorff distance (HD95). Results: The proposed method achieved robust registration. For the public dataset, prostate DSC was $0.93 \pm 0.05$, MSD $0.87 \pm 0.10$ mm, and HD95 $1.58 \pm 0.37$ mm. For the institutional dataset, prostate CTV achieved DSC $0.88 \pm 0.09$, MSD $1.21 \pm 0.38$ mm, and HD95 $2.09 \pm 1.48$ mm. Bladder and rectum performance was lower due to ultrasound's limited field of view. Visual assessments confirmed accurate alignment with minimal discrepancies. Conclusion: This study introduces a novel weakly supervised SINR-based approach for 3D MRI-US deformable registration. By leveraging sparse surface supervision and spatial priors, it achieves accurate, robust, and computationally efficient registration, enhancing real-time image guidance in HDR prostate brachytherapy and improving treatment precision.

physics.med-ph↗

arXiv2024

Chance-Constrained Economic Dispatch with Flexible Loads and RES

With the increasing penetration of intermittent renewable energy sources (RESs), it becomes increasingly challenging to maintain the supply-demand balance of power systems by solely relying on the generation side. To combat the volatility led by the uncertain RESs, demand-side management by leveraging the multi-dimensional flexibility (MDF) has been recognized as an economic and efficient approach. Thus, it is important to integrate MDF into existing power systems. In this paper, we propose an enhanced day-ahead energy market, where the MDFs of aggregate loads are traded to minimize the generation cost and mitigate the volatility of locational marginal prices (LMPs) in the transmission network. We first explicitly capture the negative impact of the uncertainty from RESs on the day-ahead market by a chance-constrained economic dispatch problem (CEDP). Then, we propose a bidding mechanism for the MDF of the aggregate loads and combine this mechanism into the CEDP for the day-ahead market. Through multiple case studies, we show that MDF from load aggregators can reduce the volatility of LMPs. In addition, we identify the values of the different flexibilities in the MDF bids, which provide useful insights into the design of more complex MDF markets.

eess.SY↗

arXiv2024

Robust Economic Dispatch with Flexible Demand and Adjustable Uncertainty Set

With more renewable energy sources (RES) integrated into the power system, the intermittency of RES places a heavy burden on the system. The uncertainty of RES is traditionally handled by controllable generators to balance the real time wind power deviation. As the demand side management develops, the flexibility of aggregate loads can be leveraged to mitigate the negative impact of the wind power. In view of this, we study the problem of how to exploit the multi-dimensional flexibility of elastic loads to balance the trade-off between a low generation cost and a low system risk related to the wind curtailment and the power deficiency. These risks are captured by the conditional value-at-risk. Also, unlike most of the existing studies, the uncertainty set of the wind power output in our model is not fixed. By contrast, it is undetermined and co-optimized based on the available load flexibility. We transform the original optimization problem into a convex one using surrogate affine approximation such that it can be solved efficiently. In case studies, we apply our model on a six-bus transmission network and demonstrate that how flexible load aggregators can help to determine the optimal admissible region for the wind power deviations.

eess.SY↗

arXiv2024

ERIC: Estimating Rainfall with Commodity Doorbell Camera for Precision Residential Irrigation

Current state-of-the-art residential irrigation systems, such as WaterMyYard, rely on rainfall data from nearby weather stations to adjust irrigation amounts. However, the accuracy of rainfall data is compromised by the limited spatial resolution of rain gauges and the significant variability of hyperlocal rainfall, leading to substantial water waste. To improve irrigation efficiency, we developed a cost-effective irrigation system, dubbed ERIC, which employs machine learning models to estimate rainfall from commodity doorbell camera footage and optimizes irrigation schedules without human intervention. Specifically, we: a) designed novel visual and audio features with lightweight neural network models to infer rainfall from the camera at the edge, preserving user privacy; b) built a complete end-to-end irrigation system on Raspberry Pi 4, costing only \$75. We deployed the system across five locations (collecting over 750 hours of video) with varying backgrounds and light conditions. Comprehensive evaluation validates that ERIC achieves state-of-the-art rainfall estimation performance ($\sim$ 5mm/day), saving 9,112 gallons/month of water, translating to \$28.56/month in utility savings. Data and code are available at https://github.com/LENSS/ERIC-BuildSys2024.git

cs.CV↗

arXiv2024

UAL-Bench: The First Comprehensive Unusual Activity Localization Benchmark

Localizing unusual activities, such as human errors or surveillance incidents, in videos holds practical significance. However, current video understanding models struggle with localizing these unusual events likely because of their insufficient representation in models' pretraining datasets. To explore foundation models' capability in localizing unusual activity, we introduce UAL-Bench, a comprehensive benchmark for unusual activity localization, featuring three video datasets: UAG-OOPS, UAG-SSBD, UAG-FunQA, and an instruction-tune dataset: OOPS-UAG-Instruct, to improve model capabilities. UAL-Bench evaluates three approaches: Video-Language Models (Vid-LLMs), instruction-tuned Vid-LLMs, and a novel integration of Vision-Language Models and Large Language Models (VLM-LLM). Our results show the VLM-LLM approach excels in localizing short-span unusual events and predicting their onset (start time) more accurately than Vid-LLMs. We also propose a new metric, R@1, TD <= p, to address limitations in existing evaluation methods. Our findings highlight the challenges posed by long-duration videos, particularly in autism diagnosis scenarios, and the need for further advancements in localization techniques. Our work not only provides a benchmark for unusual activity localization but also outlines the key challenges for existing foundation models, suggesting future research directions on this important task.

cs.CV↗

arXiv2024

Enhanced Profile-Preserving Phase-Field model of Two-Phase Flow with Surfactant Interfacial Transport and Marangoni Effects

Using a regularized delta function to distribute surfactant interfacial concentration can simplify the computation of the surface gradient operator $\nabla_s$, enabling the phase-field model to effectively simulate Marangoni flows involving surfactant transport. However, the exact conservation of total surfactant mass is compromised due to deviation from the equilibrium phase field profile, numerical diffusion, and mass non-conservation in each phase. To overcome these limitations, we have developed a new model for simulating two-phase flow with surfactant transport along the interface. This model employs a profile-preserving strategy to maintain the equilibrium interface profile, ensuring accurate calculation of the regularized delta function and better surfactant mass conservation. Within the framework of the advective Chan-Hilliard phase-field model, we utilize a regularized delta function with a reduced gradient to minimize numerical diffusion. Furthermore, we introduce a hybrid surface tension model that integrates the free-energy and the continuum-surface force models to mitigate spatial discretization errors, particularly in scenarios with high density and viscosity ratio. Verification tests demonstrates the model's effectiveness in simulating surface diffusion on stationary and expanding drop, suppressing spurious currents, and capturing the deformation of two-dimensional drops in shear flow. The results closely align with analytical solutions and previous numerical studies. Finally, we apply the model to investigate the contraction and oscillation dynamics of a surfactant-laden liquid filament, revealing the role of the Marangoni force in shaping filament behavior.

physics.flu-dyn↗

arXiv2024

The Neglected Tails in Vision-Language Models

Vision-language models (VLMs) excel in zero-shot recognition but their performance varies greatly across different visual concepts. For example, although CLIP achieves impressive accuracy on ImageNet (60-80%), its performance drops below 10% for more than ten concepts like night snake, presumably due to their limited presence in the pretraining data. However, measuring the frequency of concepts in VLMs' large-scale datasets is challenging. We address this by using large language models (LLMs) to count the number of pretraining texts that contain synonyms of these concepts. Our analysis confirms that popular datasets, such as LAION, exhibit a long-tailed concept distribution, yielding biased performance in VLMs. We also find that downstream applications of VLMs, including visual chatbots (e.g., GPT-4V) and text-to-image models (e.g., Stable Diffusion), often fail to recognize or generate images of rare concepts identified by our method. To mitigate the imbalanced performance of zero-shot VLMs, we propose REtrieval-Augmented Learning (REAL). First, instead of prompting VLMs using the original class names, REAL uses their most frequent synonyms found in pretraining texts. This simple change already outperforms costly human-engineered and LLM-enriched prompts over nine benchmark datasets. Second, REAL trains a linear classifier on a small yet balanced set of pretraining data retrieved using concept synonyms. REAL surpasses the previous zero-shot SOTA, using 400x less storage and 10,000x less training time!

cs.CV↗

arXiv2024

Learning using granularity statistical invariants for classification

Learning using statistical invariants (LUSI) is a new learning paradigm, which adopts weak convergence mechanism, and can be applied to a wider range of classification problems. However, the computation cost of invariant matrices in LUSI is high for large-scale datasets during training. To settle this issue, this paper introduces a granularity statistical invariant for LUSI, and develops a new learning paradigm called learning using granularity statistical invariants (LUGSI). LUGSI employs both strong and weak convergence mechanisms, taking a perspective of minimizing expected risk. As far as we know, it is the first time to construct granularity statistical invariants. Compared to LUSI, the introduction of this new statistical invariant brings two advantages. Firstly, it enhances the structural information of the data. Secondly, LUGSI transforms a large invariant matrix into a smaller one by maximizing the distance between classes, achieving feasibility for large-scale datasets classification problems and significantly enhancing the training speed of model operations. Experimental results indicate that LUGSI not only exhibits improved generalization capabilities but also demonstrates faster training speed, particularly for large-scale datasets.

cs.LG↗

arXiv2024

Cross-Shaped Windows Transformer with Self-supervised Pretraining for Clinically Significant Prostate Cancer Detection in Bi-parametric MRI

Biparametric magnetic resonance imaging (bpMRI) has demonstrated promising results in prostate cancer (PCa) detection using convolutional neural networks (CNNs). Recently, transformers have achieved competitive performance compared to CNNs in computer vision. Large scale transformers need abundant annotated data for training, which are difficult to obtain in medical imaging. Self-supervised learning (SSL) utilizes unlabeled data to generate meaningful semantic representations without the need for costly annotations, enhancing model performance on tasks with limited labeled data. We introduce a novel end-to-end Cross-Shaped windows (CSwin) transformer UNet model, CSwin UNet, to detect clinically significant prostate cancer (csPCa) in prostate bi-parametric MR imaging (bpMRI) and demonstrate the effectiveness of our proposed self-supervised pre-training framework. Using a large prostate bpMRI dataset with 1500 patients, we first pretrain CSwin transformer using multi-task self-supervised learning to improve data-efficiency and network generalizability. We then finetune using lesion annotations to perform csPCa detection. Five-fold cross validation shows that self-supervised CSwin UNet achieves 0.888 AUC and 0.545 Average Precision (AP), significantly outperforming four comparable models (Swin UNETR, DynUNet, Attention UNet, UNet). Using a separate bpMRI dataset with 158 patients, we evaluate our method robustness to external hold-out data. Self-supervised CSwin UNet achieves 0.79 AUC and 0.45 AP, still outperforming all other comparable methods and demonstrating good generalization to external data.

eess.IV↗

arXiv2024

Assessing Bilateral Neurovascular Bundles Function with Pulsed Wave Doppler Ultrasound: Implications for Reducing Erectile Dysfunction Following Prostate Radiotherapy

This study aims to evaluate the functional status of bilateral neurovascular bundles (NVBs) using pulsed wave Doppler ultrasound in patients undergoing prostate radiotherapy (RT). Sixty-two patients (mean age: 66.1 +/- 7.2 years) underwent transrectal ultrasound scan using a conventional ultrasound scanner, a 7.5 MHz bi-plane probe and a mechanical stepper. The ultrasound protocol comprised 3 steps: 1) 3D B-mode scans of the entire prostate, 2) localization of NVBs using color flow Doppler imaging, and 3) measurement of NVB function using pulsed wave Doppler. Five pulsed Doppler waveform features were extracted: peak systolic velocity (PSV), end-diastolic velocity (EDV), mean velocity (Vm), resistive index (RI), and pulsatile index (PI). In summary, this study presents a Doppler evaluation of NVBs in patients undergoing prostate RT. It highlights substantial differences in Doppler ultrasound waveform features between bilateral NVBs. The proposed ultrasound method may prove valuable as clinicians strive to deliver NVB-sparing RT to preserve sexual function effectively and enhance patients' overall well-being.

physics.med-ph↗

arXiv2024

Technical Report: On the Convergence of Gossip Learning in the Presence of Node Inaccessibility

Gossip learning (GL), as a decentralized alternative to federated learning (FL), is more suitable for resource-constrained wireless networks, such as Flying Ad-Hoc Networks (FANETs) that are formed by unmanned aerial vehicles (UAVs). GL can significantly enhance the efficiency and extend the battery life of UAV networks. Despite the advantages, the performance of GL is strongly affected by data distribution, communication speed, and network connectivity. However, how these factors influence the GL convergence is still unclear. Existing work studied the convergence of GL based on a virtual quantity for the sake of convenience, which failed to reflect the real state of the network when some nodes are inaccessible. In this paper, we formulate and investigate the impact of inaccessible nodes to GL under a dynamic network topology. We first decompose the weight divergence by whether the node is accessible or not. Then, we investigate the GL convergence under the dynamic of node accessibility and theoretically provide how the number of inaccessible nodes, data non-i.i.d.-ness, and duration of inaccessibility affect the convergence. Extensive experiments are carried out in practical settings to comprehensively verify the correctness of our theoretical findings.

cs.LG↗