arXiv subjects

Mingjun Li

Publications and source records attributed to Mingjun Li.

13 recordsLinked to original sources

arXiv2026

GLIP: Graph and LLM Joint Pretraining for Graph-Level Tasks

Graphs are widely used to model relational systems, with applications in domains such as social networks, finance, and biomedicine. Graph neural networks (GNNs) have become a mainstream approach for learning graph representations. With the rise of large language models (LLMs), recent studies have attempted to combine GNNs with LLMs. However, most existing works concentrate on node-level and edge-level tasks, while graph-level tasks, which require capturing more complex structural and feature information, remain relatively underexplored. Moreover, graph pretraining is a widely adopted strategy to alleviate the challenge of label scarcity. Most existing approaches are designed solely for GNNs such as GraphCL, leaving LLMs uninvolved in the process. To address these limitations, we propose GLIP, a Graph-LLM JoInt Pretraining framework for graph-level tasks. GLIP first performs graph augmentation to construct positive and negative pairs and introduces a multi-token selection strategy to identify patches informative in both structure and features. It further leverages a diffusion-based projector to enrich them with contextual information, enabling GLIP to capture signals from both global and local perspectives. Finally, GLIP employs a joint objective that integrates the LLM's semantic judgments with a contrastive alignment loss, ensuring consistent supervision at both the semantic and structural levels. After pretraining, GLIP is fine-tuned with limited labeled data for downstream tasks, and extensive experiments show that it outperforms state-of-the-art methods on graph-level classification and reasoning tasks. Our source code is publicly available at https://anonymous.4open.science/r/GLIP.

cs.LG↗

arXiv2026

Task-Level AI Readiness Assessment for Business Process Management:The T-IPO Model and LARA Matrix in Financial-Services IT Operations

Which tasks inside an enterprise workflow can a large-language-model agent reliably handle, and under what conditions? Most business process modeling frameworks still answer this at the activity level, even though a single activity can bundle work of radically different difficulty. This paper takes the analysis a step smaller. We describe two design artifacts developed in a financial-services IT setting: T-IPO, which represents each task as an eight-element tuple, and LARA (LLM Agent Readiness Assessment), a five-dimension rubric that scores a task's readiness for agent substitution. Compliance Sensitivity carries $1.5\times$ weight, a value we fixed through a three-round Delphi study and cross-checked with AHP. The rubric produces four levels, L1 to L4, and applies a floor rule so that a task with maximum compliance load cannot be classified below L3 no matter what the other scores say. Both artifacts sit inside a larger methodology (PARTIS) that we map onto BWW ontology in Section 3. We evaluate the instruments across 127 tasks. Inter-rater agreement reaches Fleiss' $\kappa = 0.80$; a replication at three further institutions returns $\kappa = 0.73$. A controlled comparison against activity-level assessment suggests, though does not prove, an improvement in predictive utility at the task level. Pilot deployment of 120 task instances confirms that auto-completion decays monotonically from $95\%$ at L1 through about $70\%$ at L2 to about $40\%$ at L3. Exploratory factor analysis points to a two-factor structure: task readiness seems to be determined jointly by cognitive-execution complexity and governance-compliance intensity. We close with a recalibration procedure (LARA-TCA) so the rubric can keep pace with evolving LLM capabilities.

cs.CY↗

arXiv2026

Interfacial optical absorptance of air-water interfaces

The photomolecular effect has been hypothesized to enhance evaporation of water at visible wavelengths. This study develops a measurement technique to investigate its presence and magnitude at the liquid-vapor interface of water. The experiment detects surface absorptance by comparing polarized reflectance for substrates with and without a water layer. The reflectance ratio is used as an indicator of interfacial absorptance or attenuation. Lightly doped silicon and platinum were used as dielectric and metal substrates, and their pseudo-optical properties were measured using ellipsometry. Experimental results were compared to a multilayer reflectance model to determine theoretical sensitivity and guide angle selection. Measurements showed close agreement with a classical model assuming zero surface absorptance. The model was then extended to include surface absorptance using a spectral response based on Lorentzian-distributed Feibelman parameters derived from a recent estimate of 0.84% interfacial absorption. Although this model predicted a 1-2% reduction in reflectance near resonant frequencies, no such spectral features were observed experimentally. These results suggest that under ambient conditions, interfacial absorption is well below 1%, indicating that the optical signature of the photomolecular effect is either much weaker than previously reported or strongly dependent on conditions not present in this experiment.

physics.optics↗

arXiv2025

MAPO: Mixed Advantage Policy Optimization

Recent advances in reinforcement learning for foundation models, such as Group Relative Policy Optimization (GRPO), have significantly improved the performance of foundation models on reasoning tasks. Notably, the advantage function serves as a central mechanism in GRPO for ranking the trajectory importance. However, existing explorations encounter both advantage reversion and advantage mirror problems, which hinder the reasonable advantage allocation across different query samples. In this work, we propose an easy but effective GRPO strategy, Mixed Advantage Policy Optimization (MAPO). We reveal that the trajectory appears with different certainty and propose the advantage percent deviation for samples with high-certainty trajectories. Furthermore, we dynamically reweight the advantage function for samples with varying trajectory certainty, thereby adaptively configuring the advantage function to account for sample-specific characteristics. Comparison with related state-of-the-art methods, along with ablation studies on different advantage variants, validates the effectiveness of our approach.

cs.AI↗

arXiv2025

Predicting 3D Motion from 2D Video for Behavior-Based VR Biometrics

Critical VR applications in domains such as healthcare, education, and finance that use traditional credentials, such as PIN, password, or multi-factor authentication, stand the chance of being compromised if a malicious person acquires the user credentials or if the user hands over their credentials to an ally. Recently, a number of approaches on user authentication have emerged that use motions of VR head-mounted displays (HMDs) and hand controllers during user interactions in VR to represent the user's behavior as a VR biometric signature. One of the fundamental limitations of behavior-based approaches is that current on-device tracking for HMDs and controllers lacks capability to perform tracking of full-body joint articulation, losing key signature data encapsulated by the user articulation. In this paper, we propose an approach that uses 2D body joints, namely shoulder, elbow, wrist, hip, knee, and ankle, acquired from the right side of the participants using an external 2D camera. Using a Transformer-based deep neural network, our method uses the 2D data of body joints that are not tracked by the VR device to predict past and future 3D tracks of the right controller, providing the benefit of augmenting 3D knowledge in authentication. Our approach provides a minimum equal error rate (EER) of 0.025, and a maximum EER drop of 0.040 over prior work that uses single-unit 3D trajectory as the input.

cs.CV↗

arXiv2024

Knowledge Transfer from Simple to Complex: A Safe and Efficient Reinforcement Learning Framework for Autonomous Driving Decision-Making

A safe and efficient decision-making system is crucial for autonomous vehicles. However, the complexity of driving environments limits the effectiveness of many rule-based and machine learning approaches. Reinforcement Learning (RL), with its robust self-learning capabilities and environmental adaptability, offers a promising solution to these challenges. Nevertheless, safety and efficiency concerns during training hinder its widespread application. To address these concerns, we propose a novel RL framework, Simple to Complex Collaborative Decision (S2CD). First, we rapidly train the teacher model in a lightweight simulation environment. In the more complex and realistic environment, teacher intervenes when the student agent exhibits suboptimal behavior by assessing actions' value to avert dangers. We also introduce an RL algorithm called Adaptive Clipping Proximal Policy Optimization Plus, which combines samples from both teacher and student policies and employs dynamic clipping strategies based on sample importance. This approach improves sample efficiency while effectively alleviating data imbalance. Additionally, we employ the Kullback-Leibler divergence as a policy constraint, transforming it into an unconstrained problem with the Lagrangian method to accelerate the student's learning. Finally, a gradual weaning strategy ensures that the student learns to explore independently over time, overcoming the teacher's limitations and maximizing performance. Simulation experiments in highway lane-change scenarios show that the S2CD framework enhances learning efficiency, reduces training costs, and significantly improves safety compared to state-of-the-art algorithms. This framework also ensures effective knowledge transfer between teacher and student models, even with suboptimal teachers, the student achieves superior performance, demonstrating the robustness and effectiveness of S2CD.

cs.RO↗

arXiv2024

Using Motion Forecasting for Behavior-Based Virtual Reality (VR) Authentication

Task-based behavioral biometric authentication of users interacting in virtual reality (VR) environments enables seamless continuous authentication by using only the motion trajectories of the person's body as a unique signature. Deep learning-based approaches for behavioral biometrics show high accuracy when using complete or near complete portions of the user trajectory, but show lower performance when using smaller segments from the start of the task. Thus, any systems designed with existing techniques are vulnerable while waiting for future segments of motion trajectories to become available. In this work, we present the first approach that predicts future user behavior using Transformer-based forecasting and using the forecasted trajectory to perform user authentication. Our work leverages the notion that given the current trajectory of a user in a task-based environment we can predict the future trajectory of the user as they are unlikely to dramatically shift their behavior since it would preclude the user from successfully completing their task goal. Using the publicly available 41-subject ball throwing dataset of Miller et al. we show improvement in user authentication when using forecasted data. When compared to no forecasting, our approach reduces the authentication equal error rate (EER) by an average of 23.85% and a maximum reduction of 36.14%.

cs.LG↗

arXiv2024

Evaluating Deep Networks for Detecting User Familiarity with VR from Hand Interactions

As VR devices become more prevalent in the consumer space, VR applications are likely to be increasingly used by users unfamiliar with VR. Detecting the familiarity level of a user with VR as an interaction medium provides the potential of providing on-demand training for acclimatization and prevents the user from being burdened by the VR environment in accomplishing their tasks. In this work, we present preliminary results of using deep classifiers to conduct automatic detection of familiarity with VR by using hand tracking of the user as they interact with a numeric passcode entry panel to unlock a VR door. We use a VR door as we envision it to the first point of entry to collaborative virtual spaces, such as meeting rooms, offices, or clinics. Users who are unfamiliar with VR will have used their hands to open doors with passcode entry panels in the real world. Thus, while the user may not be familiar with VR, they would be familiar with the task of opening the door. Using a pilot dataset consisting of 7 users familiar with VR, and 7 not familiar with VR, we acquire highest accuracy of 88.03\% when 6 test users, 3 familiar and 3 not familiar, are evaluated with classifiers trained using data from the remaining 8 users. Our results indicate potential for using user movement data to detect familiarity for the simple yet important task of secure passcode-based access.

cs.HC↗

arXiv2023

End-to-End Latency Optimization of Multi-view 3D Reconstruction for Disaster Response

In order to plan rapid response during disasters, first responder agencies often adopt `bring your own device' (BYOD) model with inexpensive mobile edge devices (e.g., drones, robots, tablets) for complex video analytics applications, e.g., 3D reconstruction of a disaster scene. Unlike simpler video applications, widely used Multi-view Stereo (MVS) based 3D reconstruction applications (e.g., openMVG/openMVS) are exceedingly time consuming, especially when run on such computationally constrained mobile edge devices. Additionally, reducing the reconstruction latency of such inherently sequential algorithms is challenging as unintelligent, application-agnostic strategies can drastically degrade the reconstruction (i.e., application outcome) quality making them useless. In this paper, we aim to design a latency optimized MVS algorithm pipeline, with the objective to best balance the end-to-end latency and reconstruction quality by running the pipeline on a collaborative mobile edge environment. The overall optimization approach is two-pronged where: (a) application optimizations introduce data-level parallelism by splitting the pipeline into high frequency and low frequency reconstruction components and (b) system optimizations incorporate task-level parallelism to the pipelines by running them opportunistically on available resources with online quality control in order to balance both latency and quality. Our evaluation on a hardware testbed using publicly available datasets shows upto ~54% reduction in latency with negligible loss (~4-7%) in reconstruction quality.

cs.CV↗

arXiv2023

FedRight: An Effective Model Copyright Protection for Federated Learning

Federated learning (FL), an effective distributed machine learning framework, implements model training and meanwhile protects local data privacy. It has been applied to a broad variety of practice areas due to its great performance and appreciable profits. Who owns the model, and how to protect the copyright has become a real problem. Intuitively, the existing property rights protection methods in centralized scenarios (e.g., watermark embedding and model fingerprints) are possible solutions for FL. But they are still challenged by the distributed nature of FL in aspects of the no data sharing, parameter aggregation, and federated training settings. For the first time, we formalize the problem of copyright protection for FL, and propose FedRight to protect model copyright based on model fingerprints, i.e., extracting model features by generating adversarial examples as model fingerprints. FedRight outperforms previous works in four key aspects: (i) Validity: it extracts model features to generate transferable fingerprints to train a detector to verify the copyright of the model. (ii) Fidelity: it is with imperceptible impact on the federated training, thus promising good main task performance. (iii) Robustness: it is empirically robust against malicious attacks on copyright protection, i.e., fine-tuning, model pruning, and adaptive attacks. (iv) Black-box: it is valid in the black-box forensic scenario where only application programming interface calls to the model are available. Extensive evaluations across 3 datasets and 9 model structures demonstrate FedRight's superior fidelity, validity, and robustness.

cs.CR↗

arXiv2022

Rethinking the Defense Against Free-rider Attack From the Perspective of Model Weight Evolving Frequency

Federated learning (FL) is a distributed machine learning approach where multiple clients collaboratively train a joint model without exchanging their data. Despite FL's unprecedented success in data privacy-preserving, its vulnerability to free-rider attacks has attracted increasing attention. Existing defenses may be ineffective against highly camouflaged or high percentages of free riders. To address these challenges, we reconsider the defense from a novel perspective, i.e., model weight evolving frequency.Empirically, we gain a novel insight that during the FL's training, the model weight evolving frequency of free-riders and that of benign clients are significantly different. Inspired by this insight, we propose a novel defense method based on the model Weight Evolving Frequency, referred to as WEF-Defense.Specifically, we first collect the weight evolving frequency (defined as WEF-Matrix) during local training. For each client, it uploads the local model's WEF-Matrix to the server together with its model weight for each iteration. The server then separates free-riders from benign clients based on the difference in the WEF-Matrix. Finally, the server uses a personalized approach to provide different global models for corresponding clients. Comprehensive experiments conducted on five datasets and five models demonstrate that WEF-Defense achieves better defense effectiveness than the state-of-the-art baselines.

cs.LG↗

arXiv2022

Eventor: An Efficient Event-Based Monocular Multi-View Stereo Accelerator on FPGA Platform

Event cameras are bio-inspired vision sensors that asynchronously represent pixel-level brightness changes as event streams. Event-based monocular multi-view stereo (EMVS) is a technique that exploits the event streams to estimate semi-dense 3D structure with known trajectory. It is a critical task for event-based monocular SLAM. However, the required intensive computation workloads make it challenging for real-time deployment on embedded platforms. In this paper, Eventor is proposed as a fast and efficient EMVS accelerator by realizing the most critical and time-consuming stages including event back-projection and volumetric ray-counting on FPGA. Highly paralleled and fully pipelined processing elements are specially designed via FPGA and integrated with the embedded ARM as a heterogeneous system to improve the throughput and reduce the memory footprint. Meanwhile, the EMVS algorithm is reformulated to a more hardware-friendly manner by rescheduling, approximate computing and hybrid data quantization. Evaluation results on DAVIS dataset show that Eventor achieves up to $24\times$ improvement in energy efficiency compared with Intel i5 CPU platform.

cs.AR↗

arXiv2019

Verification of the Herschel-Bulkley Model by a Visualization Experiment of a Elastoviscoplastic Yield-Stress Fluid Flow in Straight and Bended Tubes

The elastoviscoplastic yield-stress fluid flows in a horizontal straight tube and a bended tube have been investigated using hydrogen bubble visualization method. The experimental results are used to verify the empirical Herschel-Bulkley model. Both experimental and theoretical investigations well predict the yield-stress fluid flow behaviors. It is found that the significant factors influencing on the predictions of the Herschel-Bulkley model are the yield stress, viscosity and viscoelasticity. For the yield-stress flows in the bended tube, a more delicate constitutive model with consideration of the viscoelastic effects is expected for accurately predicting the flow behaviors.

physics.flu-dyn↗