Qifan Chen

Yuxing Chen, Basem Suleiman, Qifan Chen

Real-world trip planning requires transforming open-ended user requests into executable itineraries under strict spatial, temporal, and budgetary constraints while aligning with user preferences. Existing LLM-based agents struggle with constraint satisfaction, tool coordination, and efficiency, often producing infeasible or costly plans. To address these limitations, we present TriFlow, a progressive multi-agent framework that unifies structured reasoning and language-based flexibility through a three-stage pipeline of retrieval, planning, and governance. By this design, TriFlow progressively narrows the search space, assembles constraint-consistent itineraries via rule-LLM collaboration, and performs bounded iterative refinement to ensure global feasibility and personalisation. Evaluations on TravelPlanner and TripTailor benchmarks demonstrated state-of-the-art results, achieving 91.1% and 97% final pass rates, respectively, with over 10x runtime efficiency improvement compared to current SOTA.

Zhongshu Xu, Yuan Chen, Qifan Chen et al.

We present a numerical method to learn an accurate predictive model for an unknown stochastic dynamical system from its trajectory data. The method seeks to approximate the unknown flow map of the underlying system. It employs the idea of autoencoder to identify the unobserved latent random variables. In our approach, we design an encoding function to discover the latent variables, which are modeled as unit Gaussian, and a decoding function to reconstruct the future states of the system. Both the encoder and decoder are expressed as deep neural networks (DNNs). Once the DNNs are trained by the trajectory data, the decoder serves as a predictive model for the unknown stochastic system. Through an extensive set of numerical examples, we demonstrate that the method is able to produce long-term system predictions by using short bursts of trajectory data. It is also applicable to systems driven by non-Gaussian noises.

Arpit Babbar, Qifan Chen, Hendrik Ranocha

Compact Runge-Kutta (cRK) methods are a class of high order methods for solving hyperbolic conservation laws characterized by their compact stencil including only immediate neighboring finite elements. A Compact Runge-Kutta flux reconstruction (cRKFR) method for solver hyperbolic conservation laws was introduced in [Babbar, A., Chen, Q., Journal of Scientific Computing, 2025] which uses a time average flux formulation to perform evolution using a single numerical flux computation at each step, making it a single stage method. Entropy or kinetic energy preserving numerical fluxes are often used for construction of high order entropy stable or kinetic energy preserving methods for hyperbolic conservation laws, and are known to enhance the robustness of numerical methods for under-resolved simulations. In this work, we show how these fluxes can be incorporated into the cRKFR framework for general hyperbolic equations that consist of fluxes and non-conservative products. We test the effectiveness of this new class of methods through numerical experiments for the compressible Euler equations, magnetohydronamics (MHD) equations and multi-ion MHD equations. It is observed that the application of entropy or kinetic energy preserving fluxes enhances the robustness of the cRKFR methods.

Qifan Chen, Zhongshu Xu, Jinjin Zhang et al.

We present a numerical framework for constructing a targeted digital twin (tDT) that directly models the dynamics of quantities of interest (QoIs) in a full digital twin (DT). The proposed approach employs memory-based flow map learning (FML) to develop a data-driven model of the QoIs using short bursts of trajectory data generated through repeated executions of the full DT. This renders the construction of the FML-based tDT an entirely offline computational process. During online simulation, the learned tDT can efficiently predict and analyze the long-term dynamics of the QoIs without requiring simulations of the full DT system, thereby achieving substantial computational savings. After introducing the general numerical procedure, we demonstrate the construction and predictive capability of the tDT in a computational fluid dynamics (CFD) example: two-dimensional incompressible flow past a cylinder. The QoIs in this problem are the hydrodynamic forces exerted on the cylinder. The resulting tDTs are compact dynamical systems that evolve these forces without explicit knowledge of the underlying flow field. Numerical results show that the tDTs yield accurate long-term predictions of the forces while entirely bypassing full flow simulations.

Qifan Chen, Jin Cui, Cindy Duan et al.

Accurate estimation of postmenstrual age (PMA) at scan is crucial for assessing neonatal development and health. While deep learning models have achieved high accuracy in predicting PMA from brain MRI, they often function as black boxes, offering limited transparency and interpretability in clinical decision support. In this work, we address the dual challenge of accuracy and interpretability by adapting a multimodal large language model (MLLM) to perform both precise PMA prediction and clinically relevant explanation generation. We introduce a parameter-efficient fine-tuning (PEFT) strategy using instruction tuning and Low-Rank Adaptation (LoRA) applied to the Qwen2.5-VL-7B model. The model is trained on four 2D cortical surface projection maps derived from neonatal MRI scans. By employing distinct prompts for training and inference, our approach enables the MLLM to handle a regression task during training and generate clinically relevant explanations during inference. The fine-tuned model achieves a low prediction error with a 95 percent confidence interval of 0.78 to 1.52 weeks, while producing interpretable outputs grounded in developmental features, marking a significant step toward transparent and trustworthy AI systems in perinatal neuroscience.

Yang Lu, Qifan Chen, Simon K. Poon

Process mining is a relatively new subject that builds a bridge between traditional process modeling and data mining. Process discovery is one of the most critical parts of process mining, which aims at discovering process models automatically from event logs. The performance of existing process discovery algorithms can be affected when there are missing activity labels in event logs. Several methods have been proposed to repair missing activity labels, but their accuracy can drop when a large number of activity labels are missing. In this paper, we propose an LSTM-based prediction model to predict the missing activity labels in event logs. The proposed model takes both the prefix and suffix sequences of the events with missing activity labels as input. Additional attributes of event logs are also utilized to improve the performance. Our evaluation of several publicly available datasets shows that the proposed method performed consistently better than existing methods in terms of repairing missing activity labels in event logs.

Yang Lu, Qifan Chen, Simon Poon

Process mining is a relatively new subject which builds a bridge between process modelling and data mining. An exclusive choice in a process model usually splits the process into different branches. However, in some processes, it is possible to switch from one branch to another. The inductive miner guarantees to return sound process models, but fails to return a precise model when there are switch behaviours between different exclusive choice branches due to the limitation of process trees. In this paper, we present a novel extension to the process tree model to support switch behaviours between different branches of the exclusive choice operator and propose a novel extension to the inductive miner to discover sound process models with switch behaviours. The proposed discovery technique utilizes the theory of a previous study to detect possible switch behaviours. We apply both artificial and publicly-available datasets to evaluate our approach. Our results show that our approach can improve the precision of discovered models by 36% while maintaining high fitness values compared to the original inductive miner.

Qifan Chen, Yang Lu, Charmaine S. Tam et al.

Process mining aims to gain knowledge of business processes via the discovery of process models from event logs generated by information systems. The insights revealed from process mining heavily rely on the quality of the event logs. Activities extracted from different data sources or the free-text nature within the same system may lead to inconsistent labels. Such inconsistency would then lead to redundancy in activity labels, which refer to labels that have different syntax but share the same behaviours. Redundant activity labels could introduce unnecessary complexities to the event logs. The identifications of these labels from data-driven process discovery are difficult and rely heavily on human intervention. Neither existing process discovery algorithms nor event data preprocessing techniques can solve such redundancy efficiently. In this paper, we propose a multi-view approach to automatically detect redundant activity labels using not only context-aware features such as control--flow relations and attribute values but also semantic features from the event logs. Our evaluation of several publicly available datasets and a real-life case study demonstrate that our approach can efficiently detect redundant activity labels even with low-occurrence frequencies. The proposed approach can add value to the preprocessing step to generate more representative event logs.

Yang Lu, Qifan Chen, Simon Poon

Business processes are bound to evolve as a form of adaption to changes, and such changes are referred as process drifts. Current process drift detection methods perform well on clean event log data, but the performance can be tremendously affected by noises. A good process drift detection method should be accurate, fast, and robust to noises. In this paper, we propose an offline process drift detection method which identifies each newly observed behaviour as a candidate drift point and checks if the new behaviour can signify significant changes to the original process behaviours. In addition, a bidirectional search method is proposed to accurately locate both the adding and removing of behaviours. The proposed method can accurately detect drift points from event logs and is robust to noises. Both artificial and real-life event logs are used to evaluate our method. Results show that our method can consistently report accurate process drift time while maintaining a reasonably fast detection speed.

Yang Lu, Qifan Chen, Simon Poon

Business processes are continuously evolving in order to adapt to changes due to various factors. One type of process changes are branching frequency changes, which are related to changes in frequencies between different options when there is an exclusive choice. Existing methods either cannot detect such changes or cannot provide accurate and comprehensive results. In this paper, we propose a method which takes both event logs and process models as input and generates a choice sequence for each exclusive choice in the process model. The method then identifies change points based on the choice sequences. We evaluate our method on a real-life event log. Results show that our method can identify branching frequency changes in process models and provide comprehensive results to users.

Qifan Chen, Yang Lu, Simon Poon

Process mining acts as a valuable tool to analyse the behaviour of an organisation by offering techniques to discover, monitor and enhance real processes. The key to process mining is to discovery understandable process models. However, real-life logs can be complex with redundant activities, which share similar behaviour but have different syntax. We show that the existence of such redundant activities heavily affects the quality of discovered process models. Existing approaches filter activities by frequency, which cannot solve problems caused by redundant activities. In this paper, we propose first to discover redundant activities in the log level and, then, use the discovery results to simplify event logs. Two publicly available data sets are used to evaluate the usability of our approach in real-life processes. Our approach can be adopted as a preprocessing step before applying any discovery algorithms to produce simplify models.