arXiv subjects

Junqi Li

Publications and source records attributed to Junqi Li.

7 recordsLinked to original sources

arXiv2026

A visual large language foundational model for medical image recognition using clinician-oriented social media

Large language models (LLMs) have demonstrated strong capabilities across diverse domains, showing considerable potential in medicine. However, their application in medical settings remains limited by the scarcity of visual question answering (VQA) datasets that capture clinical reasoning and explicit image-text alignment. Here, we leverage de-identified medical images and expert commentaries shared on clinician-oriented social media. By combining an advanced LLM with clinician-in-the-loop verification, we established a rigorous pipeline to construct ThoughtMed-1M, a long-form medical VQA dataset containing over one million VQA pairs and designed to capture structured clinical logic and medical image-text alignment. To demonstrate its utility, we developed a FOundational LLM Trained on ThoughtMed-1M (FOLTMed). FOLTMed achieved state-of-the-art performance across 42 medical VQA benchmark datasets, with a macro accuracy of 85.4%, and generated more clinically coherent responses on the ThoughtMed-1M test set. It outperformed state-of-the-art models by 3--5% across factuality and similarity metrics, highlighting a scalable paradigm for advancing research on clinically grounded multimodal LLMs.

cs.AI↗

arXiv2025

A simulation platform calibration method for automated vehicle evaluation: accurate on both vehicle level and traffic flow level

Simulation testing is a fundamental approach for evaluating automated vehicles (AVs). To ensure its reliability, it is crucial to accurately replicate interactions between AVs and background traffic, which necessitates effective calibration. However, existing calibration methods often fall short in achieving this goal. To address this gap, this study introduces a simulation platform calibration method that ensures high accuracy at both the vehicle and traffic flow levels. The method offers several key features:(1) with the capability of calibration for vehicle-to-vehicle interaction; (2) with accuracy assurance; (3) with enhanced efficiency; (4) with pipeline calibration capability. The proposed method is benchmarked against a baseline with no calibration and a state-of-the-art calibration method. Results show that it enhances the accuracy of interaction replication by 83.53% and boosts calibration efficiency by 76.75%. Furthermore, it maintains accuracy across both vehicle-level and traffic flow-level metrics, with an improvement of 51.9%. Notably, the entire calibration process is fully automated, requiring no human intervention.

cs.RO↗

arXiv2025

PortAgent: LLM-driven Vehicle Dispatching Agent for Port Terminals

Vehicle Dispatching Systems (VDSs) are critical to the operational efficiency of Automated Container Terminals (ACTs). However, their widespread commercialization is hindered due to their low transferability across diverse terminals. This transferability challenge stems from three limitations: high reliance on port operational specialists, a high demand for terminal-specific data, and time-consuming manual deployment processes. Leveraging the emergence of Large Language Models (LLMs), this paper proposes PortAgent, an LLM-driven vehicle dispatching agent that fully automates the VDS transferring workflow. It bears three features: (1) no need for port operations specialists; (2) low need of data; and (3) fast deployment. Specifically, specialist dependency is eliminated by the Virtual Expert Team (VET). The VET collaborates with four virtual experts, including a Knowledge Retriever, Modeler, Coder, and Debugger, to emulate a human expert team for the VDS transferring workflow. These experts specialize in the domain of terminal VDS via a few-shot example learning approach. Through this approach, the experts are able to learn VDS-domain knowledge from a few VDS examples. These examples are retrieved via a Retrieval-Augmented Generation (RAG) mechanism, mitigating the high demand for terminal-specific data. Furthermore, an automatic VDS design workflow is established among these experts to avoid extra manual interventions. In this workflow, a self-correction loop inspired by the LLM Reflexion framework is created

cs.AI↗

arXiv2025

OPOR-Bench: Evaluating Large Language Models on Online Public Opinion Report Generation

Online Public Opinion Reports consolidate news and social media for timely crisis management by governments and enterprises. While large language models have made automated report generation technically feasible, systematic research in this specific area remains notably absent, particularly lacking formal task definitions and corresponding benchmarks. To bridge this gap, we define the Automated Online Public Opinion Report Generation (OPOR-GEN) task and construct OPOR-BENCH, an event-centric dataset covering 463 crisis events with their corresponding news articles, social media posts, and a reference summary. To evaluate report quality, we propose OPOR-EVAL, a novel agent-based framework that simulates human expert evaluation by analyzing generated reports in context. Experiments with frontier models demonstrate that our framework achieves high correlation with human judgments. Our comprehensive task definition, benchmark dataset, and evaluation framework provide a solid foundation for future research in this critical domain.

cs.CL↗

arXiv2019

A Novel DDoS Attack Detection Method Using Optimized Generalized Multiple Kernel Learning

Distributed Denial of Service (DDoS) attack has become one of the most destructive network attacks which can pose a mortal threat to Internet security. Existing detection methods can not effectively detect early attacks. In this paper, we propose a detection method of DDoS attacks based on generalized multiple kernel learning (GMKL) combining with the constructed parameter R. The super-fusion feature value (SFV) and comprehensive degree of feature (CDF) are defined to describe the characteristic of attack flow and normal flow. A method for calculating R based on SFV and CDF is proposed to select the combination of kernel function and regularization paradigm. A DDoS attack detection classifier is generated by using the trained GMKL model with R parameter. The experimental results show that kernel function and regularization parameter selection method based on R parameter reduce the randomness of parameter selection and the error of model detection, and the proposed method can effectively detect DDoS attacks in complex environments with higher detection rate and lower error rate.

cs.CR↗

arXiv2019

Adaptive DDoS attack detection method based on multiple-kernel learning

Distributed denial of service (DDoS) attacks have caused huge economic losses to society. They have become one of the main threats to Internet security. Most of the current detection methods based on a single feature and fixed model parameters cannot effectively detect early DDoS attacks in cloud and big data environment. In this paper, an adaptive DDoS attack detection method (ADADM) based on multiple kernel learning (MKL) is proposed. Based on the burstiness of DDoS attack flow, the distribution of addresses and the interactivity of communication, we define five features to describe the network flow characteristic. Based on the ensemble learning framework, the weight of each dimension is adaptively adjusted by increasing the inter-class mean with a gradient ascent and reducing the intra-class variance with a gradient descent, and the classifier is established to identify an early DDoS attack by training simple multiple kernel learning (SMKL) models with two characteristics including inter-class mean squared difference growth (M-SMKL) and intra-class variance descent (S-SMKL). The sliding window mechanism is used to coordinate the S-SMKL and M-SMKL to detect the early DDoS attack. The experimental results indicate that this method can detect DDoS attacks early and accurately.

cs.CR↗

arXiv2018

An empirical study of public data quality problems in cross project defect prediction

Background: Two public defect data, including Jureczko and NASA datasets, have been widely used in cross project defect prediction (CPDP). The quality of defect data have been reported as an important factor influencing the defect prediction performance and Shepperd et al. have researched the data quality problems in NASA datasets. However, up to now, there is no research focusing on the quality problems of Jureczko datasets which are most widely used in CPDP. Aims: In this paper, we intend to investigate the problems of identical and inconsistent cases in Jureczko datasets and validate whether removing these problematic cases will make a difference to defect prediction performance in CPDP. Method: The problems of identical and inconsistent cases are reported from two aspects, respectively in each individual dataset and in a pair of datasets from different releases of a software project. Then a cleaned version of Jureczko datasets is provided by removing duplicate and inconsistent cases. Finally three training data selection methods are employed to compare the defect prediction performance of cleaned datasets with that of original datasets. Results: The experimental results in terms of AUC and F-Measure show that most datasets obtain very different defect prediction performance. Conclusions: It is very necessary to study the data quality problems in CPDP and the cleaned Jureczko datasets may provide more reliable defect prediction performance in CPDP.

cs.SE↗