Showing 1–11 of 11 results for author: Lee, T K

The Pre-Training Study of Expanded-SPLADE Models on Web Document Titles

Authors: Hiun Kim, Tae Kwan Lee, Taeryun Won

Abstract: Masked Language Modeling (MLM) pre-training is one of the primary ways to initialize Neural Information Retrieval (IR) models prior to retrieval fine-tuning. However, studies show that MLM pre-trained models have limited readiness and transfer learning issues for fine-tuning them into Neural Bi-Encoder models. This paper studies the effect of different pre-training datasets and pre-training option… ▽ More Masked Language Modeling (MLM) pre-training is one of the primary ways to initialize Neural Information Retrieval (IR) models prior to retrieval fine-tuning. However, studies show that MLM pre-trained models have limited readiness and transfer learning issues for fine-tuning them into Neural Bi-Encoder models. This paper studies the effect of different pre-training datasets and pre-training options on the MLM pre-trained models for retrieval fine-tuning. The study focuses on the SPLADE-style model, which uses the MLM layer also at fine-tuning time. More specifically, we experimented with Expanded-SPLADE (ESPLADE) models, a specific instance of SPLADE models, and in-house web document titles are used as datasets. Pre-training, fine-tuning, and evaluation with optional test-time pruning of sparse vectors are conducted. Our observations are three-fold: First, fine-tuned models of higher retrieval effectiveness at both unpruned and most strict pruned settings are mostly pre-trained on a general corpus, and pre-trained with a higher learning rate, showing lower MLM accuracies. Second, in the most strict pruned setting, those models show higher-level retrieval cost and a higher variance in the length of the individual postings list. Third, the repetition of the general pre-training dataset does not have much effect on retrieval effectiveness. The experimentation empirically identifies the potential limitations for aligning MLM pre-training to ESPLADE fine-tuning. Also, the experimentation provides an empirical observation that, at most strict pruned settings, the retrieval effectiveness is better maintained by the higher-level retrieval cost, showing the trade-off relationship between the two in our setting. △ Less

Submitted 2 May, 2026; originally announced May 2026.

Composed Vision-Language Retrieval for Skin Cancer Case Search via Joint Alignment of Global and Local Representations

Authors: Yuheng Wang, Yuji Lin, Jiayue Cai, Z. Jane Wang, Tim K. Lee

Abstract: Medical image retrieval aims to identify clinically relevant lesion cases to support diagnostic decision making, education, and quality control. In practice, retrieval queries often combine a reference lesion image with textual descriptors such as dermoscopic features. We study composed vision-language retrieval for skin cancer, where each query consists of an image to text pair and the database c… ▽ More Medical image retrieval aims to identify clinically relevant lesion cases to support diagnostic decision making, education, and quality control. In practice, retrieval queries often combine a reference lesion image with textual descriptors such as dermoscopic features. We study composed vision-language retrieval for skin cancer, where each query consists of an image to text pair and the database contains biopsy-confirmed, multi-class disease cases. We propose a transformer based framework that learns hierarchical composed query representations and performs joint global-local alignment between queries and candidate images. Local alignment aggregates discriminative regions via multiple spatial attention masks, while global alignment provides holistic semantic supervision. The final similarity is computed through a convex, domain-informed weighting that emphasizes clinically salient local evidence while preserving global consistency. Experiments on the public Derm7pt dataset demonstrate consistent improvements over state-of-the-art methods. The proposed framework enables efficient access to relevant medical records and supports practical clinical deployment. △ Less

Submitted 20 April, 2026; v1 submitted 9 March, 2026; originally announced March 2026.

Efficiency and Effectiveness of SPLADE Models on Billion-Scale Web Document Title

Authors: Taeryun Won, Tae Kwan Lee, Hiun Kim, Hyemin Lee

Abstract: This paper presents a comprehensive comparison of BM25, SPLADE, and Expanded-SPLADE models in the context of large-scale web document retrieval. We evaluate the effectiveness and efficiency of these models on datasets spanning from tens of millions to billions of web document titles. SPLADE and Expanded-SPLADE, which utilize sparse lexical representations, demonstrate superior retrieval performanc… ▽ More This paper presents a comprehensive comparison of BM25, SPLADE, and Expanded-SPLADE models in the context of large-scale web document retrieval. We evaluate the effectiveness and efficiency of these models on datasets spanning from tens of millions to billions of web document titles. SPLADE and Expanded-SPLADE, which utilize sparse lexical representations, demonstrate superior retrieval performance compared to BM25, especially for complex queries. However, these models incur higher computational costs. We introduce pruning strategies, including document-centric pruning and top-k query term selection, boolean query with term threshold to mitigate these costs and improve the models' efficiency without significantly sacrificing retrieval performance. The results show that Expanded-SPLADE strikes the best balance between effectiveness and efficiency, particularly when handling large datasets. Our findings offer valuable insights for deploying sparse retrieval models in large-scale search engines. △ Less

Submitted 27 November, 2025; originally announced November 2025.

The Role of Vocabularies in Learning Sparse Representations for Ranking

Authors: Hiun Kim, Tae Kwan Lee, Taeryun Won

Abstract: Learned Sparse Retrieval (LSR) such as SPLADE has growing interest for effective semantic 1st stage matching while enjoying the efficiency of inverted indices. A recent work on learning SPLADE models with expanded vocabularies (ESPLADE) was proposed to represent queries and documents into a sparse space of custom vocabulary which have different levels of vocabularic granularity. Within this effort… ▽ More Learned Sparse Retrieval (LSR) such as SPLADE has growing interest for effective semantic 1st stage matching while enjoying the efficiency of inverted indices. A recent work on learning SPLADE models with expanded vocabularies (ESPLADE) was proposed to represent queries and documents into a sparse space of custom vocabulary which have different levels of vocabularic granularity. Within this effort, however, there have not been many studies on the role of vocabulary in SPLADE models and their relationship to retrieval efficiency and effectiveness. To study this, we construct BERT models with 100K-sized output vocabularies, one initialized with the ESPLADE pretraining method and one initialized randomly. After finetune on real-world search click logs, we applied logit score-based queries and documents pruning to max size for further balancing efficiency. The experimental result in our evaluation set shows that, when pruning is applied, the two models are effective compared to the 32K-sized normal SPLADE model in the computational budget under the BM25. And the ESPLADE models are more effective than the random vocab model, while having a similar retrieval cost. The result indicates that the size and pretrained weight of output vocabularies play the role of configuring the representational specification for queries, documents, and their interactions in the retrieval engine, beyond their original meaning and purposes in NLP. These findings can provide a new room for improvement for LSR by identifying the importance of representational specification from vocabulary configuration for efficient and effective retrieval. △ Less

Submitted 19 April, 2026; v1 submitted 20 September, 2025; originally announced September 2025.

Comments: fix citation style; add some previous work description at the beginning of section 3;

Integrating Clinical Knowledge Graphs and Gradient-Based Neural Systems for Enhanced Melanoma Diagnosis via the 7-Point Checklist

Authors: Yuheng Wang, Tianze Yu, Jiayue Cai, Sunil Kalia, Harvey Lui, Z. Jane Wang, Tim K. Lee

Abstract: The 7-point checklist (7PCL) is a widely used diagnostic tool in dermoscopy for identifying malignant melanoma by assigning point values to seven specific attributes. However, the traditional 7PCL is limited to distinguishing between malignant melanoma and melanocytic Nevi, and falls short in scenarios where multiple skin diseases with appearances similar to melanoma coexist. To address this limit… ▽ More The 7-point checklist (7PCL) is a widely used diagnostic tool in dermoscopy for identifying malignant melanoma by assigning point values to seven specific attributes. However, the traditional 7PCL is limited to distinguishing between malignant melanoma and melanocytic Nevi, and falls short in scenarios where multiple skin diseases with appearances similar to melanoma coexist. To address this limitation, we propose a novel diagnostic framework that integrates a clinical knowledge-based topological graph (CKTG) with a gradient diagnostic strategy featuring a data-driven weighting system (GD-DDW). The CKTG captures both the internal and external relationships among the 7PCL attributes, while the GD-DDW emulates dermatologists' diagnostic processes, prioritizing visual observation before making predictions. Additionally, we introduce a multimodal feature extraction approach leveraging a dual-attention mechanism to enhance feature extraction through cross-modal interaction and unimodal collaboration. This method incorporates meta-information to uncover interactions between clinical data and image features, ensuring more accurate and robust predictions. Our approach, evaluated on the EDRA dataset, achieved an average AUC of 88.6%, demonstrating superior performance in melanoma detection and feature prediction. This integrated system provides data-driven benchmarks for clinicians, significantly enhancing the precision of melanoma diagnosis. △ Less

Submitted 24 August, 2025; v1 submitted 23 July, 2024; originally announced July 2024.

Comments: The paper was officially accepted for publication in IEEE Transactions on Neural Networks and Learning Systems in August 2025

SSD-KD: A Self-supervised Diverse Knowledge Distillation Method for Lightweight Skin Lesion Classification Using Dermoscopic Images

Authors: Yongwei Wang, Yuheng Wang, Tim K. Lee, Chunyan Miao, Z. Jane Wang

Abstract: Skin cancer is one of the most common types of malignancy, affecting a large population and causing a heavy economic burden worldwide. Over the last few years, computer-aided diagnosis has been rapidly developed and make great progress in healthcare and medical practices due to the advances in artificial intelligence. However, most studies in skin cancer detection keep pursuing high prediction acc… ▽ More Skin cancer is one of the most common types of malignancy, affecting a large population and causing a heavy economic burden worldwide. Over the last few years, computer-aided diagnosis has been rapidly developed and make great progress in healthcare and medical practices due to the advances in artificial intelligence. However, most studies in skin cancer detection keep pursuing high prediction accuracies without considering the limitation of computing resources on portable devices. In this case, knowledge distillation (KD) has been proven as an efficient tool to help improve the adaptability of lightweight models under limited resources, meanwhile keeping a high-level representation capability. To bridge the gap, this study specifically proposes a novel method, termed SSD-KD, that unifies diverse knowledge into a generic KD framework for skin diseases classification. Our method models an intra-instance relational feature representation and integrates it with existing KD research. A dual relational knowledge distillation architecture is self-supervisedly trained while the weighted softened outputs are also exploited to enable the student model to capture richer knowledge from the teacher model. To demonstrate the effectiveness of our method, we conduct experiments on ISIC 2019, a large-scale open-accessed benchmark of skin diseases dermoscopic images. Experiments show that our distilled lightweight model can achieve an accuracy as high as 85% for the classification tasks of 8 different skin diseases with minimal parameters and computing requirements. Ablation studies confirm the effectiveness of our intra- and inter-instance relational knowledge integration strategy. Compared with state-of-the-art knowledge distillation techniques, the proposed method demonstrates improved performances for multi-diseases classification on the large-scale dermoscopy database. △ Less

Submitted 29 March, 2022; v1 submitted 22 March, 2022; originally announced March 2022.

Comments: 14 pages, 5 figures

Understanding the Dynamics of Drivers' Locations for Passengers Pickup Performance: A Case Study

Authors: Punit Rathore, Ali Zonoozi, Omid Geramifard, Tan Kian Lee

Abstract: With the emergence of e-hailing taxi services, a growing number of scholars have attempted to analyze the taxi trips data to gain insights from drivers' and passengers' flow patterns and understand different dynamics of urban public transportation. Existing studies are limited to passengers' location analysis e.g., pick-up and drop-off points, in the context of maximizing the profits or better man… ▽ More With the emergence of e-hailing taxi services, a growing number of scholars have attempted to analyze the taxi trips data to gain insights from drivers' and passengers' flow patterns and understand different dynamics of urban public transportation. Existing studies are limited to passengers' location analysis e.g., pick-up and drop-off points, in the context of maximizing the profits or better managing the resources for service providers. Moreover, taxi drivers' locations at the time of pick-up requests and their pickup performance in the spatial-temporal domain have not been explored. In this paper, we analyze drivers' and passengers' locations at the time of booking request in the context of drivers' pick-up performances. To facilitate our analysis, we implement a modified and extended version of a co-clustering technique, called sco-iVAT, to obtain useful clusters and co-clusters from big relational data, derived from booking records of Grab ride-hailing service in Singapore. We also explored the possibility of predicting timely pickup for a given booking request, without using entire trajectories data. Finally, we devised two scoring mechanisms to compute pickup performance score for all driver candidates for a booking request. These scores could be integrated into a booking assignment model to prioritize top-performing drivers for passenger pickups. △ Less

Submitted 9 September, 2020; originally announced September 2020.

Comments: Submitted to IEEE Transactions on Inelligent Transportation Systems

Vision-based techniques for gait recognition

Authors: Tracey K. M. Lee, Mohammed Belkhatir, Saeid Sanei

Abstract: Global security concerns have raised a proliferation of video surveillance devices. Intelligent surveillance systems seek to discover possible threats automatically and raise alerts. Being able to identify the surveyed object can help determine its threat level. The current generation of devices provide digital video data to be analysed for time varying features to assist in the identification pro… ▽ More Global security concerns have raised a proliferation of video surveillance devices. Intelligent surveillance systems seek to discover possible threats automatically and raise alerts. Being able to identify the surveyed object can help determine its threat level. The current generation of devices provide digital video data to be analysed for time varying features to assist in the identification process. Commonly, people queue up to access a facility and approach a video camera in full frontal view. In this environment, a variety of biometrics are available - for example, gait which includes temporal features like stride period. Gait can be measured unobtrusively at a distance. The video data will also include face features, which are short-range biometrics. In this way, one can combine biometrics naturally using one set of data. In this paper we survey current techniques of gait recognition and modelling with the environment in which the research was conducted. We also discuss in detail the issues arising from deriving gait data, such as perspective and occlusion effects, together with the associated computer vision challenges of reliable tracking of human movement. Then, after highlighting these issues and challenges related to gait processing, we proceed to discuss the frameworks combining gait with other biometrics. We then provide motivations for a novel paradigm in biometrics-based human recognition, i.e. the use of the fronto-normal view of gait as a far-range biometrics combined with biometrics operating at a near distance. △ Less

Submitted 30 April, 2020; originally announced May 2020.

Compounding the Performance Improvements of Assembled Techniques in a Convolutional Neural Network

Authors: Jungkyu Lee, Taeryun Won, Tae Kwan Lee, Hyemin Lee, Geonmo Gu, Kiho Hong

Abstract: Recent studies in image classification have demonstrated a variety of techniques for improving the performance of Convolutional Neural Networks (CNNs). However, attempts to combine existing techniques to create a practical model are still uncommon. In this study, we carry out extensive experiments to validate that carefully assembling these techniques and applying them to basic CNN models (e.g. Re… ▽ More Recent studies in image classification have demonstrated a variety of techniques for improving the performance of Convolutional Neural Networks (CNNs). However, attempts to combine existing techniques to create a practical model are still uncommon. In this study, we carry out extensive experiments to validate that carefully assembling these techniques and applying them to basic CNN models (e.g. ResNet and MobileNet) can improve the accuracy and robustness of the models while minimizing the loss of throughput. Our proposed assembled ResNet-50 shows improvements in top-1 accuracy from 76.3\% to 82.78\%, mCE from 76.0\% to 48.9\% and mFR from 57.7\% to 32.3\% on ILSVRC2012 validation set. With these improvements, inference throughput only decreases from 536 to 312. To verify the performance improvement in transfer learning, fine grained classification and image retrieval tasks were tested on several public datasets and showed that the improvement to backbone network performance boosted transfer learning performance significantly. Our approach achieved 1st place in the iFood Competition Fine-Grained Visual Recognition at CVPR 2019, and the source code and trained models are available at https://github.com/clovaai/assembled-cnn △ Less

Submitted 13 March, 2020; v1 submitted 17 January, 2020; originally announced January 2020.

Comments: 9 pages, 2 figures, 18 tables

A Benchmark on Tricks for Large-scale Image Retrieval

Authors: Byungsoo Ko, Minchul Shin, Geonmo Gu, HeeJae Jun, Tae Kwan Lee, Youngjoon Kim

Abstract: Many studies have been performed on metric learning, which has become a key ingredient in top-performing methods of instance-level image retrieval. Meanwhile, less attention has been paid to pre-processing and post-processing tricks that can significantly boost performance. Furthermore, we found that most previous studies used small scale datasets to simplify processing. Because the behavior of a… ▽ More Many studies have been performed on metric learning, which has become a key ingredient in top-performing methods of instance-level image retrieval. Meanwhile, less attention has been paid to pre-processing and post-processing tricks that can significantly boost performance. Furthermore, we found that most previous studies used small scale datasets to simplify processing. Because the behavior of a feature representation in a deep learning model depends on both domain and data, it is important to understand how model behave in large-scale environments when a proper combination of retrieval tricks is used. In this paper, we extensively analyze the effect of well-known pre-processing, post-processing tricks, and their combination for large-scale image retrieval. We found that proper use of these tricks can significantly improve model performance without necessitating complex architecture or introducing loss, as confirmed by achieving a competitive result on the Google Landmark Retrieval Challenge 2019. △ Less

Submitted 23 April, 2020; v1 submitted 27 July, 2019; originally announced July 2019.

Convolution with Logarithmic Filter Groups for Efficient Shallow CNN

Authors: Tae Kwan Lee, Wissam J. Baddar, Seong Tae Kim, Yong Man Ro

Abstract: In convolutional neural networks (CNNs), the filter grouping in convolution layers is known to be useful to reduce the network parameter size. In this paper, we propose a new logarithmic filter grouping which can capture the nonlinearity of filter distribution in CNNs. The proposed logarithmic filter grouping is installed in shallow CNNs applicable in a mobile application. Experiments were perform… ▽ More In convolutional neural networks (CNNs), the filter grouping in convolution layers is known to be useful to reduce the network parameter size. In this paper, we propose a new logarithmic filter grouping which can capture the nonlinearity of filter distribution in CNNs. The proposed logarithmic filter grouping is installed in shallow CNNs applicable in a mobile application. Experiments were performed with the shallow CNNs for classification tasks. Our classification results on Multi-PIE dataset for facial expression recognition and CIFAR-10 dataset for object classification reveal that the compact CNN with the proposed logarithmic filter grouping scheme outperforms the same network with the uniform filter grouping in terms of accuracy and parameter efficiency. Our results indicate that the efficiency of shallow CNNs can be improved by the proposed logarithmic filter grouping. △ Less

Submitted 14 September, 2017; v1 submitted 31 July, 2017; originally announced July 2017.

Comments: 8 pages, 4 figures, 3 tables. Changes in abstract, result representations and typo corrections