Publications
Publications by categories in reversed chronological order.
2026
- Pattern-Guided Forecasting Framework for Metal Price Prediction with Grouping Decomposed SeriesDongbin Kim, Jaewook Lee, and Hoki KimFinancial Innovation, 2026
Accurate forecasting of precious metal prices is increasingly critical in modern financial markets, as these metals function as industrial commodities and as strategic financial instruments for portfolio diversification and risk management. Although recent advances in financial technology have produced a range of forecasting approaches from traditional econometric methods to sophisticated deep learning models—the complex dynamics of metal prices continue to challenge existing methodologies. This paper introduces a significant innovation in financial forecasting by revealing and leveraging previously unrecognized pattern relationships in decomposed time series data. Our comprehensive analysis of metal price dynamics reveals distinct grouped patterns in decomposed time series components, challenging the conventional assumption of independence in current forecasting methods. Based on these insights, we propose the pattern-guided forecasting framework (PGFF), which enhances forecasting accuracy by leveraging cross-dimensional pattern relationships in decomposed time series. Our framework employs a novel two-stage approach: first, categorizing decomposed time series based on their temporal characteristics and autocorrelation patterns; then, implementing cross-dimensional forecasting to capture complex market dynamics. Empirical analysis of four major precious metals demonstrates that PGFF consistently outperforms existing forecasting frameworks, offering significant implications for investment decision-making and portfolio management in modern financial markets.
@article{kim2026pgff, title = {Pattern-Guided Forecasting Framework for Metal Price Prediction with Grouping Decomposed Series}, author = {Kim, Dongbin and Lee, Jaewook and Kim, Hoki}, journal = {Financial Innovation}, volume = {12}, number = {1}, pages = {45}, year = {2026}, doi = {10.1186/s40854-025-00833-5}, } - Local Geometry Attention for Time Series Forecasting under Realistic CorruptionsDongbin Kim, Youngjoo Park, Woojin Jeong, and 1 more authorIn The Fourteenth International Conference on Learning Representations, 2026
Transformers have demonstrated strong performance in time series forecasting, yet they often fail to capture the intrinsic structure of temporal data, making them susceptible to real-world noise and anomalies. Unlike in vision or language, the local geometry of temporal patterns is a critical feature in time series forecasting, but it is frequently disrupted by corruptions. In this work, we address this gap with two key contributions. First, we propose Local Geometry Attention (LGA), a novel attention mechanism theoretically grounded in local Gaussian process theory. LGA adapts to the intrinsic data geometry by learning query-specific distance metrics, enabling it to model complex temporal dependencies and enhance resilience to noise. Second, we introduce TSRBench, the first comprehensive benchmark for evaluating forecasting robustness under realistic, statistically-grounded corruptions. Experiments on TSRBench show that LGA significantly reduces performance degradation, consistently outperforming both Transformer and linear model. These results establish a foundation for developing robust time series models that can be deployed in real-world applications where data quality is not guaranteed. Our code is available at: https://github.com/dongbeank/LGA.
@inproceedings{kim2026lga, title = {Local Geometry Attention for Time Series Forecasting under Realistic Corruptions}, author = {Kim, Dongbin and Park, Youngjoo and Jeong, Woojin and Lee, Jaewook}, booktitle = {The Fourteenth International Conference on Learning Representations}, year = {2026}, } - A Locally Tokenized Generative Model for Robust Time-Series WatermarkingDongbin Kim, Geonwoo Shin, Yujin Choi, and 2 more authorsAdvances in Neural Information Processing Systems, 2026
Watermarking is a central tool for provenance in generative models, yet its application to multivariate time series remains hindered by reliability failures under post-editing attacks. We show that existing detectors, which rely on globally coupled re-encoding, suffer from bidirectional drift of the null distribution: post-editing attacks can shift the z-score of non-watermarked samples in either direction, invalidating clean-calibrated thresholds. We argue that this instability is a property of the re-encoding, and that reliable detection requires each recovered unit to depend only on a bounded temporal neighborhood. Guided by this principle, we propose L-VQVAE, a generative model in which each discrete token is produced from a short contiguous window, and LVQMark, a watermarking method over this token space that combines logit-bias injection with robust re-encoding for attack-time detection. Experiments on four benchmarks spanning finance, energy, and neuroimaging show that our approach preserves generation quality while stabilizing both detection power and false-positive behavior under post-editing attacks.
@article{kim2026lvqmark, title = {A Locally Tokenized Generative Model for Robust Time-Series Watermarking}, author = {Kim, Dongbin and Shin, Geonwoo and Choi, Yujin and Park, Soyeon and Lee, Jaewook}, journal = {Advances in Neural Information Processing Systems}, volume = {39}, year = {2026}, } - Discretizing Continuous Time Series for Imputation with Masked Diffusion TrainingDongbin Kim, Seungyun Lee, Geonwoo Shin, and 1 more authorAdvances in Neural Information Processing Systems, 2026
Time series imputation is a crucial area for reliable time series analysis, yet it remains challenging due to the complex temporal dynamics and noise of real-world data. Existing approaches, however, exhibit two limitations: missing and observed values are embedded within the same representation space without explicit structural separation, and continuous diffusion-based methods are trained to predict added noise rather than the original signal. To address these, we propose the Masked Diffusion Time-series Imputation Model (MDTIM), which leverages the training paradigm of masked diffusion model for imputation tasks. The MASK token is structurally orthogonal to valid observations, and the model directly predicts the original values, naturally aligning both the representation and the learning objective with the imputation task. To bridge the gap between discrete masked diffusion and the continuous, ordinal nature of time series, we further introduce Stochastic Discretization, which maps continuous values to ordinal-aware tokens while preserving continuous dynamics. Our experiments on diverse benchmarks confirm that MDTIM achieves superior robustness and scalability, consistently outperforming state-of-the-art deterministic and generative baselines across various missing scenarios.
@article{kim2026mdtim, title = {Discretizing Continuous Time Series for Imputation with Masked Diffusion Training}, author = {Kim, Dongbin and Lee, Seungyun and Shin, Geonwoo and Lee, Jaewook}, journal = {Advances in Neural Information Processing Systems}, volume = {39}, year = {2026}, } - Leveraging Pathology Co-occurrence for Test-Time Adaptation in Chest X-Ray DiagnosisWoojin Jeong, Yujin Choi, Dongbin Kim, and 2 more authorsIn Medical Image Computing and Computer Assisted Intervention – MICCAI 2026, 2026
Medical imaging models often degrade when deployed at new clinical sites due to differences in imaging equipment, protocols, and patient populations. Test-time adaptation (TTA) addresses this by updating a pretrained model using only unlabeled target data, without access to source data. However, existing TTA methods were designed for single-label classification on natural image benchmarks, minimizing entropy uniformly across all samples without considering label dependencies. This overlooks a key property of multi-label medical imaging: pathologies do not occur independently but exhibit structured co-occurrence patterns. In this work, we propose Co-occurrence Weighted Adaptation (CoWA), which leverages disease co-occurrence patterns as a reliability signal for adaptation. CoWA estimates label co-occurrence structure from model predictions and downweights samples that deviate from expected patterns, enabling adaptation to rely more on consistent predictions while reducing the impact of noisy ones. We evaluate CoWA on chest X-ray benchmarks under domain shifts and demonstrate consistent improvements over established baselines.
@inproceedings{jeong2026cowa, title = {Leveraging Pathology Co-occurrence for Test-Time Adaptation in Chest X-Ray Diagnosis}, author = {Jeong, Woojin and Choi, Yujin and Kim, Dongbin and Park, Soyeon and Lee, Jaewook}, booktitle = {Medical Image Computing and Computer Assisted Intervention -- MICCAI 2026}, series = {Lecture Notes in Computer Science}, publisher = {Springer Nature Switzerland}, pages = {326--336}, year = {2026}, doi = {10.1007/978-3-032-38072-2_32}, }
2025
- Temporal consistency ensemble empirical mode decomposition for forecasting practical metal priceYujin Choi, Dongbin Kim, and Jaewook LeeEngineering Applications of Artificial Intelligence, 2025
Accurately forecasting metal prices is critical to economic, industrial, and energy markets. However, traditional time series models often rely on future data, limiting their real-world applicability. This study found that decomposition methods utilizing future data inflate model performance, and the high accuracy of forecasting models is mainly due to these unrealistic assumptions. In this paper, we propose a novel Temporal Consistency Ensemble Empirical Mode Decomposition (TC-EEMD) method d...
@article{choi2025temporal, title = {Temporal consistency ensemble empirical mode decomposition for forecasting practical metal price}, author = {Choi, Yujin and Kim, Dongbin and Lee, Jaewook}, journal = {Engineering Applications of Artificial Intelligence}, volume = {158}, pages = {111490}, year = {2025}, doi = {10.1016/j.engappai.2025.111490}, }
2024
- Are Self-Attentions Effective for Time Series Forecasting?Dongbin Kim, Jinseong Park, Jaewook Lee, and 1 more authorAdvances in Neural Information Processing Systems, 2024
Time series forecasting is crucial for applications across multiple domains and various scenarios. Although Transformers have dramatically advanced the landscape of forecasting, their effectiveness remains debated. Recent findings have indicated that simpler linear models might outperform complex Transformer-based approaches, highlighting the potential for more streamlined architectures. In this paper, we shift the focus from evaluating the overall Transformer architecture to specifically examining the effectiveness of self-attention for time series forecasting. To this end, we introduce a new architecture, Cross-Attention-only Time Series transformer (CATS), that rethinks the traditional transformer framework by eliminating self-attention and leveraging cross-attention mechanisms instead. By establishing future horizon-dependent parameters as queries and enhanced parameter sharing, our model not only improves long-term forecasting accuracy but also reduces the number of parameters and memory usage. Extensive experiment across various datasets demonstrates that our model achieves superior performance with the lowest mean squared error and uses fewer parameters compared to existing models. The implementation of our model is available at: https://github.com/dongbeank/CATS.
@article{kim2024self, title = {Are Self-Attentions Effective for Time Series Forecasting?}, author = {Kim, Dongbin and Park, Jinseong and Lee, Jaewook and Kim, Hoki}, journal = {Advances in Neural Information Processing Systems}, volume = {37}, pages = {114180--114209}, year = {2024}, }