研究者詳細

顔写真

ホウ　ドンヂエ
侯　登哲
Dengzhe Hou
所属
大学院情報科学研究科　国際交流推進室
職名
特任助教（研究）
学位

  • 学士 （同済大学）

  • 修士 （東北大学）

  • 博士 （東北大学）

e-Rad 研究者番号
71039427
プロフィール

東北大学大学院情報科学研究科 助教（山田研究室）。2026年4月より未踏スケールデータアナリティクスセンター（UDAC）を兼務。2025–2026年度 日本学術振興会特別研究員（DC2）。AI4Cog（認知科学のためのAI）、Cog4AI（AIのための認知科学）、AI4Brain（神経科学のためのAI）、Brain4AI（AIのための神経科学）の4つの方向から研究を進めており、信頼性の高い脳波デコーディング、自発的注意、言語モデル・意思決定モデルの認知評価、脳とAIの表現アライメントに取り組んでいる。2026年 東北大学 博士（情報科学）。

経歴 4

  • 2026年4月 ～ 継続中
    東北大学　東北大学未踏スケールデータアナリティクスセンター　助教

  • 2026年4月 ～ 継続中
    東北大学　大学院情報科学研究科　助教

  • 2025年4月 ～ 2026年3月
    日本学術振興会　特別研究員（DC2）

  • 2024年8月 ～ 2025年1月
    ハーバード大学医学大学院／マサチューセッツ総合病院　神経内科　訪問研究員

委員歴 1

  • Interdisciplinary Information Sciences　編集委員

    2026年7月 ～ 継続中

研究キーワード 9

  • Brain4AI（AIのための神経科学）

  • AI4Brain（神経科学のためのAI）

  • Cog4AI（AIのための認知科学）

  • AI4Cog（認知科学のためのAI）

  • 脳波デコーディング

  • 大規模言語モデル

  • 脳波（EEG）

  • 認知脳科学

  • 自発的注意

研究分野 2

  • 情報通信 / 知能情報学 /

  • ライフサイエンス / 認知脳科学 /

受賞 7

  1. Silver Medal, ROGII - Wellbore Geology Prediction (89/6125 teams)

    2026年8月　Kaggle

  2. Bronze Medal, CSIRO Image2Biomass Prediction (355/3805 teams)

    2026年1月　Kaggle

  3. Bronze Medal, Santa 2024: The Perplexity Permutation Puzzle (148/1514 teams)

    2025年1月　Kaggle

  4. 第32回博士学生発表会 Best Presentation Award

    2024年12月　東北大学大学院情報科学研究科

  5. Student Travel Award

    2023年8月　European Conference on Visual Perception (ECVP)

  6. 外国人留学生奨学金（2021年4月–2023年3月）

    2021年4月　公益財団法人亀井記念財団

  7. 新入生奨学金

    2016年9月　同済大学

︎全件表示 ︎最初の5件までを表示

論文 8

  1. TimePre: Bridging Accuracy, Efficiency, and Stability in Probabilistic Time-Series Forecasting 査読有り

    Lingyu Jiang, Lingyu Xu, Peiran Li, Dengzhe Hou, Qianwen Ge, Dingyi Zhuang, Shuo Xing, Wenjing Chen, Xiangbo Gao, Ting-Hsuan Chen, Xueying Zhan, Xin Zhang, Ziming Zhang, Zhengzhong Tu, Michael Zielewski, Kazunori Yamada, Fangzhou Lin

    Transactions on Machine Learning Research　2026年8月

  2. A compact Kolmogorov–Arnold network mixer for long-term time series forecasting 査読有り

    Lingyu Jiang, Dengzhe Hou, Yuping Wang, Yao Su, Shuo Xing, Wenjing Chen, Xin Zhang, Zhengzhong Tu, Ziming Zhang, Fangzhou Lin, Michael Zielewski, Kazunori Yamada

    Scientific Reports　2026年7月

    DOI： 10.1038/s41598-026-59667-5 　

  3. Self-Evolving Agent Engineering for Healthcare: Methodologies and Applications

    Dengzhe Hou, Zihao Wu, Yuwen Zeng, Lingyu Jiang, Fangzhou Lin, Kazunori Yamada

    2026年5月22日

    DOI： 10.20944/preprints202605.1547.v1 　

  4. PathCal: State-Aware Reflection-Marker Calibration for Efficient Reasoning

    Lingyu Jiang, Zirui Li, Shuo Xing, Peiran Li, Tsubasa Takahashi, Dengzhe Hou, Zhengzhong Tu, Kazunori Yamada, Fangzhou Lin

    CoRR　abs/2605.23074　2026年5月21日

    DOI： 10.48550/arXiv.2605.23074 　

    詳細を見る 詳細を閉じる

    The emergence of Large Reasoning Language Models (LRMs) has paved the way for tackling complex reasoning tasks through test-time scaling by generating long-form Chain-of-Thought (CoT) trajectories during inference. Meanwhile, these trajectories often contain explicit reflection markers such as ``wait'', ``but'', and ``alternatively'', signaling hesitation, revision, and the consideration of alternative explorations, respectively. Recent studies on test-time control leverage such markers as lightweight handles for steering reasoning, typically treating them as a single coarse-grained category rather than distinguishing their distinct functional roles. In this paper, we conduct type-wise suppression and fixed-prefix intervention, revealing that reflection markers differ not only in their functional roles but also in when they exert the greatest influence. Specifically, different marker classes affect accuracy and generation length in distinct ways, and marker choices are most consequential before the model settles into a stable reasoning trajectory. Motivated by these findings, we introduce PathCal, a novel training-free decoding controller that calibrates reasoning paths by distinguishing marker types and intervening only at locally uncertain states. At each decoding step, PathCal utilizes the distribution over reflection-markers to estimate local competition between maintaining the current reasoning trajectory and initiating a competing branch, and softly rebalances marker logits when competing-branch evidence becomes excessive. Experiments across six reasoning benchmarks demonstrate that PathCal achieves a better efficiency--performance trade-off, improving or preserving accuracy while reducing generation length, without relying on external verifiers or additional sampling.

  5. Subject-Specific Analysis of Self-Initiated Attention Shifts from EEG with Controlled Internal and External Attention Conditions 査読有り

    Yuwen Zeng, Dengzhe Hou, Zhang Zhang, Sai Sun, Yongsong Huang, Chia-huei Tseng, Satoshi Shioiri

    IEEE International Conference on Systems, Man, and Cybernetics (SMC 2026)　2026年5月18日

    DOI： 10.48550/arXiv.2605.18251 　

    詳細を見る 詳細を閉じる

    Self-initiated attention shifts play a critical role in voluntary behavior but are difficult to study due to the absence of explicit temporal markers. While previous studies have examined their neural correlates, it remains unclear how multi-dimensional electroencephalography (EEG) features contribute to their characterization within an interpretable computational framework. In this study, we build on an experimental paradigm developed in our previous work, which enables controlled comparison between task-constrained self-initiated shifts and externally instructed shifts under identical visual stimulation. Within this setting, we investigate whether preparatory EEG activity can distinguish these two types of attention shifts. We adopt a machine learning-based approach and conduct two complementary analyses: (1) a performance-oriented assessment of frequency-specific topographic patterns, and (2) a model-based feature attribution analysis using SHapley Additive exPlanations (SHAP). These analyses provide a structured view of how spectral features across regions of interest contribute to model behavior. Our results demonstrate reliable within-subject classification performance, indicating that preparatory EEG activity contains subject-specific discriminative information within this paradigm. The analysis shows that higher-frequency bands and frontal regions contribute strongly to model decisions, although such contributions should be interpreted cautiously due to the potential influence of non-neural artifacts in high-frequency EEG signals. Overall, this work highlights the value of interpretable machine learning for analyzing subject-specific EEG signal patterns in a controlled experimental setting, with potential applications in personalized and asynchronous brain-machine interface systems.

  6. Task-constrained self-initiated attention shifts are indexed by frontal-midline theta ramping 査読有り

    Dengzhe Hou, Sai Sun, Yasuhiro Hatori, Chia-huei Tseng, Satoshi Shioiri

    FRONTIERS IN HUMAN NEUROSCIENCE　19　1708257　2025年12月

    DOI： 10.3389/FNHUM.2025.1708257 　

  7. Commonality of neuronal coherence for motor skill acquisition and interlimb transfer

    Jun Zhao, Yifan Wang, Dengzhe Hou, Sai Sun, János Négyesi, Hitoshi Inada, Satoshi Shioiri, Ryoichi Nagatomi

    Scientific Reports　2025年7月19日

    DOI： 10.1038/s41598-025-11943-6 　

  8. EEG activity over ipsilateral and contralateral M1 during simple and complex hand tasks: variations with motor learning 査読有り

    Jun Zhao, Yifan Wang, Dengzhe Hou, János Négyesi, De-Lai Qiu, Ryoichi Nagatomi

    FRONTIERS IN NEUROSCIENCE　19　1681250　2025年

    DOI： 10.3389/FNINS.2025.1681250 　

︎全件表示 ︎最初の5件までを表示

MISC 10

  1. S1CogBench: What System One Decision Models Can and Cannot Do on Cognitive Experiments

    Dengzhe Hou, Kazunori D. Yamada

    arXiv preprint　arXiv:2611.00911　2026年10月31日

    DOI： 10.48550/arXiv.2611.00911 　

  2. Control-Diverse Reinforcement Fine-Tuning: Decoupling the Shared Control Bottleneck of RL Post-Training

    Binwen Tan, Jingchao Wang, Dengzhe Hou, Lingyu Jiang, Zeyuan Wu, Yunhan Shen, Fangzhou Lin, Kazunori Yamada, Atsushi Koike

    CoRR　abs/2608.08224　2026年8月

    DOI： 10.48550/arXiv.2608.08224 　

  3. CogBattery: Human-Referenced Evaluation of Cognitive Task Behavior and Structure in Language Models

    Dengzhe Hou, Lingyu Jiang, Fangzhou Lin, Kazunori D. Yamada

    CoRR　abs/2607.24999　2026年7月

    DOI： 10.48550/arXiv.2607.24999 　

  4. CogEEGAgent: Toward Autonomous Cognitive EEG Analysis with Grounded Execution and Selection-Aware Verification

    Dengzhe Hou, Lingyu Jiang, Fangzhou Lin, Kazunori D. Yamada

    CoRR　abs/2607.25045　2026年7月

    DOI： 10.48550/arXiv.2607.25045 　

  5. Same Brain, Different Prediction: How Preprocessing Choices Undermine EEG Decoding Reliability

    Dengzhe Hou, Zihao Wu, Lingyu Jiang, Zirui Li, Fangzhou Lin, Kazunori D. Yamada

    CoRR　abs/2605.07212　2026年5月8日

    DOI： 10.48550/arXiv.2605.07212 　

    詳細を見る 詳細を閉じる

    Electroencephalography (EEG) is a cornerstone of brain-computer interfaces and clinical neuroscience, yet deep learning models are typically trained and evaluated under a single, unreported preprocessing pipeline. We formalize preprocessing choices as a counterfactual intervention space and show that EEG predictions are surprisingly unstable under this space: across six datasets spanning four paradigms, up to 42% of trial-level predictions flip when only the preprocessing changes, a variability that standard uncertainty methods do not explicitly quantify because they condition on a fixed preprocessing pipeline. We provide three tools to make this instability measurable, decomposable, and reducible. First, a Walsh-Hadamard decomposition of the 2^7 pipeline space reveals that sensitivity is near-additive in practice under the binary intervention design, enabling efficient step-by-step optimization. Second, we introduce Preprocessing Uncertainty (PU), a per-trial diagnostic that captures a dimension of instability complementary to model-based confidence. Third, we study Normalized Adaptive PGI (NA-PGI), a graph-structured regularizer that exploits the compositional structure of preprocessing interventions as one mitigation strategy with clear scope conditions.

  6. Physics-Aware Video Instance Removal Benchmark

    Zirui Li, Xinghao Chen, Lingyu Jiang, Dengzhe Hou, Fangzhou Lin, Kazunori Yamada, Xiangbo Gao, Zhengzhong Tu

    CVPR 2026 Workshop on Video Generation and Beyond Evaluation (VGBE)　2026年4月7日

    DOI： 10.48550/arXiv.2604.05898 　

    詳細を見る 詳細を閉じる

    Video Instance Removal (VIR) requires removing target objects while maintaining background integrity and physical consistency, such as specular reflections and illumination interactions. Despite advancements in text-guided editing, current benchmarks primarily assess visual plausibility, often overlooking the physical causalities, such as lingering shadows, triggered by object removal. We introduce the Physics-Aware Video Instance Removal (PVIR) benchmark, featuring 95 high-quality videos annotated with instance-accurate masks and removal prompts. PVIR is partitioned into Simple and Hard subsets, the latter explicitly targeting complex physical interactions. We evaluate four representative methods, PISCO-Removal, UniVideo, DiffuEraser, and CoCoCo, using a decoupled human evaluation protocol across three dimensions to isolate semantic, visual, and spatial failures: instruction following, rendering quality, and edit exclusivity. Our results show that PISCO-Removal and UniVideo achieve state-of-the-art performance, while DiffuEraser frequently introduces blurring artifacts and CoCoCo struggles significantly with instruction following. The persistent performance drop on the Hard subset highlights the ongoing challenge of recovering complex physical side effects.

  7. Vibe Medicine: Redefining Biomedical Research Through Human-AI Co-Work

    Zihao Wu, Shu Xu, Boyu Chen, Shuo Wan, Yu Li, Wei Ruan, Yuxuan Lyu, Shuo Li, Dajiang Zhu, Tianming Liu, Dengzhe Hou, Lu Zhao

    CoRR　abs/2604.23674　2026年4月

    DOI： 10.48550/arXiv.2604.23674 　

  8. WMIP: A Working-Memory-Inspired Programmable Probe of Cumulative State Tracking in Large Language Models

    Dengzhe Hou, Lingyu Jiang, Deng Li, Zirui Li, Fangzhou Lin, Kazunori D Yamada

    arXiv preprint　arXiv:2603.27343　2026年3月28日

    DOI： 10.48550/arXiv.2603.27343 　

    詳細を見る 詳細を閉じる

    Existing large language models (LLMs) evaluations use fixed-difficulty benchmarks that cannot adapt as models improve, and rarely isolate specific cognitive processes. We introduce Working Memory Fidelity-Active Manipulation (WMF-AM), a probe of cumulative state tracking, the ability to maintain and update intermediate results across K sequential operations within a single query, without a scratchpad. Unlike multi-step agent benchmarks that stress task orchestration, WMF-AM isolates within-pass cumulative load by parameterizing depth K. The core probe uses arithmetic accumulation on 28 models from 12 families (0.5B to frontier); a matched non-arithmetic extension (permissions, schedules, inventories) confirms the design generalizes beyond arithmetic. Three construct-isolation ablations confirm that cumulative load, not arithmetic skill or entity tracking, drives difficulty. We release WMF-AM as a lightweight, recalibratable diagnostic for characterizing where models degrade under cumulative load. Code and data can be accessed at https://github.com/dengzhe-hou/WMF-AM

  9. KANMixer: a minimal KAN-centered mixer for long-term time series forecasting

    Lingyu Jiang, Dengzhe Hou, Yuping Wang, Yao Su, Shuo Xing, Wenjing Chen, Xin Zhang, Zhengzhong Tu, Ziming Zhang, Fangzhou Lin, Michael Zielewski, Kazunori D Yamada

    2025年8月3日

    詳細を見る 詳細を閉じる

    Long-term time series forecasting (LTSF) underpins critical applications from energy management to weather prediction, yet achieving reliable multi-step-ahead accuracy remains challenging. Existing LTSF approaches, dominated by MLP- and Transformer-based architectures, either rely on simple linear mappings or introduce increasingly complex hand-crafted inductive biases, raising the question of whether a more expressive and principled nonlinear core could offer a better alternative. Therefore, we investigate whether Kolmogorov-Arnold Networks (KANs), a recently proposed model featuring adaptive basis functions capable of granular modulation of nonlinearities, can improve LTSF performance, and under which design choices they are most effective. Specifically, we propose KANMixer, a minimal KAN-centered architecture consisting of a multi-scale pooling frontend, a KAN-based temporal mixing backbone, and prediction heads. By avoiding heavy auxiliary modules, KANMixer enables a clear assessment of KAN components in LTSF. Across 28 benchmark-horizon settings against nine baselines, KANMixer achieves the best MSE in 16 settings and the best MAE in 11. Furthermore, extensive ablations on three representative datasets show that KAN effectiveness depends strongly on the choice of edge function; B-spline bases outperform Fourier and Wavelet alternatives; the prediction head contributes most to the gains; moderate depth is preferred over deeper unstable stacks; and decomposition priors help MLP but harm KAN. Beyond practical guidance for integrating KAN into LTSF, these results reveal an underexplored dependency between structural priors and backbone nonlinearity: design choices that benefit MLP can degrade KAN.

  10. 機械学習による自発的注意シフト予兆の脳活動解析

    金成慧, HOU Dengzhe, 塩入諭

    電子情報通信学会技術研究報告(Web)　125　(291(HIP2025 64-73))　2025年

    ISSN： 2432-6380

︎全件表示 ︎最初の5件までを表示

講演・口頭発表等 18

  1. AIとニューロテクノロジーのガバナンス 招待有り

    侯登哲, マイク・ザイルースキ

    第44回日本ロボット学会学術講演会 オープンフォーラム OF7「テクノロジーの質的進化と組織統制」　2026年9月1日

  2. From Preprocessing Choices to LLM Agents: Automated and Verifiable Cognitive EEG Analysis 国際会議 招待有り

    Dengzhe Hou

    The 12th Annual CWRU-Tohoku Data Science in Engineering and Life Sciences Symposium　2026年8月4日

  3. 自発的注意シフトの脳内メカニズムに関するデータ駆動型研究

    Yuwen Zeng, Dengzhe Hou, Yongsong Huang, Chia-huei Tseng, Satoshi Shioiri

    ヒューマン情報処理研究会（HIP）　2025年12月15日

  4. 機械学習による自発的注意シフト予兆の脳活動解析

    金成慧, Dengzhe Hou, 塩入 諭

    ヒューマン情報処理研究会（HIP）　2025年12月15日

  5. 自由観察中の自発的注意シフトに先立つ脳活動の機械学習的解析

    金成 慧, Dengzhe Hou, 塩入 諭

    日本視覚学会 2025年夏季大会　2025年9月10日

  6. EEG-Based Decoding of Voluntary Attentional Control: Differentiating Self-Initiated Shifts in a Visual Search Paradigm

    Dengzhe Hou, Chia-huei Tseng, Satoshi Shioiri

    EPC & APCV Joint Meeting 2025 (The 17th Asia-Pacific Conference on Vision)　2025年6月20日

  7. Task complexity-related neurophysiological dynamics underlying visuomotor skill learning and interlimb transfer 国際会議

    J. Zhao, Y. Wang, D. Hou, S. Sun, J. Négyesi, H. Inada, S. Shioiri, R. Nagatomi

    Society for Neuroscience (SfN) Annual Meeting 2024　2024年10月9日

  8. Attention guides reward-based decision-making as measured by SSVEPs 国際会議

    Dengzhe Hou, Sai Sun, Satoshi Shioiri

    Society for Neuroscience (SfN) Annual Meeting 2024　2024年10月9日

  9. Exploring Mechanisms of Self-initiated Attention Shifts: Analysis of Theta Wave

    Dengzhe Hou, Sai Sun, Yasuhiro Hatori, Chia-huei Tseng, Satoshi Shioiri

    The 16th Asia-Pacific Conference on Vision (APCV 2024)　2024年7月12日

  10. Exploring Mechanisms of Self-initiated Attention Shifts – Analysis of Alpha Wave 国際会議

    Satoshi Shioiri, Dengzhe Hou, Sai Sun, Yasuhiro Hatori, Chia-huei Tseng

    The 16th Asia-Pacific Conference on Vision (APCV 2024)　2024年7月10日

  11. Criticality-State-Based Mechanism for Self-Initiated Attention

    侯 登哲, 塩入 諭

    日本視覚学会 2023年夏季大会　2023年9月19日

  12. Neural activities preceding self-initiated attention 国際会議

    Satoshi Shioiri, Wei Wu, Kazuya Kobayashi, Dengzhe Hou, Shin Ono, Yoshiyuki Sato, Yasuhiro Hatori, Chia-huei Tseng

    European Conference on Visual Perception (ECVP 2023)　2023年8月28日

  13. Self-initiation of attentional shift during visual search

    Dengzhe Hou, Sai Sun, Yoshiyuki Sato, Yasuhiro Hatori, Chia-huei Tseng, Satoshi Shioiri

    European Conference on Visual Perception (ECVP 2023)　2023年8月27日

  14. Investigation of Self-initiated Attention by EEG Measurements

    侯 登哲, 塩入 諭

    日本視覚学会 2023年冬季大会　2023年1月18日

  15. Analysis of EEG measures for self-initiated attention

    塩入 諭, Wei Wu, 小林 司弥, Dengzhe Hou, 小野 真, 佐藤 好幸, 羽鳥 康裕, 曾 加蕙

    日本視覚学会 2022年夏季大会　2022年9月5日

  16. Critical brain states related with self-initiated attentional shift

    Wei Wu, Kunihiro Kobayashi, Dengzhe Hou, Satoshi Ono, Yoshiyuki Sato, Yasuhiro Hatori, Chia-huei Tseng, Satoshi Shioiri

    Vision Sciences Society (VSS) Annual Meeting 2022　2022年5月15日

  17. Searching indexes for self-initiated attention shift

    侯 登哲, 吴 韡, 小林 司弥, 塩入 諭

    日本視覚学会 2022年冬季大会　2022年1月19日

  18. 注意シフトを開始する臨界脳状態の検討

    電子情報通信学会技術研究報告（HIP）　2021年10月1日

︎全件表示 ︎最初の5件までを表示

共同研究・競争的資金等の研究課題 7

  1. 注意テンプレートの表象形式：計算モデル階層・脳波・視線追跡による解明

    侯 登哲

    提供機関：Japan Society for the Promotion of Science

    制度名：Grant-in-Aid for Research Activity Start-up

    研究機関：Tohoku University

    2026年7月 ～ 2028年3月

  2. 自発的脳機能の神経基盤理解

    塩入 諭, 侯 登哲

    2024年4月 ～ 2028年3月

  3. テクノロジーガバナンスと社会的受容性に基づく課題解決と人材育成

    2026年6月 ～ 2027年10月

  4. 脳波基盤モデルにおける非脳源アーティファクト依存性の体系的検証

    2026年7月 ～ 2027年3月

  5. 認知科学実験パラダイムに基づくAIシステム認知能力評価プラットフォームの開拓

    侯 登哲

    2026年4月 ～ 2027年3月

  6. 自発的な注意の脳機能メカニズムの解明：脳波-眼球運動同時計測

    HOU DENGZHE

    提供機関：Japan Society for the Promotion of Science

    制度名：Grants-in-Aid for Scientific Research

    研究種目：Grant-in-Aid for JSPS Fellows

    研究機関：Tohoku University

    2025年4月1日 ～ 2026年3月31日

  7. 自発的な注意シフトの脳内メカニズムの解明

    侯 登哲

    2023年4月 ～ 2025年3月

︎全件表示 ︎最初の5件までを表示

担当経験のある科目(授業) 2

  1. データ科学トレーニング II・データ科学 チャレンジ　東北大学

  2. 機械学習基礎　東北大学

学術貢献活動 7

  1. International Joint Conference on Neural Networks (IJCNN)

    2026年4月1日 ～ 継続中

    学術貢献活動種別: 査読等

  2. European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML PKDD 2026)

    2026年4月1日 ～ 継続中

    学術貢献活動種別: 査読等

  3. Discover Neuroscience

    2025年1月1日 ～ 継続中

    学術貢献活動種別: 査読等

  4. Scientific Reports

    2025年1月1日 ～ 継続中

    学術貢献活動種別: 査読等

  5. Journal of NeuroEngineering and Rehabilitation

    2025年1月1日 ～ 継続中

    学術貢献活動種別: 査読等

  6. Cognitive Neurodynamics

    2025年1月1日 ～ 継続中

    学術貢献活動種別: 査読等

  7. npj Science of Learning

    2025年1月1日 ～ 継続中

    学術貢献活動種別: 査読等

︎全件表示 ︎最初の5件までを表示