publications

publications by categories in reversed chronological order. generated by jekyll-scholar.

2026

  1. okawa2026selection.jpg
    How Selection Shapes Diversity in LLM Ecosystems
    Maya Okawa
    In Advances in Neural Information Processing Systems (NeurIPS), 2026
    HTML
  2. okawa2026consensus.jpg
    Emergence of Biased Consensus in Multi-Agent LLM Debates
    Maya Okawa
    In International Conference on Machine Learning (ICML), 2026
    arXiv HTML
  3. okawa2026emotion.jpg
    Emergence of Hierarchical Emotion Organization in Large Language Models
    Maya Okawa, Bo Zhao, Eric Bigelow, and 4 more authors
    In International Conference on Machine Learning (ICML), 2026
    arXiv HTML
  4. Aggregated Multi-output Gaussian Processes with Knowledge Transfer Across Domains
    Yusuke Tanaka, Toshiyuki Tanaka, Tomoharu Iwata, and 4 more authors
    Machine Learning, 2026
    DOI arXiv HTML

2025

  1. Kindness or Sycophancy? Understanding and Shaping Model Personality via Synthetic Games
    Maya Okawa, Ekdeep Singh Lubana, Mai Uchida, and 1 more author
    In NeurIPS 2025 Workshop on Cognitive Interpretability (CogInterp), 2025
  2. nishi2025representation.jpeg
    Representation Shattering in Transformers: A Synthetic Study with Knowledge Editing
    Kento Nishi, Rahul Ramesh, Maya Okawa, and 3 more authors
    In International Conference on Machine Learning (ICML), 2025
    arXiv HTML
  3. park2025incontext.gif
    In-Context Learning of Representations
    Core F. Park, Andrew Lee, Ekdeep Singh Lubana, and 5 more authors
    In International Conference on Learning Representations (ICLR), 2025
    arXiv HTML
  4. Swing-by Dynamics in Concept Learning and Compositional Generalization
    Yongyi Yang, Core F. Park, Ekdeep Singh Lubana, and 3 more authors
    In International Conference on Learning Representations (ICLR), 2025
    arXiv HTML

2024

  1. okawa2024emergence.gif
    Emergence of Hidden Capabilities: Exploring Learning Dynamics in Concept Space
    Maya Okawa, Core F Park, Andrew Lee, and 2 more authors
    In Advances in Neural Information Processing Systems (NeurIPS), 2024
    Spotlight
    arXiv HTML
  2. Towards an Understanding of Stepwise Inference in Transformers: A Synthetic Graph Navigation Model
    Mikail Khona, Maya Okawa, Jan Hula, and 5 more authors
    In International Conference on Machine Learning (ICML), 2024
    arXiv HTML
  3. Emergence of Hierarchical Emotion Representations in Large Language Models
    Bo Zhao, Maya Okawa, Eric J. Bigelow, and 3 more authors
    In NeurIPS 2024 Workshop on Scientific Methods for Understanding Deep Learning (SciForDL), 2024
  4. Dynamics of Concept Learning and Compositional Generalization
    Yongyi Yang, Core F. Park, Ekdeep Singh Lubana, and 3 more authors
    In NeurIPS 2024 Workshop, 2024
  5. Hidden Learning Dynamics of Capability before Behavior in Diffusion Models
    Core F. Park, Maya Okawa, Andrew Lee, and 2 more authors
    In ICML 2024 Workshop, 2024
  6. Meta-Learning for Neural Network-based Temporal Point Processes
    Yoshiaki Takimoto, Yusuke Tanaka, Tomoharu Iwata, and 4 more authors
    arXiv preprint arXiv:2401.15846, 2024
    arXiv HTML

2023

  1. okawa2023compositional.gif
    Compositional Abilities Emerge Multiplicatively: Exploring Diffusion Models on a Synthetic Task
    Maya Okawa, Ekdeep S Lubana, Robert Dick, and 1 more author
    In Advances in Neural Information Processing Systems (NeurIPS), 2023
    arXiv HTML Video
  2. Compositional Abilities Emerge Multiplicatively: Exploring Diffusion Models on a Synthetic Task
    Maya Okawa, Ekdeep Singh Lubana, Robert P. Dick, and 1 more author
    In ICML 2023 Workshop, 2023
  3. Stepwise Inference in Transformers: Exploring a Synthetic Graph Navigation Task
    Mikail Khona, Maya Okawa, Rahul Ramesh, and 4 more authors
    In NeurIPS 2023 Workshop, 2023

2022

  1. okawa2022predicting.jpeg
    Predicting Opinion Dynamics via Sociologically-Informed Neural Networks
    Maya Okawa and Tomoharu Iwata
    In 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD), 2022
    arXiv HTML
  2. Context-Aware Spatio-Temporal Event Prediction via Convolutional Hawkes Processes
    Maya Okawa, Tomoharu Iwata, Yusuke Tanaka, and 3 more authors
    Machine Learning, 2022
    DOI HTML

2021

  1. Dynamic Hawkes Processes for Discovering Time-evolving Communities’ States behind Diffusion Processes
    Maya Okawa, Tomoharu Iwata, Yusuke Tanaka, and 3 more authors
    In 27th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD), 2021
    arXiv

2019

  1. Deep Mixture Point Processes: Spatio-temporal Event Prediction with Rich Contextual Information
    Maya Okawa, Tomoharu Iwata, Takeshi Kurashima, and 3 more authors
    In 25th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD), 2019
    arXiv
  2. Refining Coarse-Grained Spatial Data Using Auxiliary Spatial Data Sets with Various Granularities
    Yusuke Tanaka, Tomoharu Iwata, Toshiyuki Tanaka, and 3 more authors
    In 32nd AAAI Conference on Artificial Intelligence (AAAI), 2019
    arXiv
  3. Spatially aggregated Gaussian processes with multivariate areal outputs
    Yusuke Tanaka, Toshiyuki Tanaka, Tomoharu Iwata, and 4 more authors
    In Advances in Neural Information Processing Systems (NeurIPS), 2019
    arXiv
  4. Marked Temporal Point Processes for Trip Demand Prediction in Bike Sharing Systems
    Maya Okawa, Yusuke Tanaka, Takeshi Kurashima, and 2 more authors
    IEICE Transactions on Information and Systems, 2019
    DOI HTML
  5. Predicting Traffic Accidents with Event Recorder Data
    Yoshiaki Takimoto, Yusuke Tanaka, Takeshi Kurashima, and 3 more authors
    In 3rd ACM SIGSPATIAL International Workshop on Prediction of Human Mobility (PredictGIS), 2019
    DOI HTML

2017

  1. Online traffic flow prediction using convolved bilinear Poisson regression
    Maya Okawa, Hideaki Kim, and Hiroyuki Toda
    In 18th IEEE International Conference on Mobile Data Management (MDM), 2017