ABOUT

I’m a Computer Science Ph.D. student at the NYU Courant Institute School, advised by Prof. Rumi Chunara. I also work closely with NYU Langone. Previously, I worked with Amazon as an Applied Scientist Intern, advised by Prof. Hengrui Cai and Dr. Rui Song. I finished my master’s and bachelor's at Seoul National University, advised by Prof. Sanghack Lee. Before my doctoral studies, I was a Researcher at LG AI Research.

I am primarily interested in Foundation Models, broadly in understanding their memorization and generalization. I am enthusiastic about a wide range of subjects, including:

• Memorization

  • Continual Learning / Unlearning for Foundation Models [W3]

  • Multi-agentic Memory for Agentic Collaboration 

• Generalization

  • Causal / Verifiable Reasoning Models [W4]

  • Foundation Models for safety-critical domains (e.g., Healthcare) [C1, W5, W2, W1]

  • Post Training (e.g., Weight Space Learning [W3], Model Collaboration [C1, W5], RLHF) 

Previously, I worked on Causal Representation Learning & Causality-inspired Generalization Algorithms. Apart from research, I’m a big fan of History👨🏻‍🏫 and Jazz🎷. I'm also interested in the application of Artificial Intelligence to humanities research (e.g., Histoinformatics). Please feel free to reach out for multi-disciplinary research collaborations!

NEWS

[2025.11.10] 🔥 Our paper on Inference-Time Model Collaboration was accepted on both ResponsibleFM - NeurIPS 2025 and Responsible Synthetic Data - AAAI 2026 as a Spotlight paper!

[2025.10.10] 🔥 Our paper on Semi-Verifiable Reasoning was accepted on AMLC 2025 Evaluation Paradigms for GenAI: Planning, Tools, and Real-World Performance as an Oral paper!

[2025.09.23] 🔥 Our paper on Semi-Verifiable Reasoning was accepted on ForLM- NeurIPS 2025 and ER- NeurIPS 2025 as a Spotlight!

[2025.08.13] Our paper on Continual Learning was accepted on CLVision- ICCV 2025. See you in Honolulu!

[2025.04.04] This summer, I will join Amazon in Seattle as an Applied Scientist Intern.

[2025.03.05] Our paper on Continual Learning was accepted at WSL - ICLR 2025. See you in Singapore!

[2025.02.26]  Our paper on Model-to-Model Regularization was accepted at CVPR 2025. See you in Nashville!

[2024.09.01] I've begun my Ph.D. journey in Computer Science at New York University!

Publications & Preprints (Google Scholar)

(C: Conference, W: Workshop, J: Journal, P: Preprint)

• [C1]  PEER Pressure: Model-to-Model Regularization for Single Source Domain Generalization 

  • • Dongkyu Cho, Inwoo Hwang, Sanghack Lee

    • CVPR 2025 (Acceptance Rate: 22.12%)

• [W5]  Expert-guided Clinical Text Augmentation via Query-Based Model Collaboration

  • • Dongkyu Cho*, Miao Zhang*, Gregory D. Lyng, Rumi Chunara

    • NeurIPS 2025, Workshop on Responsible FM 

    • AAAI 2026, Workshop on Responsible Synthetic Data (Spotlight)

    • Collaboration with Optum/United Healthcare

• [W4]  Correct Reasoning Paths Visit Shared Decision Pivots

  • • Dongkyu Cho, Amy B.Z. Zhang, Bilel Fehri, Rumi Chunara, Hengrui Cai, Rui Song

    • NeurIPS 2025, Workshop on Foundations of Reasoning in Language Models

    • NeurIPS 2025, Workshop on Efficient Reasoning (Spotlight)

    • AMLC 2025, Evaluation Paradigms for GenAI: Planning, Tools, and Real-World Performance (Oral)

    • Work done during Dongkyu's internship @ Amazon

• [W3]  Cost-Efficient Continual Learning with Sufficient Exemplar Memory

  • • Dongkyu Cho, Taesup Moon, Rumi Chunara, Kyunghyun Cho, Sungmin Cha

    • ICLR 2025, Workshop on Weight Space Learning

    • ICCV 2025, 6th Workshop on Continual Learning in Computer Vision

• [W2]  ShERPA: Leveraging Neuron Alignment for Knowledge-preserving Fine-tuning

  • • Dongkyu Cho, Jinseok Yang, Jun Seo, Seohui Bae, Dongwan Kang, Hyeokjun Choe, Woohyung Lim

    • ICLR 2024, Workshop on Mathematical and Empirical Understanding of Foundation Models

• [W1]  Learning to ignore: Single Source Domain Generalization via Oracle Regularization

  • • Dongkyu Cho, Sanghack Lee

    • NeurIPS 2023, Causal Representation Learning Workshop

• [P4]  Towards Factual Measure of Verifiable Reasoning

• [P3]  Tree of Concepts: Towards Interpretable Continual Learning Models  (Working Paper with NYU Langone)

• [P2]  Towards Clinical Reasoning Models: Verifiable Rewards via Expert Supervision

• [P1]  Dealing with the Evil Twins: Improving Random Augmentation by Addressing Catastrophic Forgetting of Diverse Augmentations

Education

[Ph.D.] Doctor of Philosophy, New York University 

Courant Institute School of Mathematics, Computing, and Data Science

• Doctor of Computer Science   (September 2024 ~ May 2029)

• Advisor: Professor Rumi Chunara

• Field of Research: Model Generalization, Foundation Models, Causality

[MS] Master of Science, Seoul National University

• Master of Data Science   (March 2021 ~ August 2023)

• Advisor: Professor Sanghack Lee

• Field of Research: Causality, Causal Representation Learning

[BA] Bachelor of Arts, Seoul National University

• Information Science & Culture/ Western History   (March 2014 ~ February 2021)

• Field of Research: Quantitative Historical Research.

Work Experience

Amazon Science - Applied Scientist Intern   (May 2025 ~ )

• Applied Scientist Intern at Amazon, Worldwide Trust.

• Advised by Prof. Hengrui Cai and Dr. Rui Song

LG AI Research - Research Scientist Intern   (July 2023 ~ July 2024)

• Research Scientist Intern at LG AI Research, Data Intelligence Lab

• Research Field: Time-Series Foundation Models, Loss Landscapes and Model Merging, Alignment of Large Language Models for Time-Series Forecasting, LLM-driven Causal Discovery

Causality Lab, SNU GSDS - Research Assistant   (July 2021 ~ August 2023)

• Research Assistant at Causality Lab, Seoul National University GSDS

• Research Field: Leveraging causality for effective Out-of-Distribution Generalization

VAIV Company - Analyst Intern   (January 2019 ~ February 2019)

• Analyst Intern at VAIV Company

• NLP-based Market Sentiment Analysis

Academic Services

• Conference Reviewer

  • ICLR (2026)

  • ICLR MCDC (2025), CoLLAs (2025), ICML (2025)

Invited Talks

• Invited Talk at NYU Center for Health Data Science (09.2025)

• Invited Talk at NYU Digital Health Work (03.2025), 

• Invited Talk at SNU GSDS Student Seminar (05.2023)