Skip to content
Tutorials
Presentations
2025
- A multi-modal, multi-agent system: A minimalistic framework to study multi-agent systems [slides]
- WMAC 2025 – AAAI 2025 Workshop on Advancing LLM-Based Multi-Agent Collaboration
- Learning to X [slides]
- NorwAI&NAIL AI Seminar at NTNU, Trondheim, Norway
- Learning to infer statistical quantities [slides]
- Samsung-Mila-NYU Seminar Series.
- Causality gets help
from deep learning [slides]
- Computational Biomedicine Seminar at NYU School of Medicine.
2024
- Learning to X [slides]
- Analytical Connectionism Summer School at the Simons Foundation
- NYU STERN
- UPenn Wharton School of Business
- You want to train language models yourself from scratch [slides]
- Musings on DPO [slides]
- Samsung-Mila-NYU Monthly Meeting.
- Introduction to Natural Language Processing and Multilingual Machine Translation
- Together with Dr. Duygu Ataman.
- Department of Field Support at United Nations.
- Lab-in-the-Loop Therapeutic Antibody Design [slides]
- Can Language Models Guess Your Identities from De-identified Clinical Notes? [slides]
- Apple Workshop on Machine Learning for Healthcare
- Beyond test accuracies for studying deep neural networks [slides]
2023
- Beyond test accuracies for studying deep neural networks [slides]
- Prescient Design’s ML research for lab-in-the-loop antibody design [slides]
- HY-IBB HY-AIR HY-IPT Joint International Symposium on Data-driven, AI-aided Molecular Modeling and Drug Design
- OxML 2023
- Health system scale language models: TED-style talk [slides]
- Health system scale language models for clinical and operational tasks [slides]
- A slight-less-magical perspective into autoregressive language modeling: count, compress and prune [slides]
- Workshop on AI for Thermal Sciences (UC Irvine)
- ALPS Winter School
- Introduction to Natural Language Processing [slides]
- Prescient Design’s Lab-in-the-Loop de novo antibody design [slides]
- Generative multitask learning with a target-causing confounder [slides]
2022
- Prescient Design’s Lab-in-the-Loop de novo antibody design [slides]
- ML in PL 2022
- Neural Concept Conference 2022
- A slight-less-magical perspective into autoregressive language modeling: count, compress and prune [slides]
- Panel discussion at Believe the Impossible | CIFAR Celebrates 40 Years [recording]
- Computer Science Chats [recording]
- Generative multitask learning with a target-causing confounder [slides]
- Korean AI Hub
- Flatiron Institute MLxScience Summer School
- Learned data augmentation for natural language processing [slides]
- Columbia University
- KAIST
- Hyundai Motors AIRS
- A Deep Manifold Sampler: Foundations behind the Prescient Design’s approach to protein design [slides] [recording]
- Google
- Roche Data & Analytics (D&A) Data Science Seminar Series
- Lunit
- Machine translation [slides]
- AI: Revolution or Evolution? [slides]
2021
- Interview at <Humans of AI: Stories, Not State> [recording]
- Oversmoothing in neural autoregressive modeling [slides]
- Rodrigo Nogueira before Passage Re-ranking with BERT [slides]
- 제8회 SW Welcomes Girls [recording]
- 카이스트 영리더 목요특강 [slides]
- Programming vs. Machine Learning [slides]
- Samsung Software Society
- Allen School NLP Speaker Series
- KAIST CS Colloquium [recording]
- Sungkyunkwan University (SKKU) AI colloquium
- Seoul National University of Science & Technology BK Seminar Series
- Few-shot learning with a large-scale language model [slides]
- KLUE: Korean Language Understanding Evaluation [slides]
- Online hyperparameter optimization [slides]
- Unreasonably shallow deep learning [slides]
- HiTZ Language Technology Webinar Series [recording]
- Pfizer
- Machine translation [slides]
2020
- A speculative lecture on SGD in deep learning [slides]
- An guest lecture in MAS480: Math and AI at KAIST Math by Wanmo Kang
- Humans and machines in screening mammograms [slides][recording]
- SSMBA: Self-Supervised Manifold Based Data Augmentation [slides][recording]
- Unreasonably shallow deep learning [slides]
- A guest lecture at CS774 AI Ethics by Alice Oh
- Learning to find evidence [slides][recording]
- Amazon NLP Seminar Series. June 2020.
- SNU GSDS Seminar Series. June 2020.
- Rapidly deploying open science platforms for COVID-19 at NYU Alumni fireside chat [slides]
- Podcast Underrated ML
- QAGS. Question Answering and Generation for evaluating Summarization [slides][recording]
- ML in Korea Social at ICLR 2020.
- LxMLS Lisbon Machine Learning Summer School 2020.
- Inconsistency of recurrent language models [slide]
- UBC ML Seminar Series. July 2020.
- Cornell Tech. March 2020.
- Google AI. Feb 2020.
- Sequential or Non-Sequential, that is the problem: UCSB Mind&Machine Intelligence Workshop. Feb 2020. [slides]
2019
2018
2017
- Recent Advances in Neural Machine Translation at SlatorCon New York 2017
- Learning to Decode at RE:WORK Deep Learning Summit Montreal 2017
- Invited talk at RepEval 2017 co-located with EMNLP’17
- Invited talk at Subword and Character Level Modelling (SCLeM) 2017 co-located with EMNLP’17
- Invited talk at Lisbon Machine Learning Summer School 2017
- Deep learning, where are you going?
- State University of New York (Stony Brook), Nov 2017
- University of Notre Dame, Nov 2017
- University of Texas (Austin), Oct 2017
- U. Massachusetts (Amherst) , Oct 2017
- CU Boulder, Oct 2017
- U. Copenhagen, Sept 2017
- Naver, July 2017
- Jeju Machine Learning Camp, July 2017
- ShanghaiTech, July 2017
- Neural Machine Translation. Open workshop with Kyunghyun Cho
- Nordic Conference on Computational Linguistics (Nodalida) 2017
- Deep learning and structures
- Deep learning, where are you going?
- Curious AI Company
- Rensselaer Polytechnic Institute
- Google NYC 2017
- Goldman Sachs (Machine Learning Workgroup) 2017
2016
2015
- New Territory of Machine Translation. Winter 2015.
- Keynote Speech at MT Summit XV. Oct 2015.
- Lost in Interpretability
- Data Science Lunch Seminar @Center for Data Science, NYU. Sep 2015.
- Neural Machine Translation and Connectionist Natural Language Processing
- at Google NYC, Oct 2015.
- at the University of Maryland, College Park, Sept 2015.
- at the 3rd Workshop on Continuous Vector Space Models and their Compositionality (CVSC) 2015 @ACL’15, July 2015.
- Deep Learning: Present, Challenges and Future
- at the School of Computer and Communications, EPFL, Feb 2015.
- at the Department of Computer Science, Virginia Tech, Feb 2015.
- at the School of Informatics, University of Edinburgh, Feb 2015.
- at the Department of Computer Science, Courant Institute of Mathematical Sciences, New York University, Feb 2015.
- at the School of Computer Science, Carnegie Mellon University, Mar 2015.
- at the Department of Computer Science, Tufts University, Mar 2015.
- at the Mind Research Network, University of New Mexico, Mar 2015.
- at the School of Electrical Engineering and Computer Science, Washington State University, Apr 2015.
Before 2015
- First Step toward Neural Machine Translation: Approaches, Challenges and By-Products
- at IBM Watson R&D Center, US, Sep 2014.
- at the University of Sherbrooke, Canada, Oct 2014.
- at Google DeepMind, UK, Oct 2014.
- at the University of Oxford, UK, Oct 2014.
- at the University of Cambridge, UK, Oct 2014.
- at the University of Montreal (Recherche appliquée en linguistique informatique), Nov 2014.
- Beyond Word Embedding: Learning Phrase Representations with RNN Encoder–Decoder
- Presented at Samsung Advanced Institute of Technology, Korea, June 2014.
- Presented at Seoul National University, Korea, June 2014.
- Deep Neural Networks: Basic Principles in a Nutshell
- Presented at the Nokia Research Center, Finland. April 2013.