SANGAH LEE / TEACHING
Teaching
My teaching spans the foundations of computational linguistics and current research on large language models. Undergraduate courses emphasize both conceptual foundations and hands-on technical skills, while graduate seminars explore a different LLM-related research theme each semester, ranging from linguistic probing and reasoning to evaluation, fine-tuning, and model behavior.
From basic topics and technical skills for computational linguistics to neural architectures: csv and json, regular expressions, numpy, pytorch, logistic regression, feed-forward networks, RNNs/LSTMs, CNNs.
From statistical language modeling to neural architectures and large language models: n-grams, vector semantics, sequence-to-sequence models, attention, Transformers, and modern LLMs.
From statistical language modeling to neural architectures and large language models: n-grams, vector semantics, sequence-to-sequence models, attention, Transformers, and modern LLMs.
Understanding and Shaping LLM Behavior
A research seminar on how language models represent, reason, remember, and behave, with topics including interpretability, knowledge, steering, multilinguality, preference, and alignment.
From basic topics and technical skills for computational linguistics to neural architectures: csv and json, regular expressions, numpy, pytorch, logistic regression, feed-forward networks, RNNs/LSTMs, CNNs, and sequence-to-sequence models.
Training, Tuning and Utilizing Large Language Models
A research seminar on the application, training, and fine-tuning of Large Language Models (LLMs) to produce actual research outcomes.
From statistical language modeling to neural architectures and large language models: n-grams, vector semantics, sequence-to-sequence models, attention, Transformers, and modern LLMs.
Large Language Models and Reasoning
A research seminar on Transformer-based large language models and reasoning abilities of such models.
From basic topics and technical skills for computational linguistics to neural architectures: csv and json, regular expressions, numpy, pytorch, logistic regression, feed-forward networks, RNNs/LSTMs, CNNs, and sequence-to-sequence models.
Understanding and Evaluating Large Language Models
A research seminar discussing ways to achieve better generation through evaluating the outputs of generative LLMs.
From statistical language modeling to neural architectures and large language models: n-grams, vector semantics, sequence-to-sequence models, attention, Transformers, and modern LLMs.
Large Language Models and Linguistic Probing
A research seminar on Transformer-based large language models and linguistic probing of such models.
Current Courses
Language and Computer
Studies in Computational Linguistics 1
Previous Courses
Computational Linguistics
Studies in Computational Linguistics 2
Language and Computer
Seminar in Computational Linguistics
Computational Linguistics
Studies in Computational Linguistics 2
Language and Computer
Seminar in Computational Linguistics
Computational Linguistics
Studies in Computational Linguistics 2