Hee-woo Jun

Towards Structured Mechanistic Reasoning in Single-cell Prediction

LLMs as Annotators: Evaluating Model–Human Alignment in Detecting Contentious Language in Historical Corpora

It is suggested that LLMs can serve as auxiliary annotators for sensitive-language detection in historical materials, provided that human oversight and contextual interpretation remain central to annotation workflows.

The Dialect Tax: Dialectal Biases Persist throughout the Language Modeling Pipeline

The study traces this dialect tax across the natural language processing pipeline, suggesting that the dialect tax is encoded and accumulated not by any one step in isolation, but at every step of the language modeling process.

Code Monitor Red Teaming for Public-Test-Passing Code

This work introduces Code Monitor Red Teaming, a monitor-red-teaming protocol that fixes a public-check information boundary while varying generator pressure, verifier scaffolding, and weak-to-strong capability.