Sangwoo Park

Check out our new preprint on achieving Contextual Integrity in LLMs via complementary self-distillation! :)

Can LLM agents build memory before seeing any user task? Memory is usually built from human tasks or deployment interactions. New tool environments often have neither, creating cold-start gap. Introducing PREPING: building agent memory without tasks. dozi01.github.io/preping-projec…

ThinkSafe: Self-Generated Safety Alignment for Reasoning Models A framework that restores safety in reasoning models without external teachers. It unlocks latent harm-identification knowledge via lightweight refusal steering, improving safety while preserving reasoning—with lower compute than GRPO.

THINKSAFE Self-Generated Safety Alignment for Reasoning Models huggingface.co/papers/2601.23…