Author

Hyunsoo Cho

1 paper indexed here

We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.

Not the right person? Other researchers publish under this name.

#machine learning Feb 2026

QUATRO: Query-Adaptive Trust Region Policy Optimization for LLM Fine-tuning

This work proposes Query-Adaptive Trust-Region policy Optimization (QUATRO), which directly enforces trust-region constraints through a principled optimization, and shows stable training under increased policy staleness and aggressive learning rates, maintaining well-controlled entropy throughout training.

Doyeon Lee, Eun-Young Lyou, Hyunsoo Cho et al. · arXiv.org · 1 citation