Hippocampus's Garden

Ordinal regression has order structure between classes and there are dedicated loss functions to use this information.

今後5年程度の未来を見据え、AI時代を生き抜くための方法について自分の考えをまとめました。

Reflecting on 2024's deep learning breakthroughs! Discover my top 10 favorite research papers that shaped the field this year.

RLHF is not the only method for AI alignment. This article introduces modern algorithms like DPO and KTO that offer simpler and more stable alternatives.

パリ移住で様々な失敗を経ながら学んだ「生活の知恵」をまとめました。

NotebookLMを使って青空文庫版『罪と罰』をまとめました。

新天地で働き始めてもうすぐ1年になるので、所感をまとめました。

This competition was all about distribution shift. Let's learn how the winners conquered the challenge.

It's hard to collect paired preferences. Can we align LLMs without them? Yes, with KTO!

DPO reduces the effort required to align LLMs. Here is how I created the Reviewer #2 Bot from TinyLlama using DPO.