My main authored paper written on NAVER has been accepted at ICDAR2024 🎉🎉
Our method, named Coordinate-aware End-to-end Document Parser (CREPE), is a simple and easy to train yet highly functional model for Visual Document Understanding.
What Do Neural Networks Learn When Trained With Random Labels?
#NeurIPS2020
ランダムラベルでニューラルネットワークを学習するとどうなる?
FineTune速度向上は直感的に納得。それだけでなくpretrainが効果的な場合と逆効果となる場合の相違点を検証。さらに悪影響を回避するコツまで提案。
My main authored paper written on NAVER has been accepted at ICDAR2024 🎉🎉
Our method, named Coordinate-aware End-to-end Document Parser (CREPE), is a simple and easy to train yet highly functional model for Visual Document Understanding.
Is the countermeasure against illegal copy videos really sufficient? Our study shows traditional video copy detection methods fail on the current mainstream short video platforms, characterized by their brief duration of just a few seconds.
なにこれ?って最初思ったけど、ちゃんと読んだら「スゲェーー!!」ってなった
The effective number of samples is defined as the volume of samples and can be calculated by a simple formula (1−β^n)/(1−β),
論文
参考にした記事