-
Add your name in a comment below to be added to the Computer Vision for Digital Heritage mailing list and have your name listed on this website in the "Participants" section of our SIG webpage: https:…
-
Hello,
I'm a PhD Candidate at the University of Washington doing research on misinformation around COVID policies and our team is looking for a dataset of manually fact checked tweets to validate our…
-
### Name and Institution (Required)
Name: Alli Moon
Institution: University of Washington
### Confirm you have reviewed the following documentation
- [x ] [Support guidelines](https://geos-c…
-
# remaining to-dos
- [x] @briesenberg07 complete shared-email setup for [uwlsemanticweb@uw.edu](mailto:uwlsemanticweb@uw.edu)
- [ ] Add [uwlsemanticweb@uw.edu](mailto:uwlsemanticweb@uw.edu) as conta…
-
https://www.barrfoundation.org/grantmaking/grants
-
## 一言でいうと
論文の概要/結論/課題を自動生成するという夢の研究。既存の論文から知識グラフを作成し、Link PredictionによりEntity間の接続を増やす。生成を行う際は既存研究のタイトルを与え、そこから知識グラフを用い関連Entityを抽出。既存研究とEntityとの関連度に応じAttentionをかけながら生成を行う。
![image](https://user-i…
-
I see lots of .pyc files in your repo. Using a .gitignore that is configured for python will filter these out from git actions and you can git rm the existing ones.
-
## 一言でいうと
言語モデルに、固有表現とその参照を組み込む研究(「スタバ」に行った、「そこ」では、など)。固有表現か否かのフラグ(当否・出現番号・残りの単語数)を導入し、生成時もこのフラグで分岐して生成を行っている。
![image](https://user-images.githubusercontent.com/544269/31000118-55433224-a514-11…
-
Participants: Kaylea Champion @kayleachampion, Wm Salt Hale @altsalt (both University of Washington)
Project: We propose to extend GrimoireLab as needed in order to support an analysis of software li…
-
## 一言でいうと
BERTに知識グラフ(構造知識)を組み込む研究。Word Pieceの表現を線形変換=>メンションスパン分集めてpooling=>スパン間でSelf-Attention=>候補エンティティ表現の重み付き平均をとりスパン表現に加算=>Word PieceのSelf-Attention時にこれを参照、という流れ。
![image](https://user-images…