|
@GoogleAI | |||||
|
Today we announce a novel, open-source method for text generation tasks (e.g., summarization or sentence fusion), which uses edit operations instead of generating text from scratch, leading to less errors and faster model execution. Read about it below. goo.gle/38XfRXU
|
||||||
|
||||||
|
Thomas G. Dietterich
@tdietterich
|
31. sij |
|
You mean "fewer errors" :-)
|
||
|
|
||
|
Eric Malmi
@ericmalmi
|
31. sij |
|
This may have been avoided if the tweet was passed through the LaserTagger model that was trained for the GEC task!
|
||
|
|
||
|
Dave 🦔
@_dmh
|
1. velj |
|
I'm surprised this hasn't been done before! I know folks have done sequence tagging for extractive summarization at the level of deciding which sentences to keep. Is this the first work on doing the same at the token level? Or is it the focus on inference speed that's novel?
|
||
|
|
||
|
Fernando Alva-Manchego
@feralvam
|
1. velj |
|
I haven't read the paper in detail yet, but I remember something similar for sentence simplification aclweb.org/anthology/P19-…
|
||
|
|
||
|
Vaclav Kosar
@vackosar
|
1. velj |
|
Listen to the paper here you can: youtu.be/a6Xp2c3JXYc
|
||
|
|
||
|
BELHABIB AMINE
@belhamin
|
1. velj |
|
@ZiyadMestour this one is for u man
|
||
|
|
||
|
Aritra Roy Gosthipaty
@ariG23498
|
31. sij |
|
|
||
|
MOHAMMAD FAIZ
@MOHAMMA31730751
|
1. velj |
|
This sounds good but it also contain lot of work and most importantly to apply what is been think or visualize...✌🙌
|
||
|
|
||
|
wellactualybot
@wellactualybot
|
1. velj |
|
@GoogleAI um it's actualy spelled generatting*
|
||
|
|
||
|
Serhat Bakis
@M_BH55
|
1. velj |
|
€**()¥7¥£¥78())98£()₩008£())000 @ft
|
||
|
|
||