|
@GoogleAI | |||||
|
Check out Meena, a new state-of-the-art open-domain conversational agent, released along with a new evaluation metric, the Sensibleness and Specificity Average, which captures basic, but important attributes for normal conversation. Learn more below! goo.gle/36zB8Wj
|
||||||
|
||||||
|
Google AI
@GoogleAI
|
28. sij |
|
|
||
|
|
||
|
Alex Engler
@AlexCEngler
|
28. sij |
|
Wow! Impressive!
And also:
brookings.edu/research/the-c…
|
||
|
|
||
|
Steve
@Zergfriend
|
29. sij |
|
I wanna chat with Meena though ::(
|
||
|
|
||
|
Joshua Herman
@zitterbewegung
|
28. sij |
|
Any timeline on releasing the source code ?
|
||
|
|
||
|
Surya Kasturi
@bysurya
|
28. sij |
|
@xpearhead, @lmthang, @quocleix How different is Meena's SSA from other human eval methods? Did you guys perform any correlation b/w Meena's SSA, PPL vs. other human evaluations such as DSTC7's Appropriateness and Informativeness/Utility.
|
||
|
|
||
|
Thang Luong
@lmthang
|
1. velj |
|
Good point! Many evaluations use 5-point Likert scale & look at other aspects, e.g., diversity, relevance, humanlikeness, etc. We use binary evaluation and think SSA
is basic to human quality & easy for crowdworkers to rate. also in paper, SSA correlates with humanlikeness. pic.twitter.com/ptrX4Ofs7m
|
||
|
|
||
|
Jero
@jeroaranda
|
29. sij |
|
|
||
|
Arturo MF
@arturomf94
|
29. sij |
|
Olv 2.6 billion params haha
|
||
|
|
||
|
Maged Samuel
@m_sam489
|
28. sij |
|
When will you release the fast voice typing dictation in Gboard to other Android phones, which is currently available in Gboard on pixel phones only?! 🤔
|
||
|
|
||
|
Henny '#TrumpResign' Roggy
@TotalSecurily
|
29. sij |
|
This is super exciting! Can't wait for an actual product demo at Google I/O!
|
||
|
|
||
|
Henny '#TrumpResign' Roggy
@TotalSecurily
|
29. sij |
|
I mean *if* they demo it by then. It would be a great opportunity, though!
|
||
|
|
||