|
lerner zhang
@
lerner_adams
Shanghai, Volksrepublik China
|
|
|
|
|
116
Tweetovi
|
465
Pratim
|
11
Osobe koje vas prate
|
| Tweetovi |
|
lerner zhang
@lerner_adams
|
14 h |
|
I wonder if the average perplexity is calculated in this way? pic.twitter.com/cx9F3hf1E4
|
||
|
|
||
| lerner zhang proslijedio/la je tweet | ||
|
Dirk Hovy
@dirk_hovy
|
30. sij |
|
I have now seen several classification tasks where a logistic regression model with bag-of-character n-grams tied or even beat neural (MTL, LSTM, CNN+attention) models. Any similar observations?
|
||
|
|
||
| lerner zhang proslijedio/la je tweet | ||
|
𝗨𝗻𝗙𝗼𝘂𝗻𝗱.𝗮𝗶
@UnFound_ai
|
31. sij |
|
Not chargram, but see this leaderboard paperswithcode.com/sota/sentiment…
|
||
|
|
||
|
lerner zhang
@lerner_adams
|
2. velj |
|
The End of an Era | Q&A 2019 Finale abc.net.au/qanda/2019-09-…
|
||
|
|
||
|
lerner zhang
@lerner_adams
|
29. sij |
|
Actually I don't know anyone who did the weighting so far.
|
||
|
|
||
|
lerner zhang
@lerner_adams
|
29. sij |
|
I wonder if it would be of any use if you are not considering a job switch..
|
||
|
|
||
|
lerner zhang
@lerner_adams
|
29. sij |
|
just my two cents
|
||
|
|
||
|
lerner zhang
@lerner_adams
|
29. sij |
|
Understood. I thought we can get a perplexity for each generated word, but all words are not equal. Maybe the perplexities can be weighted by the tf-idf score or the alike.
|
||
|
|
||
|
lerner zhang
@lerner_adams
|
29. sij |
|
Genau
|
||
|
|
||
|
lerner zhang
@lerner_adams
|
29. sij |
|
Danke
|
||
|
|
||
|
lerner zhang
@lerner_adams
|
29. sij |
|
By perplexity do you mean average perplexity? I wonder if a weighted average perplexity would be better?
|
||
|
|
||
| lerner zhang proslijedio/la je tweet | ||
|
Quoc Le
@quocleix
|
28. sij |
|
New paper: Towards a Human-like Open-Domain Chatbot. Key takeaways:
1. "Perplexity is all a chatbot needs" ;)
2. We're getting closer to a high-quality chatbot that can chat about anything
Paper: arxiv.org/abs/2001.09977
Blog: ai.googleblog.com/2020/01/toward… pic.twitter.com/5SOBa58qx3
|
||
|
|
||
|
lerner zhang
@lerner_adams
|
26. sij |
|
我只用quora和stackexchange,知乎上充满了歧视和偏见。
|
||
|
|
||
|
lerner zhang
@lerner_adams
|
24. sij |
|
How do you determine how much data to use/annotate(like outsourcing)? datascience.stackexchange.com/q/66963/9191
|
||
|
|
||
|
lerner zhang
@lerner_adams
|
14. srp |
|
Can I treat you to a meal or a coffee?
|
||
|
|
||
| lerner zhang proslijedio/la je tweet | ||
|
Physics-astronomy.org
@OrgPhysics
|
16. lip |
|
One word this video... pic.twitter.com/D1jpArOdco
|
||
|
|
||
|
lerner zhang
@lerner_adams
|
17. lip |
|
The best explanation of CRF I have ever seen!
|
||
|
|
||
|
lerner zhang
@lerner_adams
|
15. lip |
|
Is this true?
|
||
|
|
||
|
lerner zhang
@lerner_adams
|
15. lip |
|
That's crazy.
|
||
|
|
||
|
lerner zhang
@lerner_adams
|
3. lip |
|
I wish I can work in such companies in the future. workplace.stackexchange.com/a/28271/44279?…
|
||
|
|
||