Twitter | Pretraživanje | |
Quoc Le 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: Blog:
Reply Retweet Označi sa "sviđa mi se"
lerner zhang
By perplexity do you mean average perplexity? I wonder if a weighted average perplexity would be better?
Reply Retweet Označi sa "sviđa mi se" More
Danny Iskandar 29. sij
Odgovor korisniku/ci @lerner_adams @quocleix i 2 ostali
Taken from Google blog: perplexity, the uncertainty of predicting the next token (in this case, the next word in a conversation) Lower perplexity is better, means the model is good at predicting the next word
Reply Retweet Označi sa "sviđa mi se"
lerner zhang 29. sij
Odgovor korisniku/ci @diskandartweet @quocleix i 2 ostali
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.
Reply Retweet Označi sa "sviđa mi se"
Thang Luong 29. sij
Odgovor korisniku/ci @lerner_adams @quocleix @xpearhead
Perplexity for a language model, by definition, is computed by first averaging all neg log predictions and then exponentiating. Does that help explain?
Reply Retweet Označi sa "sviđa mi se"
lerner zhang 29. sij
Odgovor korisniku/ci @lmthang @quocleix @xpearhead
Danke
Reply Retweet Označi sa "sviđa mi se"