What were the most trending papers at #EMNLP2020 ?
Here is a snapshot of the main papers leaderboard based on scraped slideslive view counts. It was taken on late Friday, a few hours before the announcement of the winning papers...
emnlp-2020.herokuapp.com
Hello #emnlp2020 folks, check out our 'DORB: dynamic optimiz multiple RL rewards via bandits' paper at 12pm ET Nov18. Also happy to chat about RL/MTL+UNC/jobs (graduating soon🙂)
Han Guo Mohit Bansal
video: slideslive.com/38939074/dorb-…
virtual.2020.emnlp.org/paper_main.222…
arxiv.org/abs/2011.07635
![GoogleがToTTo(Table To Text)というデータセットを提案.
Wikipediaのテーブルからそれを忠実に説明するテキストを生成するというタスク.
paper: arxiv.org/abs/2004.14373 [EMNLP2020]
repo: github.com/google-researc…
blog: ai.googleblog.com/2021/01/totto-…](https://pbs.twimg.com/media/Er0IB4KVQAIxqYw.jpg "綿岡 (@Wataoka_Koki) on Twitter photo 2021-01-16 00:51:41 GoogleがToTTo(Table To Text)というデータセットを提案.
Wikipediaのテーブルからそれを忠実に説明するテキストを生成するというタスク.
paper: arxiv.org/abs/2004.14373 [EMNLP2020]
repo: github.com/google-researc…
blog: ai.googleblog.com/2021/01/totto-…")
#EMNLP2020 We’ve released 1.2 million reviews in 6 languages in the Multilingual Amazon Reviews Corpus (MARC)! w/ Y. Lu G. Szarvas Noah A. Smith arxiv.org/abs/2010.02573
The dataset was carefully cleaned and is balanced across star ratings. Try it out at registry.opendata.aws/amazon-reviews… 1/3
So many fascinating ideas at yesterday's #blackboxNLP workshop at #emnlp2020 . Too many bookmarked papers. Some takeaways:
1- There's more room to adopt input saliency methods in NLP. With Grad*input and Integrated Gradients being key gradient-based methods.
What kind of scientific ideas are more likely to spread beyond academia? In our recent #emnlp2020 findings paper, we carried out a large-scale data analysis to understand the knowledge transfer of scientific concepts across text corpora from research to practice. #CompSocSci
My 'present' to Hugging Face for being awarded as Best Demo Paper at #emnlp2020 .
mT5-small fine-tuned on tydiQA for multilingual QA. EM = 41.65
huggingface.co/mrm8488/mT5-sm…
We are honored to be awarded the Best Demo Paper for 'Transformers: State-of-the-Art Natural Language Processing' at #emnlp2020 😍
Thank you to our wonderful team members and the fantastic community of contributors who make the library possible 🤗🤗🤗
aclweb.org/anthology/2020…
Check out our #emnlp2020 paper now on arXiv: we develop an energy-based version of BERT called 'Electric' and show this essentially re-invents the ELECTRA pre-training method. Electric is useful for re-ranking the outputs of text generation systems. arxiv.org/pdf/2012.08561….
If you like ELECTRA, check out Electric, our new #emnlp2020 work that brought over energy-based model perspective ;) Electric produces fast pseudo-likelihood to improve speech recognition & translation reranking. Github: github.com/google-researc… & paper aclweb.org/anthology/2020…
Excited to share my #emnlp2020 paper w/ John Wieting, Mohit Iyyer on unsupervised text style transfer!
1. simple SoTA method using diverse paraphrasing
2. survey of eval in 23 papers + fixes
3. New dataset w/ 11 diverse styles
paper, code, *demo*: style.cs.umass.edu
1/9👇
Our 'Efficient Estimation of Influence of a Training Instance' will appear in @SustaiNLP2020/ #EMNLP2020
Dropout with instance-specific masks can efficiently evaluate influence (= how instances affect a prediction). Useful for analysis and data filtering. aclweb.org/anthology/2020…
Consistency of a Recurrent LM is at #emnlp2020 ♾️
Check out our talk and stop by the live session at 7pm EST (Nov 17)!
paper: aclweb.org/anthology/2020…
talk: slideslive.com/38939066/consi…
EMNLP page: virtual.2020.emnlp.org/paper_main.219…
<eos>
Hey <wake-word>, tell me about Punta Cana🇩🇴. Our #emnlp2020 paper introduces a conversational information-seeking dataset on geographic entities.
📜Paper + 📁Dataset + 💻Code: curiosity.pedro.ai
Gather 5H: Nov 18 18UTC
w/Paul Crook, Seungwhan Shane Moon, Stephen Wang 1/4
Congrats to the winners of the best paper awards at the #EMNLP2020 Workshop on Noisy Text!
noisy-text.github.io/2020/pdf/2020.…
noisy-text.github.io/2020/pdf/2020.…
Great keynote by Janet Pierrehumbert at #emnlp2020 about how psychology experiments test humans' ability to process language, and therefore how we should test NLP systems' ability to understand language. +1 for highlighting social variation in language!
