Interesting speech-recognition SOTA from Google. Would be nice to see how their data-augmentation strategy performs with other non-LAS models. ai.googleblog.com/2019/04/specau…

The war between ML frameworks has raged on since the rebirth of deep learning. Who is winning? Horace He's data analysis shows clear trends: PyTorch is winning dramatically among researchers, while Tensorflow still dominates industry. #PyTorch #Tensorflow thegradient.pub/state-of-ml-fr…

This is mental.

It took me a while to recap bc some papers are complex. Key points: - Destructing black box: convergence, generalization, neural tangent kernel - New approaches: Bayesian & uncertainty, graph, convex optimization - Neuroscience & bio-inspired algos - ... huyenchip.com/2019/12/18/key…

Sounds like Berkeley instructors are handling #COVID19 well 😂

Delighted that our long paper "Neural Generation of Dialogue Response Timings" was accepted for #acl2020nlp. The work was done with Naomi Harte at the The ADAPT Centre in TCD. #NLProc

Since the paper "Attention Is All You Need", so many new things have happened to improve the Transformer model, e.g. to make the attention span longer, to reduce memory & compute cost, etc. That's what my post is about - lilianweng.github.io/lil-log/2020/0… 🤓

absolutely devastated after going down to pitch in at the capitol hill autonomous zone organizing tent only to find out the anarchist commune already has an L2 AWS cloud solutions engineer

Seems like a lot of people are working on their EMNLP reviews at the moment! Please remember that the following are NOT valid primary reasons for rejection, and the authors will be able to appeal to these guidelines in the rebuttal: 2020.emnlp.org/blog/2020-05-1…

Cool project looking at false positives for wakewords. I especially appreciate the direct links to edit privacy settings. unacceptable-privacy.github.io

It is not just the data. Popular compression techniques amplify bias in deep neural networks. Work co-led with Nyalleng Moorosi, w Gregory Clark, Samy Bengio and Remi Denton "Characterizing the bias of compressed models" -- pre-print - arxiv.org/abs/2010.03058

TWO positions in Trinity College Dublin, Ireland, available from Dec 1st, 2020 for 14 months. A Research Assistant (qualified to Masters level) A Research Fellow (holds a PhD) Please email [email protected]<mailto:[email protected]> for a more detailed description of either role

#aclnlp2022 will be in Dublin, Ireland. #NLProc 60th ACL! 😲

Has your neural net ever NaN’d so hard you thought about chucking your laptop in the trash and moving to British Columbia to be a tree planter instead?

I have a new paper on how to represent part-whole hierarchies in neural networks. arxiv.org/abs/2102.12627

"This is the story of how affect recognition came to be part of the AI industry, and the problems that presents." An excerpt from #AtlasofAI is out today in The Atlantic drawn from my chapter on emotion AI. Thanks to Adrienne LaFrance for editing 🙏📚✍ theatlantic.com/technology/arc…

Here's the December 2021 release of draft chapters for Speech and Language Processing, just in time for Jim and I to wish you all a Happy New Year! Enjoy! web.stanford.edu/~jurafsky/slp3/

There is a lot of talk on my timeline of #NLProc / #AI walls being hit and/or smashed, goalposts being missed and/or moved, etc. I don’t like the notion of goalposts (as if “intelligence” were a binary), but here’s what I think is required for linguistic intelligence: »

I had a blast last week at #INTERSPEECH2023. Here's a round-up of a few of the papers. link.medium.com/zjoFUGJQJCb