Joe Stacey

@_joestacey_

NLP postdoc at @SheffieldNLP Ex @Imperial_NLP PhD, @Apple AI/ML Scholar, @UCL MSc Model robustness and now uncertainty quantification

Joined September 2020

2,613

Following

2,497

Followers

Replies Reposts Media

Pinned
Joe Stacey
@_joestacey_
Apr 15
Excited to share my first postdoc paper with @SheffieldNLP ! 🤩 In this work we argue that supervised uncertainty quantification (UQ) needs better evaluation Want to know more? Here's a little summary 🧵
6
Joe Stacey
@_joestacey_
Aug 26
Sure, terrible pay (in the UK at least), career progression is hard and mostly requires relocation, and no job stability at early stages of your career But wow academia is amazing! So stimulating, so enjoyable, and with so many cool problems you can work on. What's not to love?
Vivek Gupta
@keviv9
Aug 26
Serious advice to young Ph.D. graduates: Don’t join academia. It’s not worth it now.
4
Joe Stacey
@_joestacey_
Aug 22
One skill I do not have is coming up with great names for papers, how do people do it?? Do people use LLMs for it, or is it better without?
4
Joe Stacey
@_joestacey_
Aug 22
Everyone I know from Budapest tells me it’s such a cool city, so can’t wait to visit for #EMNLP 😊 I’ll be there with our nice paper on probe robustness
Joe Stacey
@_joestacey_
Apr 15
Replying to @_joestacey_
This work has been really fun to work on! Massive thank you my fantastic collaborators @OrgadHadas @inuikentaro @benbenhh and @NafiseSadat You can find the paper here: arxiv.org/abs/2604.11662
1
Joe Stacey
@_joestacey_
Jul 16
🎯 bang on imo
Djamé..
@zehavoc
Jul 16
Replying to @_joestacey_
this is why I strongly prefer the rebuttal period of softconf. 900 chars per reviewer's answer. You were only here to explain or answer questions. No new experiments, no thread about who is going to have the last word.
1