doctoral student at brown-- nlp, cognitive science. he/him.
Joined September 2016
- Tweets46
- Following114
- Followers115
- Likes108
Aaron Traylor retweeted
Mechanistic interpreters: Aaron Traylor and Jack Merullo of the LUNAR Lab got under the hood of a ChatGPT-like AI to show that this type of AI can adopt some of the same processes the human brain uses. bit.ly/Carneyfabcollab @LnccBrown @Brown_NLP @attraylor @jack_merullo_
does anyone have any work talking about meme morphology, i guess specifically the productive "-champ" morpheme that comes from PogChamp?
e.g. maldchamp, sadgechamp
Aaron Traylor retweeted
For those (like me) who weren't really following ICLR last week, some more context on this cool work led by Charles Lovering:
Checkout our new @iclr_conf paper w/ @tallinzen: We found that models use “good” features (e.g. S-V agreement) over “bad” features (e.g. lexical overlap) when they’re easier to extract from the model’s pre-trained representations.
Paper: openreview.net/pdf?id=mNtmha…
(1/3)
Aaron Traylor retweeted
Applying to the CS PhD or MS/PhD program at @umasscs?
As part of the PhD Applicant Support Program, we're offering a **live Info Webinar on November 23rd, 11AM EST**! A panel of PhD students will be ready to answer your questions.
Register here: cics.umass.edu/event/phd-app…
#phdchat
Aaron Traylor retweeted
if you are applying to PhD programs in CS, this is for you! specifically, we've seen lots of opportunities tailored towards applicants from underrepresented groups, so several grad students at @browncs compiled a list that we hope is generally helpful cs.brown.edu/degrees/doctora… 1/6
not back home to read my copy of @GretchenAMcC’s “because internet” so my mom is reading it and she loves it! she’s figuring out how to include it in her 1st-3rd grade class’ study of the history of written language
Aaron Traylor retweeted
Brandon Prickett, Aaron Traylor, and Joe Pater. 2019. Learning Reduplication with a Variable-Free Neural Network.
"...we successfully simulate the three experiments presented by Marcus et al. (1999)..."
works.bepress.com/joe_pater/…
@GaryMarcus @attraylor @bpricket @ryandcotterell
Aaron Traylor retweeted
Who said AI will take our jobs? I just asked BERT to generate *new scientific disciplines* by filling the blanks for: "I did my phd in MASK MASK for the last four years." favorites so far:
Space Linguistics
Quantum Arts
Food Archaeology
Educational Magic
Applied Theology
🤖❤️👩🎓
just preordered @GretchenAMcC's book on internet language!! the next 6 months of waiting for it to come in the mail will be hell!
Aaron Traylor retweeted
Ok, I realize now that
if you didn't do half a PhD studying human vision with fMRI, this quote doesn't make sense; and
"OMG" isn't an explanation;
also @goodfellow_ian is messaging me on chat asking great questions;
so let me just broadcast an explanation of why "OMG": 1/n
Aaron Traylor retweeted
.@strubell and @pat_verga accepting the Best Long Paper award at #emnlp2018 - congrats!
Aaron Traylor retweeted
First #emnlp2018 best long paper talk! Emma Strubell on "Linguistically-Informed Self-Attention for Semantic Role Labeling"
Aaron Traylor retweeted
1/ There's a bunch of UMass NLP papers at #emnlp2018 from @UMassCS and @LinguistsUMass! I'm doing this late so catching up on a few already presented. There were two workshop papers looking at linguistic properties of deep learning models:
Aaron Traylor retweeted
I've set up a discussion blog post with links to "Generative Linguistics and Neural Networks at 60" and the commentaries. Please contribute your thoughts!
@attraylor @ChrisGPotts @tallinzen @sleepinyourhat @GaryMarcus @adinamwilliams @ewandun @yoavgo
blogs.umass.edu/cogsci/2018/… …
Aaron Traylor retweeted
"To Live is to Suffer, to Survive is to Find Some Meaning in the Suffering: Randomized Hyperparameter Optimization for Batching Dynamic Computation Graphs"