Dr. Matsubara !!!
Big thanks to my advisor, Prof. Marco Levorato, and @s_mandt for being part of my committee members!
The story of my PhD journey was pretty complex and too long to be posted here
work is NOT reproducible with the code, and requested update
The author has never responded to the issue, but I just noticed that it was silently deleted
Code: github.com/yoshitomo-mats
論文: link.springer.com/chapter/10.100
プレプリント: arxiv.org/abs/2011.12913
Intern Positions
Our team in Amazon Alexa AI is hiring Applied Scientist Interns!
If you have background in dialogue systems/information retrieval, send me a DM!
It should also be a great research opportunity
e.g., 2 of 2 my AS intern works were published (EMNLP & SIGIR)
transformers, accelerate & datasets packages for deep learning & knowledge distillation experiments with 
LOW coding cost
github.com/yoshitomo-mats
Paper, new results, trained models and Google Colab are 
1/n
Feature requests and discussions are always welcome!
Ask me at its Discussions page or here 
github.com/yoshitomo-mats
I want to give my wife a huge credit
He is definitely a life changer, and I'm very excited for our new endeavor with him 
(10.2%, 1,197 of 11,725 reviewers)
This is back-to-back(-to-back) award wining from NeurIPS 2022
neurips.cc/Conferences/20
Our team is hiring another Research Scientist
Send me a DM if you have strong research background in deep learning and NLP and want to be considered for the position
Paper+Datasets
Our great intern 's work was accepted at #ACL2023NLP (Findings)
TL;DR: We proposed cross-lingual knowledge distillation to train non-English QA models with NO annotated data
Preprint: arxiv.org/abs/2305.16302
Details+Datasets
Remember that ALL authors MUST have complete OpenReview (OR) profiles
If your coauthors don't have OR profiles, you won't be able to add them
Check this page and have your coauthors create complete profiles
cvpr.thecvf.com/Conferences/20
: github.com/yoshitomo-mats
論文
:Plagiarism Alert
So unhappy to share that a paper recently accepted to ELSEVIER Neurocomputing sciencedirect.com/science/articl
plagiarized at least 3 of our papers on split computing
arxiv.org/abs/2007.15818
dl.acm.org/doi/abs/10.114
dl.acm.org/doi/abs/10.114
More details...
1/n
Same as last year, I collected #ICLR2024 review data, modifying my previous script for OpenReview APIv2
Here are the histograms based on 7,331 submissions
Hope this helps!
Thank you Papers with Code team for the recognition and this cool package
I'll keep contributing to this community by publishing papers with code and ML OSS for & lowering barriers to reproducible studies
Looking forward to working on exciting research projects beyond what I've done as an intern
Sequential Reranking: dl.acm.org/doi/abs/10.114
Multi-Head Student:
e.g., You can use MobileNetV3 by as a teacher/student by editing a sample config (1st, resnet152) like 2nd/3rd
