Jihoon Han

[Emily Alger @MLE_Alger]

Inspired to get creative with my poster for @ictmc2024 this week! I shared my work on bayesian generalised linear mixed models to optimise treatment dosage in early phase trials using Patient-reported Outcomes. #ICTMC2024 #clinicaltrials

[Soo Ahn Lee @SooAhnLee]

Come see my poster #2518 on Wed & Thurs about modeling brain responses of sustained pleasure and pain dynamically changing over time! Working with @choongwanwoo @torwager #OHBM2024

[Eui-Jin @euigenie_]

#OHBM2024 🧠
Check out my poster #2500 (Wed/Thur) on how multi-echo improves the pain prediction model!
Working with @SuhwanGim, @dong_hee_lee, @_LeeJaeJoong_ and @choongwanwoo

[Youngeun (Grace) Park @YoungeunPark_]

#OHBM2024 🧠🇰🇷
Come to my poster #2519 on Wed/Thurs (June 26-27) about a personalized pain biomarker using densely-sampled fMRI data from one individual🙋🏻‍♂️ (@choongwanwoo) and its comparison to a population-level model (N=124).
@sungwoo320

[andrew gao @itsandrewgao]

Oh

[huze @HuzeYann]

0/. Brain is not ViT. we scored 70.8 in the Algonauts 2023 visual brain competition, w/o ensemble we can do 66.8 score. other teams (including me in the past) are struggling with 60.

“memory” is the secret.

paper: https://t.co/v8BXs25oz1
code&webpage: https://t.co/Pd31thAzDA

[Sasha Rush @srush_nlp]

Blog Post (w/ @gail_w): On "Thinking Like Transformers"

In which, I get a bit obsessed with learning how to code in Transformer lang🤖.

https://t.co/Nb6G52vuiK

(You can follow along or do the exercises yourself in a colab notebook.)

[Choong-Wan (Wani) Woo @choongwanwoo]

A new preprint from our lab on the brain representations and dynamics of spontaneous thought using a free association-based thought sampling task https://t.co/F4hXQ5lQJi

[Alexander Huth @alex_ander]

Excited to share this work from @AmandaLeBel3 on a detailed breakdown of language representations in the cerebellum! Take-away 1: it represents language! But data are consistent with cerebellum ONLY representing the HIGHEST-LEVEL, MOST CONCEPTUAL aspects of language.

[Jae-Joong Lee @_LeeJaeJoong_]

Excited to announce that our recent work is finally out in @NatureMedicine, with @choongwanwoo, @torwager, @HongjiKim, @martaceko, @BoyongPark1, @SooahnLee, Hyunjin Park, Mathieu Roy, and Seong-Gi Kim!

https://t.co/rRY8OGGWXq

Thanks to all co-authors for their contributions.

[Hasson Lab @HassonLab]

New preprint from the Hasson Lab with @ArielYGoldstein and many collaborators at NYU and Google Research showing deep connections between the way the human brain and deep language models like GPT2 process language in natural settings: https://t.co/GtmgzGsi7n

[Oriol Vinyals @OriolVinyalsML]

Recent conversation with a friend:

@ilyasut: what's your take on https://t.co/fqVhQNaBWQ?
@OriolVinyalsML: my take is: farewell convolutions : )

[Jean-Rémi King @JeanRemiKing]

Language processing in brains and deep networks: computational convergence and its limits.

Check out our latest preprint: https://t.co/HBVkNVtxUl, by @c_caucheteux and I

[summerfieldlab @summerfieldlab.bsky.social @summerfieldlab]

Neuroscientists are turning to deep networks as computational theories for biology. But can they be used to make falsfiable predictions about neural representation, dynamics, behavior, and cognition? Answers in a new review with @SaxeLab @steph_nelli
https://t.co/DBvcCIcQn1

[Google AI @GoogleAI]

Announcing XTREME, a new #NaturalLanguageProcessing benchmark for cross-lingual generalization, which covers 40 typologically diverse languages using nine tasks that collectively require reasoning about different levels of syntax or semantics. Learn more ↓https://t.co/F7pgTQdbuo

[Dive into Deep learning (D2L.ai) @D2L_ai]

We have re-organized Chapter: NLP pretraining (https://t.co/gUKNmrySw7) and Chapter: NLP applications (https://t.co/I5QvkSPEVa), and added sections of BERT (model, data, pretraining, fine-tuning, application) and natural language inference (data, model).

[Quoc Le @quocleix]

Cool results from our collaboration with colleagues at @DeepMind on searching for new layers as alternatives for BatchNorm-ReLU. Excited with the potential use of AutoML for discovering novel ML concepts from low level primitives.

[Choong-Wan (Wani) Woo @choongwanwoo]

Our new paper is finally out "Toward a unified framework for interpreting machine-learning models in neuroimaging" on Nature Protocols with @la_da_k @sungwoo320 @taesupmoon, Juyeon Heo, Sungmin Cha, and @torwager https://t.co/ycJppox8qu

[AI at Meta @AIatMeta]

With 4.5B parallel sentences in 576 language pairs, CCMatrix is the largest data set of high-quality, web-based bitexts for training translation models. Now Facebook AI is sharing tools for other researchers to use this corpus for their work. https://t.co/uvBbfjPTk5

[Kyle McDonald @kcimc]

um.. google's latest chatbot is 😳 https://t.co/jJyW8pMeMZ