Picture

Seungwook Han @seungwookh

phd-ing @MIT_CSAIL, prev @MITIBMLab @columbia hanseungwook.github.io Joined June 2017
  • Tweets
    139
  • Followers
    411
  • Following
    589
  • Likes
    659
Seungwook Han retweeted
jaschasd Profile Picture

Jascha Sohl-Dickstein @jaschasd

3 days ago

Title: Advice for a young investigator in the first and last days of the Anthropocene Abstract: Within just a few years, it is likely that we will create AI systems that outperform the best humans on all intellectual tasks. This will have implications for your research and…

Seungwook Han retweeted
jxbz Profile Picture

Jeremy Bernstein @jxbz

5 days ago

I wrote this blog post that tries to go further toward design principles for neural nets and optimizers The post presents a visual intro to optimization on normed manifolds and a Muon variant for the manifold of matrices with unit condition number x.com/thinkymachines…

thinkymachines Profile Picture

Thinking Machines @thinkymachines

5 days ago

I wrote this blog post that tries to go further toward design principles for neural nets and optimizers The post presents a visual intro to optimization on normed manifolds and a Muon variant for the manifold of matrices with unit condition number x.com/thinkymachines…

seungwookh Profile Picture

Seungwook Han @seungwookh

4 weeks ago

Why do models forget less with RL than SFT?

jyo_pari Profile Picture

Jyo Pari @jyo_pari

4 weeks ago

Why do models forget less with RL than SFT?

Seungwook Han retweeted
jyo_pari Profile Picture

Jyo Pari @jyo_pari

a month ago

We have a fun collaboration of @GPU_MODE x @scaleml coming up! We’re hosting a week-long online bootcamp that explores the core components of GPT-OSS while also diving into cutting-edge research that pushes beyond what’s currently in GPT-OSS! For example, how can MoE's power…

seungwookh Profile Picture

Seungwook Han @seungwookh

2 months ago

uncertainty-aware reasoning, akin to how humans leverage our confidence

MehulDamani2 Profile Picture

Mehul Damani @MehulDamani2

2 months ago

uncertainty-aware reasoning, akin to how humans leverage our confidence

seungwookh Profile Picture

Seungwook Han @seungwookh

2 months ago

was actually wondering with @hyundongleee the fundamental differences between diffusion and autoregressive modeling other than the structure imposed in the modeling of the sequential conditional distribution and how they manifest. a poignant paper that addresses this thought

mihirp98 Profile Picture

Mihir Prabhudesai @mihirp98

2 months ago

was actually wondering with @hyundongleee the fundamental differences between diffusion and autoregressive modeling other than the structure imposed in the modeling of the sequential conditional distribution and how they manifest. a poignant paper that addresses this thought

seungwookh Profile Picture

Seungwook Han @seungwookh

2 months ago

omw to trying this out 👀

pika_labs Profile Picture

Pika @pika_labs

2 months ago

omw to trying this out 👀

seungwookh Profile Picture

Seungwook Han @seungwookh

2 months ago

how particles can act differently under different scales and conditions and how we can equip it as part of design is cool

MITarchitecture Profile Picture

MIT Architecture @MITarchitecture

9 months ago

how particles can act differently under different scales and conditions and how we can equip it as part of design is cool

Seungwook Han retweeted
LakerNewhouse Profile Picture

Laker Newhouse @LakerNewhouse

2 months ago

[1/9] We created a performant Lipschitz transformer by spectrally regulating the weights—without using activation stability tricks: no layer norm, QK norm, or logit softcapping. We think this may address a “root cause” of unstable training.

seungwookh Profile Picture

Seungwook Han @seungwookh

3 months ago

But actually this is the og way of doing it and should stop by E-2103 to see @jxbz and Laker Newhouse whiteboard the whole paper.

jxbz Profile Picture

Jeremy Bernstein @jxbz

3 months ago

But actually this is the og way of doing it and should stop by E-2103 to see @jxbz and Laker Newhouse whiteboard the whole paper. https://t.co/NjV3qnxCaK

Seungwook Han retweeted
jyo_pari Profile Picture

Jyo Pari @jyo_pari

3 months ago

If you are interested in questioning how we should pretrain models and create new architectures for general reasoning - then checkout E606 @ ICML, our position by @seungwookh and I on potential directions for the next generation reasoning models!

seungwookh Profile Picture

Seungwook Han @seungwookh

3 months ago

At #ICML 🇨🇦 this week. I'm convinced that the core computations are shared across modalities (vision, text, audio, etc). The real question is the (synthetic) generative process that ties them. Reach out if you have thoughts or want to chat!

seungwookh Profile Picture

Seungwook Han @seungwookh

3 months ago

wholeheartedly agree with this direction that games can be a good playground for learning reasoning. makes us think what other synthetic environments we can design and grow over complexity

Benjamin_eecs Profile Picture

Bo Liu (Benjamin Liu) @Benjamin_eecs

3 months ago

wholeheartedly agree with this direction that games can be a good playground for learning reasoning. makes us think what other synthetic environments we can design and grow over complexity

Trends for United States

Jane Goodall
100 B posts
Santana
38,1 B posts
Mason Miller
3.558 posts
Yom Kippur
43,6 B posts
Barca
184 B posts
Xbox
95,6 B posts
Cease
17,1 B posts
Olmo
41,6 B posts
Pope
74,1 B posts
Caresha
7.593 posts
Rashford
45,5 B posts
Nuno Mendes
65,9 B posts
Manny
8.949 posts
Barcelona
239 B posts
Game 3
45 B posts
Pedri
61,3 B posts
Yamal
161 B posts
Happ
2.720 posts
Halloween
200 B posts
Padres
47,1 B posts

You might like

PhD student at @MIT_CSAIL working on deep learning || interning at @AIatMeta, previously at @GoogleDeepMind
Shobhita Sundaram
@shobsund
1K Followers
Research scientist @RedHat & @MITIBMLab, PhD @Harvard. Research interests: information theory, statistical learning theory, trustworthy machine learning.
Hao Wang
@HW_HaoWang
122 Followers
PhD in probabilistic programing @tudelft advised by @sebdumancic. research engineer at MIT-IBM Lab.
aldo pareja
@ParejaAldo
171 Followers
eng at @vannevarlabs, research @FSIStanford, tinkering @stanfordauv 🌊
Selena Sun
@selenxsun
249 Followers
at openai. im just here to read papers.
minyoung huh (jacob)
@minyoung_huh
180 Followers
PhD student @berkeley_ai | Previous: CS + Music @MIT | underfit to the demands of reality
Stephanie Fu
@xkungfu
592 Followers
Senior Research Scientist and PI @MITIBMLab, @IBMResearch. Applying statistics and information theory to generative AI. PhD @umich, postdoc @harvard.
Kristjan Greenewald
@KGreenewald
240 Followers
PhD student @MIT, Student Researcher @GoogleDeepMind
idan shenfeld
@IdanShenfeld
276 Followers
Boston University NLP PhD @llamagrp. Researching interpretability and generalization of language models. Made in China. (Biased) Opinions are my own.
Zilu Tang (Peter)
@Zilu_Tang_Peter
233 Followers
Postdoc at Stanford CS. Previously: MIT PhD, Cornell BS
Sam Park
@smsampark
435 Followers
CS PhD student at MIT, interested in reinforcement learning and biologically inspired computation
Zhang-Wei Hong
@ZhangWeiHong9
162 Followers
AI Visualization & (re)Interpretability Researcher @IBMResearch @GeorgiaTech
Ben Hoover
@Ben_Hoov
1K Followers
Research Engineer, LLM Alignment @Redhat AI Innovation Team
Shiv
@shivsr98
51 Followers
Research Scientist @Meta FAIR | CS PhD @UT Austin | Former Research Intern @DeepMind, @Nvidia, @Baidu
Bo Liu
@cranialxix
483 Followers
PhD Student @MIT_CSAIL RS Intern @SakanaAILabs RL, Meta-Learning, Emergence, Open-Endedness, ALife
Akarsh Kumar
@akarshkumar0101
2K Followers