Tamanna Hossain-Kay
@tamannahossainkay.bsky.social
PhD Candidate at UCI NLP
Reposted by Tamanna Hossain-Kay
Artem Kirsanov, Chi-Ning Chou, Kyunghyun Cho, SueYeon Chung
The Geometry of Prompting: Unveiling Distinct Mechanisms of Task Adaptation in Language Models
https://arxiv.org/abs/2502.08009
The Geometry of Prompting: Unveiling Distinct Mechanisms of Task Adaptation in Language Models
https://arxiv.org/abs/2502.08009
February 13, 2025 at 8:05 AM
Artem Kirsanov, Chi-Ning Chou, Kyunghyun Cho, SueYeon Chung
The Geometry of Prompting: Unveiling Distinct Mechanisms of Task Adaptation in Language Models
https://arxiv.org/abs/2502.08009
The Geometry of Prompting: Unveiling Distinct Mechanisms of Task Adaptation in Language Models
https://arxiv.org/abs/2502.08009
Reposted by Tamanna Hossain-Kay
New video 📽️ from me:
I had the privilege of being a keynote speaker at the International Association for Safe & Ethical AI.
I focused on solutions and ideas for moving forward.
They made it public, and here it is.
Enjoy, and hope it's helpful! 🤗
oecdtv.webtv-solution.com/embed-or-en-...
I had the privilege of being a keynote speaker at the International Association for Safe & Ethical AI.
I focused on solutions and ideas for moving forward.
They made it public, and here it is.
Enjoy, and hope it's helpful! 🤗
oecdtv.webtv-solution.com/embed-or-en-...
IASEAI’25 Safe & Ethical AI Conference (Plenary)
oecdtv.webtv-solution.com
February 8, 2025 at 2:45 PM
New video 📽️ from me:
I had the privilege of being a keynote speaker at the International Association for Safe & Ethical AI.
I focused on solutions and ideas for moving forward.
They made it public, and here it is.
Enjoy, and hope it's helpful! 🤗
oecdtv.webtv-solution.com/embed-or-en-...
I had the privilege of being a keynote speaker at the International Association for Safe & Ethical AI.
I focused on solutions and ideas for moving forward.
They made it public, and here it is.
Enjoy, and hope it's helpful! 🤗
oecdtv.webtv-solution.com/embed-or-en-...
Reposted by Tamanna Hossain-Kay
Remember this @neuripsconf.bsky.social workshop paper? We spent the past month writing a newer, better, longer version!!! You can find it online here: arxiv.org/abs/2502.00561
February 4, 2025 at 3:28 PM
Remember this @neuripsconf.bsky.social workshop paper? We spent the past month writing a newer, better, longer version!!! You can find it online here: arxiv.org/abs/2502.00561
Reposted by Tamanna Hossain-Kay
What to know about DeepSeek
youtu.be/0eMzc-WnBfQ?...
In which we attempt to figure out MoE, o1, scaling, tech reporting, modern semiconductors, microeconomics, and international geopolitics.
youtu.be/0eMzc-WnBfQ?...
In which we attempt to figure out MoE, o1, scaling, tech reporting, modern semiconductors, microeconomics, and international geopolitics.
How DeepSeek Changes the LLM Story
YouTube video by Sasha Rush 🤗
youtu.be
February 4, 2025 at 3:41 PM
What to know about DeepSeek
youtu.be/0eMzc-WnBfQ?...
In which we attempt to figure out MoE, o1, scaling, tech reporting, modern semiconductors, microeconomics, and international geopolitics.
youtu.be/0eMzc-WnBfQ?...
In which we attempt to figure out MoE, o1, scaling, tech reporting, modern semiconductors, microeconomics, and international geopolitics.
Reposted by Tamanna Hossain-Kay
ICYMI – we released a new model today, proving yet again that the performance gap between open vs. closed models is shrinking. Meet Tülu 3 405B.
January 30, 2025 at 9:10 PM
ICYMI – we released a new model today, proving yet again that the performance gap between open vs. closed models is shrinking. Meet Tülu 3 405B.
Reposted by Tamanna Hossain-Kay
Read only the first 1-2 sentences of each and go with your gut. You'll likely get the perfect score!
Kind of thing where I probably prefer an unaligned model output to an aligned one..
Kind of thing where I probably prefer an unaligned model output to an aligned one..
November 21, 2024 at 2:19 AM
Read only the first 1-2 sentences of each and go with your gut. You'll likely get the perfect score!
Kind of thing where I probably prefer an unaligned model output to an aligned one..
Kind of thing where I probably prefer an unaligned model output to an aligned one..