You should've seen the look on my face as someone whose lack of tolerance (and honestly, competence) for anything beyond minimum degree physics requirements played no small role in choosing CS as my major and becoming interested in AI!
Joking aside, many congrats to both! 🤩
BREAKING NEWS
The Royal Swedish Academy of Sciences has decided to award the 2024 #NobelPrize in Physics to John J. Hopfield and Geoffrey E. Hinton “for foundational discoveries and inventions that enable machine learning with artificial neural networks.”
@yoavartzi@sivareddyg Such an IMPORTANT issue too. I mean, just put yourself in the shoes of someone in Saratoga Springs deciding on which way to go for a bagel.
https://t.co/mb3QACZHrk
@tallinzen - heard: I’ve heard many good things about The Bear, Fallout, and Shogun. But they might be too anxiety-inducing for my current PhD student life. So I’m saving them for a special time.
And if you happen to need a new “good guilty pleasure” too, “The Franchise” looks promising.
@tallinzen Ah, ok, I think I get the taxonomy now!
- watched: Andor and Severance (I know you said you’ve watched Severance and might’ve watched Andor too. But those light up my soul, and I had to honor them.)
@sharakelyan I mean, in addition to this, a good rule of thumb for the reviewers is that if there’s something that you wouldn’t have even noticed with two \vspace’s, maybe do not use that as grounds for rejection!!
Sorry you have to deal with this 😢
Hey @soldni
Since you’re the most outspoken I know when it comes to tokenization 😅
Do you think that this is something that can be related to tokenization:
Please keep the discussion going till Summer 2026: “We hope to hold SCiL 2026 in conjunction with the meeting of the ACL, location and date TBD.”
https://t.co/aE6bglkhKo
Okay, okay here's a chapter on rejection sampling.
https://t.co/BQMYWmIh9L
I need to get comfortable writing in markdown, at least there's a canoncial link for it. will polish and add more discussion soon.
@m2saxon I have typed it (in an informal setting, of course). But English is my second language. So not sure if I’d feel differently about typing it if I were a native speaker.
here is my meticulously curated (and highly biased) summer paper reading list 📚:
Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks (2020)
╰╴ https://t.co/IbaElWz8na
LoRA: Low-Rank Adaptation of Large Language Models (2021)
╰╴ https://t.co/AArmQ3QaT8
Ring Attention with Blockwise Transformers for Near-Infinite Context (2023)
╰╴ https://t.co/G7fBGD09vV
Mamba: Linear-Time Sequence Modeling with Selective State Spaces
╰╴ https://t.co/G7fBGD09vV
The Era of 1-bit LLMs: All Large Language Models are in 1.58 Bits
╰╴ https://t.co/HTMr03Dt6N
Quiet-STaR: Language Models Can Teach Themselves to Think Before Speaking
╰╴ https://t.co/PWSIUfI4YP
Leave No Context Behind: Efficient Infinite Context Transformers with Infini-attention
╰╴https://t.co/glgLLlpqu3
The Illusion of State in State-Space Models
╰╴https://t.co/Yb4ZM1nahx
σ-GPTs: A New Approach to Autoregressive Models
╰╴https://t.co/MrfTbkge6Q
Rethinking LLM Memorization through the Lens of Adversarial Compression
╰╴https://t.co/CRx5ZLkyuF
Better & Faster Large Language Models via Multi-token Prediction
╰╴https://t.co/LgOsy1um63
xLSTM: Extended Long Short-Term Memory
╰╴https://t.co/2V32fpCdXn
Chameleon: Mixed-Modal Early-Fusion Foundation Models
╰╴https://t.co/kv4wlMajg5
Transformers Can Do Arithmetic with the Right Embeddings
╰╴https://t.co/MRwKcYgoyY
Scalable MatMul-free Language Modeling
╰╴https://t.co/HeqbyHvn5a