🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

One core learning we had with Chameleon is that the intended form of the modality is a modality in itself. Visual Perception and Visual Generation are two separate modalities and must be treated as such; hence, using discretized tokens for perception is wrong.

So here's a story of, by far, the weirdest bug I've encountered in my CS career. Along with Maciej Wołczyk we've been training a neural network that learns how to play NetHack, an old roguelike game, that looks like in the screenshot. Recenlty, something unexpected happened.

lol.. wrappers be like

"A random half of panelists were shown a CV and only a one-paragraph summary of the proposed research, while the other half were shown a CV and a full proposal. We find that withholding proposal texts from panelists did not detectibly impact rankings." link.springer.com/article/10.100…

New essay: ML seems to promise discovery without understanding, but this is fool's gold that has led to a reproducibility crisis in ML-based science. aisnakeoil.com/p/scientists-s… (with Sayash Kapoor). In 2021 we compiled evidence that an error called leakage is pervasive in ML models

Nathan Benaich THEY WROTE ERROR BOUNDS SO IT’S REAL SCIENCE

Thoughts exist without language.

PLEASE release this custom html parse PLEASE 🙏

I agree with much of both @emilymbender.bsky.social’s #ACL2024 presidential talk and (((ل()(ل() 'yoav))))👾’s rejoinder, but I want to comment on just one aspect where I disagree with both: the definition and domain of CL vs NLP. 🧵👇

as SAC for EMNLP I was asked to read the discussions between the authors and reviewers and had every intention to do so but the length of the discussions is out of control. many tables with results of new experiments, hundreds of lines of code (!). bring back word limits please.

Unless you are an OpenAI employee working on improving their products, I don’t understand why such efforts are science. Why are we (question to faculty) spending taxpayer dollars in doing QA for a closed product by a well-capitalized company that does not give back to science?

Today, we release several Moshi artifacts: a long technical report with all the details behind our model, weights for Moshi and its Mimi codec, along with streaming inference code in Pytorch, Rust and MLX. More details below 🧵 ⬇️ Paper: kyutai.org/Moshi.pdf Repo:

Great workshop at AAAI about low-rank representations! These have important consequences for Neurosymbolic: Logical circuits can be understood as low-rank factorisations.

"How to train a model on 10k H100 GPUs?" has now been immortalized on my blog: soumith.ch/blog/2024-10-0…

This “English(/French/etc) is the new code” trope bemuses me, even as someone bullish about LLMs substantially changing how we code. A 🧵(1/9)

great opportunity to do a #PhD in #Europe in #ML #AI 🚨🚨🚨 I'll hire 2 students via #ELLIS this year, reach out if you want to do research in: - #reliable and #efficient ML in the wild - scalable #neurosymbolic #nesy AI - #lowrank representations - #tractable inference 🚨🚨🚨

i think this is the wrong question. yes, CS graduates are very bad in software development, and a dedicated LLM can be better. but put these graduates in a job, and some will develop to be "senior devs" at some point, capable of working on real-life, large systems. LLMs won't.

the term "experts" in "mixture of experts" in the context of LLMs is highly misleading and does way more harm than good in coming up with a conceptual representation of what this component brings to the table.

Papers at #EMNLP2024 #3 A counter-example to the frequently adopted mech interp linear representation hypothesis: Recurrent Neural Networks Learn … Non-Linear Representations Fri Nov 15 BlackboxNLP 2024 poster aclanthology.org/2024.blackboxn… CC Csordás Róbert Christopher Potts

(a) how did MMLU become the defacto standard benchmark every LLM is trying to beat? (b) it is estimated to contain 9% questions that human experts think are wrong. do we know if humans and models agree on which ones belong in this 9%?