Emilien Dupont

phd student in machine learning @oxcsml @UniofOxford 🐳 previously research intern @Apple, computational maths @Stanford, theoretical physics @imperialcollege

[Emilien Dupont @emidup]

We introduce 🌸✨ AlphaEvolve ✨🌸, an evolutionary coding agent using LLMs coupled with automatic evaluators, to tackle open scientific problems 🧑‍🔬 and optimize critical pieces of compute infra ⚙️

https://t.co/ZcGN5AaRf9

[Jean-François Ton @jeanfrancois287]

📢New Paper on Reward Modelling📢

Ever wondered how to choose the best comparisons when building a preference dataset for LLMs?

Our latest paper revives classic statistical methods to do it optimally!

Here’s a 🧵on how it works 👇
https://t.co/8TlYZCh6ra

[Richie (Yat Long) Lo @chrysalis_ai]

Take a look at our latest work. One of my proudest works so far :)

[Stephen James @stepjamUK]

🚨Important update from our Robot Learning Lab in London.
Following recent news, we’re moving on after a wonderful 2 years…
Today, we unveil 4 big pieces of research from our incredible team. Check out the compilation video and thread below to see our final work! 📽️👇

[Emilien Dupont @emidup]

Joint work with @hyunjik11, @MatthiasSBauer, @lucastheis, @schwarzjn_

[Emilien Dupont @emidup]

We build neural codecs from a *single* image or video, achieving compression performance close to SOTA models trained on large datasets, while requiring ~100x fewer FLOPs for decoding ⚡ #CVPR2024

https://t.co/Y7rvJ7Mj7e

[Emilien Dupont @emidup]

For technical details, please refer to the paper and code.

📜: https://t.co/LlBKqFoera
🧑‍💻: https://t.co/VGcH2nkUUm
⚙️: https://t.co/Y7rvJ7Mj7e

We hope this is a step towards making neural codecs a practical reality ✨

[Emilien Dupont @emidup]

Work done with the amazing @ber24, @BarekatainAmin, @SashaVNovikov, @matejbalog, @Mpawankumar123, @franciscuto, @JSEllenberg, @PengmingWang, Omar Fawzi, @pushmeet, @AlhusseinFawzi!

[Emilien Dupont @emidup]

We present #FunSearch in @Nature today - a system combining LLMs with evolutionary search to generate new discoveries in math and computer science! 👩‍🔬🔬✨

[Google DeepMind @GoogleDeepMind]

Introducing FunSearch in @Nature: a method using large language models to search for new solutions in mathematics & computer science. 🔍

It pairs the creativity of an LLM with an automated evaluator to guard against hallucinations and incorrect ideas. 🧵 https://t.co/MC5ttgvZeM

[Jin Xu @jinxu06]

We construct neural processes by iteratively transforming a simple stochastic process into an expressive one, similar to flow/diffusion-based models, but in function space!

Join us at our #NeurIPS2023 poster session: https://t.co/UBYmJtnOwN on Wednesday morning!

[Miguel Angel Bautista @itsbautistam]

Introducing Manifold Diffusion Fields (MDF), our new work on learning generative models over fields defined on curved geometries. This is joint work with our intern @Ahmed_AI035 (who hasn’t even started his PhD yet!) and @jsusskin at @Apple MLR https://t.co/EcupOb38b6 🧵

[Jin Xu @jinxu06]

We introduce a new class of stochastic process models, which are constructed by stacking sequences of neural parameterised Markov transition operators in function space.
🗞️ https://t.co/Fw3zbkn1bP
w/ @emidup @kasparmartens @tom_rainforth
@yeewhye
A thread

[Hyunjik Kim @hyunjik11]

Drop by our #ICLR2023 workshop tmrw (Thurs) on "Neural Fields Across Fields: Methods and Applications of INRs"!
Schedule: https://t.co/YH6aOarN2Z
Accepted Papers: https://t.co/2Rr5k7DQBu
In-Person: Room MH4 (room used for poster sessions)
Virtual: https://t.co/jSzxyHxBOT

[Jonathan Richard Schwarz @schwarzjn_]

Very happy to announce that our latest paper on Neural data compression with INRs, Meta Learning & Sparse Subnetwork selection has been accepted to #ICML2023 (Scores 7, 7, 7). 1/N

Paper: https://t.co/qYt6LPpYD6

[Bobby @bobby_he]

Can deep transformers be trained without skip connections nor normalisation layers?

Our ICLR 2023 paper shows you how, using wide NN signal propagation ideas. We hope this can potentially pave the way to more efficient deep LLMs! (1/9)

Paper: https://t.co/DPpTQ9VT39

[Hyunjik Kim @hyunjik11]

Previously we had introduced *functa*, a framework for representing data as neural functions (aka neural fields, INRs) and doing deep learning on them.
In our recent work *spatial functa* we show how to scale up the approach to ImageNet-1k 256x256.
📝https://t.co/Gw4Fu37V4W