@drugilsberg Profile picture

Matteo Manica

@drugilsberg

Research Scientist at @IBMResearch | Interested in AI and machine learning applications to science | Views are my own

Similar User
LIAC at EPFL photo

@SchwallerGroup

Teodoro Laino photo

@teodorolaino

Miruna Cretu photo

@MirunaCretu2

Philippe Schwaller (he/him) photo

@pschwllr

Luigi Bonati photo

@LuigiBonati

Oliver Schilter photo

@OSchilter

Molecular Machine Learning photo

@molecularML

Jannis Born photo

@JannisBorn

Francesca Grisoni photo

@fra_grisoni

Kevin M Jablonka photo

@kmjablonka

Maria Rodriguez Martinez photo

@MariaRoCompBio

BOjana Ranković photo

@6ojaHa

Merck Foundation photo

@MerckFoundation

Jean-Louis Reymond photo

@jrjrjlr

Ryan-Rhys Griffiths photo

@Ryan__Rhys

Pinned

🚀 Exciting news! Our paper on GT4SD has been published in @Nature_NPJ Computational Materials! Learn more about it here: doi.org/10.1038/s41524…. Try it out: GitHub - github.com/GT4SD/gt4sd-co… Demo apps on HF Spaces - huggingface.co/GT4SD


Matteo Manica Reposted

In "Language models can identify enzymatic binding sites in protein sequences", we identify binding site positions based on SMILES and protein AA sequences. Compared to predictions based on PFAM alignments, we significantly lower the false-positive rate. sciencedirect.com/science/articl…

Tweet Image 1

Matteo Manica Reposted

Friends, as the saying approximately goes: "Fortune favours the bald" 👴 👉 My proposal "HOLES: Higher-Order Learning of Essential Structures with Geometry and Topology" was successfully funded with a 1.5M € @ERC_Research Starting Grant! 👈 🤯🤯🤯 This will allow me to pursue…

Tweet Image 1

Matteo Manica Reposted

Our ICML paper: ML + quantum computing + optimal transport = 🚀! Learn distributions conditionally with quantum. Including a 24-qubit hardware run on the assignment problem (predicting doubly-stochastic matrices) @IBMResearch 📖 ibm.biz/qontot-icml 📹 ibm.biz/qontot-yt


Matteo Manica Reposted

This was done mostly by @maykcaldas and @DrChrisCollison - congrats on huge effort! We wrote for both newcomers to LLMs in chem and for experts. @maykcaldas also set-up a repo for continuing to track the field! Paper: arxiv.org/abs/2407.01603 Repo: github.com/ur-whitelab/LL…


Matteo Manica Reposted

Some news on my side 🎉 I will join @WUR as an assistant professor in September. My research remains at the intersection of bioinformatics and cheminformatics and I will focus on applications contributing to pharmacology and sustainability in (bio)chemical synthesis and ag.


Matteo Manica Reposted

Consider these fantastic opportunities in #Lausanne if you're into ML for Biomedicine and are looking for a PhD, PostDoc or master thesis! I can speak from experience and it has been super inspiring to work with @marianna_raps!

📢 We are hiring! 📢 Are you a MSc/PhD👩‍🎓👨‍🎓 passionate about #AI #ML in #cancer, looking for your next career step? Join our newly-founded AI/ML for Biomedicine group part of the Biomedical Data Science Center at @unil and @CHUVLausanne in beautiful Lausanne!🇨🇭🏔🌈 👇



Matteo Manica Reposted

🚀 Excited to share our latest work on how language models can revolutionize the identification of enzymatic binding sites in protein sequences! Check it out here: sciencedirect.com/science/articl… 🧬 #LLM #Proteins #Biocatalysis #Enzyme

Tweet Image 1

Matteo Manica Reposted

I'm very excited about our first paper on benchmarking LLMs for chemistry!

Are large language models superhuman chemists? We have been wondering about this, too and developed the first automated and comprehensive benchmark for evaluating large language models in chemistry and let LLMs compete with human chemists 👩🏼‍🔬 📝 arxiv.org/abs/2404.01475

Tweet Image 1


Matteo Manica Reposted

🚀🧪Today marks more than just news: it's a proud moment in which @Syngenta and @IBM teams celebrate the development and integration of cutting-edge technology into business operations through a long-lasting collaboration. shootsbysyngenta.com/success-story-… (1/5)


Matteo Manica Reposted

👨🏾‍🎓 👩‍🎓 Attention MSc candidates! 📢 are you passionate about digital innovation in chemistry? Here’s your chance! 🌐 Join @IBMResearch 🇨🇭for a funded MSc thesis and explore the frontiers of digital chemistry. Secure your spot now for 2024: 👇#MScThesis #NCCRCatalysis

Next stop MSc thesis? 📣 Apply now for an #NCCRCatalysis Young Talents Fellowship to conduct your funded Master’s thesis project on sustainable chemistry & catalysis 🌱🧪🌍 within our member labs in 🇨🇭! ⏰ 30 April 2024 📑 More info bit.ly/2024YoungTalen… 🔁 Appreciated 🫶🏻

Tweet Image 1


Matteo Manica Reposted

Our paper got accepted at #AAAI24! 🎉 We propose two new sequence generation algorithms with "error bars", by adapting beam search to conformal predictions. If you use LLMs for science, like predicting molecules or proteins that verify some conditions, check it out!

Here's our preprint on conformal sequence generation! w/ @malberts99 @MariaRoCompBio arxiv.org/abs/2309.03797 Two new algorithms for sequence prediction sets with guarantees: - Conformal sub-beams: pruning standard beams - Conformal dynamic beams: step-by-step conformal decoding.



Matteo Manica Reposted

The IBM Research Zürich lab is at #NeurIPS2023 in NOLA! Come chat with us about all things research and life in Zürich, if you haven’t already 😉 @IBMResearch

Tweet Image 1

Matteo Manica Reposted

Can transformers follow instructions? We explore this in: "Looped Transformers as Programmable Computers" arxiv.org/abs/2301.13196 led by Angeliki (@AngelikiGiannou) and Shashank (@shashank_r12) in collaboartion with the @Kangwook_Lee and @jasondeanlee Here is a 🧵

Tweet Image 1

Matteo Manica Reposted

Steering Large Language Models for Machine Translation with Finetuning and In-Context (virtual poster) with @duartealves @nunonmg @JoãoAlves @josepombal @RicardoRei7 @accezz @andre_t_martins Do not hesitate to pass by the poster and say hi to Duarte !


Matteo Manica Reposted

IBM and Meta launched a coalition of over 50 AI companies and research institutions that are pushing an open model of AI 🚀wsj.com/articles/meta-… via @WSJ


Matteo Manica Reposted

Still in disbelief about the positive achievements of this year! Our paper on "Uncovering Unique Concept Vectors through Latent Space Decomposition" is now in the @TmlrPub 🤩 Read it: openreview.net/pdf?id=LT4DXqU… Try it: github.com/maragraziani/c… And let us know what you think! [1/3]


Matteo Manica Reposted

My article on explainable machine learning for enzyme-catalysed reaction classification has just been published in @jcheminf More on it tomorrow (I want to try something new with a video tl;dr :-)) jcheminf.biomedcentral.com/articles/10.11…

Tweet Image 1

Matteo Manica Reposted

After I resisted for a long time, my students finally convinced me that it’s a good idea to create a group account. @Stefan_P_Schmid’s thorough and data-driven analysis revealed untapped market potential. Follow for group updates and to feel the DCL “vibe” 😉

Get ready for a *spacey* event 🚀🌌Join us for the inaugural lecture of @kjelljorner on Oct. 25! Lots of futuristic Research guaranteed! 🦾See you there



Matteo Manica Reposted

1/6 Molecular set representation learning Do we need molecular topology or are sets of vectors of atomic invariants (~ECFP0) enough? They can be; and enhance existing GNNs; and be used for binding affinity, molecular property, and rxn yield prediction. chemrxiv.org/engage/chemrxi…

Tweet Image 1

Matteo Manica Reposted

Great to see our work @IBMResearch on prompt driven retrosynthesis in @ACSCentSci highlighted by @OPRD_ACS alongside several former colleagues and collaborators! 🌟Highlight: pubs.acs.org/doi/10.1021/ac… 📰Paper: pubs.acs.org/doi/10.1021/ac… #NLP #MachineLearning #Chemistry #AI


Loading...

Something went wrong.


Something went wrong.