Kevin Jablonka
@kjablonka.com
Trying to teach computers how to design materials. Leading a research group at FSU Jena/HIPOLE Jena. Increasing entropy since 1996.
When an "AI scientist" produces a result, is that knowledge? Philosophers I taught last semester kept reminding me: knowledge is justified true belief. The process matters. New preprint: "AI scientists produce results without reasoning scientifically." 25,000+ runs.
I just published a Substack post highlighting ChemBench by Mirza and coauthors, which evaluates how well LLMs perform on chemistry tasks. From strengths in structural analysis to key limitations, it’s a timely look at AI’s role in chemical reasoning. open.substack.com/pub/bravoaba...
Evaluating chemical reasoning capabilities of LLMs with ChemBench
As large language models (LLMs) sweep across scientific domains, chemistry confronts a critical question: can they rival—or even surpass—trained chemists in core reasoning tasks?
open.substack.com
Our team has been spending a lot of time building evaluations for machine learning systems. We have learned some lessons and wrote them down arxiv.org/abs/2503.10837
🚀ChemBench just leveled up! We’re thrilled to announce the latest release of ChemBench—now smarter and smoother! Dive into benchmarking any chemistry AI model with our revamped framework, designed for flexibility and ease. #ChemistryAI #MachineLearning #OpenScience #Innovation
Many of our benchmarks underwent a large revision in the last weeks. We now host HuggingFace spaces for them and the dataset. The revised article and post below give more details
Jablonka Lab (Lab for AI for Materials)@jablonkagroup.bsky.social · last yr.🚀Our revised MaCBench paper is now on arxiv! arxiv.org/pdf/2411.16955 Key updates! 🌟Robust reproducibility: 5x experiment runs + error bars for statistical confidence 🌟Full dataset & leaderboard: Now live on HuggingFace with model comparisons huggingface.co/spaces/jablo...
From @kjablonka.com, @mvictoriagil.bsky.social, @pepe-marquez.bsky.social and colleagues. 'From text to insight: large language models for chemical data extraction' #OpenAccess 🔓 pubs.rsc.org/en/content/a...
From text to insight: large language models for chemical data extraction
The vast majority of chemical knowledge exists in unstructured natural language, yet structured data is crucial for innovative and systematic materials design. Traditionally, the field has relied on m...
pubs.rsc.org
We at @digital-discovery.bsky.social are very happy to announce a new paper type called "Commit". Inspired by version control systems such as git, the idea is that if you have an update on a short and pointed publication, you can send it as a commit. We envision commits to be co-cited with the
I've been thinking about how reasoning models will change AI applied to science. The recent papers from Deepseek/AI2/MoonShotAI are showing that we can exceed humans on reasoning tasks and I've written up some reflections on the consequences diffuse.one/p/d1-007
diffuse.one
andrew white's blog.
diffuse.one
Super excited that we are able to advertise 12(!) fully funded PhD positions in the field of biomolecular technology of protein interactions. Check out the exciting projects, spread the word and join BioToP. #biotechnology #compchem @bokuvienna.bsky.social @fwf-at.bsky.social biotop.boku.ac.at
BioToP stands for "Biomolecular Technology of Proteins" and aims in interdisciplinary training of PhD students.
biotop.boku.ac.at
If you are excited about ML for materials science and chemistry, we might have good news for you. We are still hiring on all levels. Simply connect by submitting your application via forms.fillout.com/t/eoGA7AhnAKus.
Join LAMA
Made with Fillout, the best way to make forms, surveys and quizzes your audience will answer.
forms.fillout.com
Our data extraction tutorial is now online in Chem. Soc. Rev. The notebooks can be run using the #jupyter4nfdi service from #base4nfdi. 📝 Paper: pubs.rsc.org/en/content/a... 💻 JupyterHub: t1p.de/matextract-cpu 📚 Online book: matextract.pub 📽️ intro from @kjablonka.com! 👇
From text to insight: large language models for chemical data extraction
The vast majority of chemical knowledge exists in unstructured natural language, yet structured data is crucial for innovative and systematic materials design. Traditionally, the field has relied on m...
pubs.rsc.org
Kevin Jablonka@kjablonka.com · 2y ago🎅🏼 A small early Christmas present from our team. To celebrate the publication of our data extraction tutorial in Chem Soc Rev, we made it easy to run it — without any installation — on a JupyterHub of the Base4NFDI. 🎥 Video intro to the JupyterHub deployment: youtu.be/l-5QNUo1fcU
Doing good science is 90% finding a science buddy to constantly talk to about the project.
10 minutes ago I am excited to share a perspective on the much-needed topic of hashtag#safety for hashtag#selfdrivinglaboratories. As the field progresses, understanding the challenges and gaps in building safe setups will be crucial for scaling up this technology! doi.org/10.26434/che...
Steering towards safe self-driving laboratories
The past decade has witnessed remarkable advancements in autonomous systems, such as automobiles that are evolving from traditional vehicles to ones capable of navigating complex environments without ...
doi.org
FutureHouse is launching an independent postdoctoral fellowship program for exceptional researchers who want to apply our automated science tools to specific problems in biology and biochemistry, in collaboration with world-leading academic labs. 1/
Lot of cool stuff in here. Consistent with my working hypothesis that the main scientific utility of LLMs at the moment is plain old NLP
Kevin Jablonka@kjablonka.com · 2y ago🎅🏼 A small early Christmas present from our team. To celebrate the publication of our data extraction tutorial in Chem Soc Rev, we made it easy to run it — without any installation — on a JupyterHub of the Base4NFDI. 🎥 Video intro to the JupyterHub deployment: youtu.be/l-5QNUo1fcU
🎅🏼 A small early Christmas present from our team. To celebrate the publication of our data extraction tutorial in Chem Soc Rev, we made it easy to run it — without any installation — on a JupyterHub of the Base4NFDI. 🎥 Video intro to the JupyterHub deployment: youtu.be/l-5QNUo1fcU
From text to insight: large language models for chemical data extraction
The vast majority of chemical knowledge exists in unstructured natural language, yet structured data is crucial for innovative and systematic materials design. Traditionally, the field has relied on m...
pubs.rsc.org
A holiday treat for direct air capture of CO₂ with metal-organic frameworks 🎄 A #preprint led by @hspark1212.bsky.social & @yunsunglim.bsky.social with: - Fine-tuned machine learning force field MACE-DAC - Integrated simulation package DAC-SIM chemrxiv.org/engage/chemr... #CompChemSky #ChemSky
Elucidating structures from spectra using multimodal embeddings and discrete optimization Representations of NMR spectra contrasted against molecular graph representations. Contrastive learning used for learned database look-up -- nice! P: chemrxiv.org/engage/chemr...
Combining the best of two worlds, de novo sequencing and database search with the awesome Casanovo and the super-fast Rust-based Sage for #metaproteomics. Yeaha!
Matthew Collins@matthewcollins.bsky.social · 2y agoI have been talking about it for ages - and now it is finally out *Orthrus: an AI-powered, cloud-ready, and open-source hybrid approach for metaproteomics* www.biorxiv.org/content/10.1...
Our multimodal follow-up to ChemBench. Our team went to the lab to create images to challenge leading vision language models. Check the paper arxiv.org/abs/2411.16955 to learn more.
Probing the limitations of multimodal language models for chemistry and materials research
Recent advancements in artificial intelligence have sparked interest in scientific assistants that could support researchers across the full spectrum of scientific workflows, from literature review to...
arxiv.org
Jablonka Lab (Lab for AI for Materials)@jablonkagroup.bsky.social · 2y agoAre Vision Language Models ready for scientific research? 🧑🔬🧪 We compared leading VLLMs on the three pillars of chemical and material science discovery: data extraction, lab experimentation and data interpretation. arxiv.org/abs/2411.16955
We are hiring (resharing appreciated)! Given recent successful grant applications (I got my SNSF Starting Grant 🚀), we are extending the LIAC team with multiple openings (PhD/postdoc) for 2025. Apply now (deadline: December 20th) by filling in this form: forms.fillout.com/t/eq5ADAw3kkus. #ChemSky
Our multimodal follow-up to ChemBench. Our team went to the lab to create images to challenge leading vision language models. Check the paper arxiv.org/abs/2411.16955 to learn more.
Probing the limitations of multimodal language models for chemistry and materials research
Recent advancements in artificial intelligence have sparked interest in scientific assistants that could support researchers across the full spectrum of scientific workflows, from literature review to...
arxiv.org
Jablonka Lab (Lab for AI for Materials)@jablonkagroup.bsky.social · 2y agoAre Vision Language Models ready for scientific research? 🧑🔬🧪 We compared leading VLLMs on the three pillars of chemical and material science discovery: data extraction, lab experimentation and data interpretation. arxiv.org/abs/2411.16955
For those of you looking for a big helping of LLMs for Thanksgiving, here you go! 🤖🦃 📝 arxiv.org/abs/2411.15221 We release 34 examples demonstrating applications of LLMs in materials science and chemistry across: 🔶molecular and material property prediction; 🔷molecular and material design; ...
I'm learning materials according to mine. Thanks, @morinryan.bsky.social for creating this service!
Alán Aspuru-Guzik@aspuru.bsky.social · 2y agoMy scholargoogler.com word cloud. Incredible service. #ChemSky #QuantumSky #SelfDrivingLabs #UniversityofToronto #UofT #VectorInsititute #AI #Materials #QuantumAlgorithms
Chemists often combine many different techniques to elucidate structures. Adrian has been building a system that mimics this using models and genetic algorithms.
Quick introduction. 👋 Here are some things I do: 🤖 Use AI/ML to supercharge science - from analysis to prediction, to automation 🔨 Build data infra so researchers can do more science with less overhead 🌟 Contribute to open code, data, and science 🤝 Build communities for modern research.
A very cool article about our ChemBench benchmark (arxiv.org/abs/2404.01475).
Are large language models superhuman chemists?
Large language models (LLMs) have gained widespread interest due to their ability to process human language and perform tasks on which they have not been explicitly trained. However, we possess only...
arxiv.org
Chemistry World@chemistryworld.com · 2y agoA new preprint shows that large language models consistently out-perform human chemists across all topics, but are limited in their ability to apply chemical reasoning or accurately evaluate their own performance. www.chemistryworld.com/news/large-l...
