Linjiang | 林江

LLMs at your service: Knowledge discovery from porous organic cages literature using a large language model doi.org/10.26434/chemr…

Tap into the unlimited potential of machine learning potentials - in this case, we have hydrogen bonding, halogen bonding, and phase transitions 🤯 doi.org/10.26434/chemr…

Physics-informed dual-objective Bayesian optimization of high-entropy-alloy nanozymes, with LLM-in-the-loop experimental data analysis, by a robotic AI chemist doi.org/10.26434/chemr…

@JunJian07285897 🫡

@linjiangchen Good work

Chemistry laboratories most commonly operate in a multi-researcher, multi-task mode. So can autonomous chemistry laboratories do. Introducing our multi-robot-multi-task scheduling system: doi.org/10.26434/chemr…

Llama 3 + agents + robots + automated exptl. stations + your demand >>> doi.org/10.26434/chemr…

Two #PhD positions open in my group once we move to @UoBChemistry in April 2024! Interested in #compchem #materials #MachineLearning? Apply by March 1 at birmingham.ac.uk/schools/chemis… #PhDjobs #chemPhD Please spread the word!

@aicooper Or build ☃️☃️☃️ in front of the MIF 😁

An honour to be named amongst the 2024 Thieme Chemistry Journal awardees! @UoBChemistry Many thanks to @thiemechemistry and congratulations to the 2024 awardees! thieme.de/en/thieme-chem…

I have two PhD opportunities in my group. If you're passionate about new materials discovery and solid-state #NMR, this could be for you 🙂 @FindAPhD findaphd.com/phds/projectde… findaphd.com/phds/projectde…

A tour de force from Muyu, a Master's student who visited the group last year, on a 6-months CSC scholarship 🎉 And, there's more to come 🤓 doi.org/10.1021/acsami…

Happy to share our latest article “Design of piezoelectric organic cage salts energy harvesting” is out now. Thanks for the contributions from all co-authors.

Congratulations to Prof Andrew Cooper on being elected as a Foreign Member of the Chinese Academy of Sciences - one of the highest honours that China bestows 🙌 Prof Cooper’s research is pushing boundaries in AI, robotics & chemistry. Read more ➡️ brnw.ch/21wFXxm

Took a long time in the making but here it is: Progress toward the computational discovery of new metal–organic framework adsorbents for energy applications nature.com/articles/s4156… Many thanks to @RandallSnurr and @DrYGChung . @ucl @uclchemeng

@MolecularXtal Nice work 👍 A genuine question: why is it a foundation model and not "just" a general purpose machine learned force field?

@LucaCataz Amazing news - congratulations Luca 👏🎉 hopefully will be able to catch up with you in the near future 😄

After 6 countries, 3 continents, countless of amazing collaborators, super mentors, and lots of cool science, I am now tenured Associate Professor at the Department of Life Sciences at @UNIMORE_univ back in Italy! Thanks to all the great people who supported me in this journey!

@Andrew_Sue @TeamISMSC @ismsc2024 Welcome to Hangzhou for ISMSC 2024. We are delighted to extend our warm invitation to visit Westlake University during your stay. If you are interested, please feel free to reach out to me via email liu@westlake.edu.cn. It is my utmost pleasure to welcome you during your visit.

The wait is over! The official website of ISMSC-2024 is now online: ismsc2024.com. Registration is also open. We look forward to seeing you all in Hangzhou, China next May! @TeamISMSC @ISMSC2024

@taylordsparks @sayeedh34 @RosseinskyGroup Molecular crystals would be a fine challenge 😉

@linjiangchen @sayeedh34 @RosseinskyGroup Thanks Linjiang! We certainly agree and think this is a pretty big deal. FUSE was already quite good- comparable with any structure prediction algorithm and we have now found a way to use ML to pretty dramatically speed it up. We have some other future ideas cooking up next.

Our latest is out! Over the last year, my student @sayeedh34 and I had the pleasure of working with @RosseinskyGroup members to bring machine learning approaches into their already capable FUSE code for crystal structure prediction. Unsurprisingly, using generative machine learning models to first identify likely candidate structures and then breaking these up into modules for FUSE was extremely helpful and efficient in identifying the lowest energy structure. We see speed-ups between 2-8x 😎 Preprint out now! doi.org/10.26434/chemr…

Meet Eunomia, our AI agent for materials datasets! 🤖 It's easy to use, has RAG and chain of verification 🤓, shows high performance across different NLP tasks 🚀, and more... 🎉 Live App: eunomia.streamlit.app Paper: arxiv.org/abs/2312.11690 Code: github.com/AI4ChemS/Eunom…