@d_armstr profile picture
Daniel Armstrong @d_armstr
Joined December 2022
  • Tweets 15
  • Following 136
  • Followers 60
5 Photos and videos
Daniel Armstrong retweeted
Philippe Schwaller (he/him)@pschwllr
14 Oct 2025
Exciting postdoc opportunity in the @SchwallerGroup at EPFL! We're hiring a postdoc to advance ML-driven synthesis planning after Zlatko Joncev’s successful exit to co-found B-12 (YC '25) 🚀 Work on: - LLMs for strategic synthesis planning - Chemical reasoning at scale - Building the next-gen framework for retrosynthesis Our recent preprint shows that LLMs can guide synthesis planning with natural-language strategies — combining AI reasoning with traditional chemical tools (arxiv.org/abs/2503.08537). Join us at the intersection of chemistry & AI. Up to 2 years. Based in Lausanne 🇨🇭 Apply: forms.fillout.com/t/nnxVE3Rc… #ChemTwitter #MachineLearning #SynthesisPlanning #PostdocPosition
15
38
7,677
Daniel Armstrong retweeted
Nature Computational Science@NatComputSci
23 Jul 2025
Out now! @pschwllr, @SchwallerGroup, @loic_roch, @VictorSabanza and colleagues provide guidelines and recommendations for when to use multi-fidelity Bayesian optimization over their single-fidelity counterparts nature.com/articles/s43588-0… 🔓rdcu.be/exyUh

Best practices for multi-fidelity Bayesian optimization in materials and molecular research

Nature Computational Science - Multi-fidelity Bayesian optimization methods are studied on molecular and material discovery tasks, and guidelines are provided to recommend cheaper and informative...

nature.com
1
8
36
2,073
Happy to present a poster of my work on starting material constrained synthesis planning, Tango*, in the AI4Materials workshop at #ICLR2025 in 🇸🇬 Thanks to collaborators @JeffGuo__ and @ZJoncev and my PI @pschwllr!
Daniel Armstrong@d_armstr
9 Dec 2024
1/ Starting material constrained synthesis planning is now possible using a general retrosynthesis algorithm *without* training a dedicated value network! Check-out our new paper, TangoStar. Preprint : arxiv.org/abs/2412.03424
4
15
1,435
Daniel Armstrong retweeted
Junwu Chen@JunwuChen25
8 Apr 2025
AdsMT is finally out in @NatureComms 🥳 It is designed for rapid prediction of global minimum adsorption energy (GMAE) from surface graphs and adsorbate descriptors. Thanks to @XuHuang461675 , @pschwllr and @NCCR_Catalysis! Paper & code: nature.com/articles/s41467-0… @SchwallerGroup

A multi-modal transformer for predicting global minimum adsorption energy

Nature Communications - The fast evaluation of global minimum adsorption energy (GMAE) is crucial for catalyst screening. Here, authors designed a multi-modal transformer called AdsMT to rapidly...

nature.com
1
8
28
2,918
New LIAC ( @SchwallerGroup ) pre-print out! Here we show that language models understand synthetic strategy, reaction mechanism and chemical structure. Many thanks to my co-authors @drecmb, @TheoNeukomm and @ZJoncev along-with our PI @pschwllr. Paper: arxiv.org/abs/2503.08537

Andrés M Bran@drecmb
12 Mar 2025
LLMs are pretty bad at writing molecules, but quite good at analyzing mols and reactions! In our new work we use LLMs search in chemical tasks, unlocking steerable synth. planning and mechanism prediction 🌟 1/ @TheoNeukomm @d_armstr @ZJoncev @pschwllr arxiv.org/abs/2503.08537
1
4
11
1,127
more replies
Reaction feasibility is a huge issue in retrosynthesis tools; models often propose reactions which are incompatible with different regions in the molecule. Here the LLM correctly tags the highlighted alcohol group as incompatible with the bromide and suggests MOM/TBS protection.
1
1
92
Thanks to the co-authors and the Swiss National Science Foundation for supporting this work.
2
76
1/ Starting material constrained synthesis planning is now possible using a general retrosynthesis algorithm *without* training a dedicated value network! Check-out our new paper, TangoStar. Preprint : arxiv.org/abs/2412.03424
1
18
38
4,199
more replies
4/ We thank developed this work using code and models provided by the Coley Group (@cwcoley) in their recent work, DESP. Code : github.com/coleygroup/desp Paper : arxiv.org/abs/2407.06334

GitHub - coleygroup/desp: Double-Ended Synthesis Planning with Goal-Constrained Bidirectional...

Double-Ended Synthesis Planning with Goal-Constrained Bidirectional Search (NeurIPS 2024) - coleygroup/desp

github.com
1
2
184
5/ Thanks to my PI @pschwllr (@SchwallerGroup), my collaborators @JeffGuo__ and @ZJoncev, and the Swiss National Science Foundation (@snsf_ch) for funding this work.
1
224
Daniel Armstrong retweeted
LIAC at EPFL@SchwallerGroup
7 Dec 2024
Another great #Ellis workshop on ML for Molecular Discovery. Thanks to @gklambauer, @fra_grisoni, and the whole organizing team! Proud of our team's contributions - 3 posters (@d_armstr, @kelmannson, @ZJoncev) and an excellent talk by @SarinaKopf! Very well done! #compchem
9
24
1,913
Daniel Armstrong retweeted
Andrés M Bran@drecmb
8 May 2024
ChemCrow is finally our in @NatMachIntell! 🥳🔥 nature.com/articles/s42256-0… It took us more than a year since the preprint, but we made it With @SamCox822 @andrewwhite01 @CarloBalda97 @OSchilter @CarloBalda97 @andrewwhite01 @pschwllr. @EPFL_en @NCCR_Catalysis @FutureHouseSF

Augmenting large language models with chemistry tools

Nature Machine Intelligence - Large language models can be queried to perform chain-of-thought reasoning on text descriptions of data or computational tools, which can enable flexible and...

nature.com
2
27
105
15,106
Load more