Philipp Marquetand
@marquetand.bsky.social
Machine learning for different domains ranging from graphic design to chemistry.
Reposted by Philipp Marquetand
📢 Paper + code release 📃💻
After 2 years of work, I'm excited to announce our newest paper, MatterGen, has been published in Nature!
www.nature.com/articles/s41...
We are also releasing all the training data, model weights, model code, and evaluation code on GitHub!
github.com/microsoft/ma...
After 2 years of work, I'm excited to announce our newest paper, MatterGen, has been published in Nature!
www.nature.com/articles/s41...
We are also releasing all the training data, model weights, model code, and evaluation code on GitHub!
github.com/microsoft/ma...
January 16, 2025 at 10:15 AM
📢 Paper + code release 📃💻
After 2 years of work, I'm excited to announce our newest paper, MatterGen, has been published in Nature!
www.nature.com/articles/s41...
We are also releasing all the training data, model weights, model code, and evaluation code on GitHub!
github.com/microsoft/ma...
After 2 years of work, I'm excited to announce our newest paper, MatterGen, has been published in Nature!
www.nature.com/articles/s41...
We are also releasing all the training data, model weights, model code, and evaluation code on GitHub!
github.com/microsoft/ma...
Reposted by Philipp Marquetand
⚠️ Readable Paper Alert ⚠️
BLT: what if we just got rid of tokenization?
Result:
* text looks a lot like audio, video, PDF, it’s all just bytes
* dynamically reduce compute based on difficulty
* new scaling axis (patch size)
ai.meta.com/research/pub...
BLT: what if we just got rid of tokenization?
Result:
* text looks a lot like audio, video, PDF, it’s all just bytes
* dynamically reduce compute based on difficulty
* new scaling axis (patch size)
ai.meta.com/research/pub...
Byte Latent Transformer: Patches Scale Better Than Tokens | Research - AI at Meta
We introduce the Byte Latent Transformer (BLT), a new byte-level LLM architecture that, for the first time, matches tokenization-based LLM performance at...
ai.meta.com
December 14, 2024 at 4:12 PM
⚠️ Readable Paper Alert ⚠️
BLT: what if we just got rid of tokenization?
Result:
* text looks a lot like audio, video, PDF, it’s all just bytes
* dynamically reduce compute based on difficulty
* new scaling axis (patch size)
ai.meta.com/research/pub...
BLT: what if we just got rid of tokenization?
Result:
* text looks a lot like audio, video, PDF, it’s all just bytes
* dynamically reduce compute based on difficulty
* new scaling axis (patch size)
ai.meta.com/research/pub...
Reposted by Philipp Marquetand
There are already many excellent expert reviews on MLPs, so why adding more to the mix?
We want to give a light intro into the field for new starters going from the historical developments to latest developments on going beyond locality to foundation models.
dx.doi.org/10.1088/1361-648X/ad9657
We want to give a light intro into the field for new starters going from the historical developments to latest developments on going beyond locality to foundation models.
dx.doi.org/10.1088/1361-648X/ad9657
December 6, 2024 at 4:33 PM
There are already many excellent expert reviews on MLPs, so why adding more to the mix?
We want to give a light intro into the field for new starters going from the historical developments to latest developments on going beyond locality to foundation models.
dx.doi.org/10.1088/1361-648X/ad9657
We want to give a light intro into the field for new starters going from the historical developments to latest developments on going beyond locality to foundation models.
dx.doi.org/10.1088/1361-648X/ad9657
Reposted by Philipp Marquetand
Scientists, academics, researchers: We’re excited to share that @altmetric.com is now tracking mentions of your research on Bluesky! 🧪
There are already many articles for which there is more attention on Bluesky than on other comparable micro-blogging sites, meaning the academic community and the general public have clearly adopted Bluesky as one of its core places to disseminate and discuss new research.
A Place of Joy.
A Place of Joy.
December 3, 2024 at 2:10 PM
Scientists, academics, researchers: We’re excited to share that @altmetric.com is now tracking mentions of your research on Bluesky! 🧪
Reposted by Philipp Marquetand
Hello! I'm posting this both here and on the X-rated site, let's see where it gets more re-posts 😇. We are looking for a research software engineer to help us develop (even) better code for #compchem #atomicscale #machinelearning. Check out the specs and apply!
www.epfl.ch/labs/cosmo/i...
www.epfl.ch/labs/cosmo/i...
Jobs
-
www.epfl.ch
December 3, 2024 at 11:56 AM
Hello! I'm posting this both here and on the X-rated site, let's see where it gets more re-posts 😇. We are looking for a research software engineer to help us develop (even) better code for #compchem #atomicscale #machinelearning. Check out the specs and apply!
www.epfl.ch/labs/cosmo/i...
www.epfl.ch/labs/cosmo/i...
Reposted by Philipp Marquetand
With less than 1M compounds, our VQM24 dataset sets a new SOTA covering more comprehensively the compositional and configurational space of small molecules. Wonderful joint efforts by current and past team-members and collaborators. Hopefully useful for
#ML
&
#compchem arxiv.org/abs/2405.05961
#ML
&
#compchem arxiv.org/abs/2405.05961
November 19, 2024 at 12:56 PM
With less than 1M compounds, our VQM24 dataset sets a new SOTA covering more comprehensively the compositional and configurational space of small molecules. Wonderful joint efforts by current and past team-members and collaborators. Hopefully useful for
#ML
&
#compchem arxiv.org/abs/2405.05961
#ML
&
#compchem arxiv.org/abs/2405.05961
Reposted by Philipp Marquetand
Timon S. Gutleb, Rhyan Barrett, Julia Westermayr, Christoph Ortner
Parameterizing Intersecting Surfaces via Invariants
https://arxiv.org/abs/2407.03731
Parameterizing Intersecting Surfaces via Invariants
https://arxiv.org/abs/2407.03731
July 8, 2024 at 7:01 AM
Timon S. Gutleb, Rhyan Barrett, Julia Westermayr, Christoph Ortner
Parameterizing Intersecting Surfaces via Invariants
https://arxiv.org/abs/2407.03731
Parameterizing Intersecting Surfaces via Invariants
https://arxiv.org/abs/2407.03731
Some steps from my life:
1. Quantum dynamics
2. ML in chemistry
3. ML for graphic design
Reflected in some publications and an advert from this year:
1. doi.org/10.1103/Phys...
2. doi.org/10.1039/D4SC...
3. www.lifeatcanva.com/en/jobs/6000...
More info below
1. Quantum dynamics
2. ML in chemistry
3. ML for graphic design
Reflected in some publications and an advert from this year:
1. doi.org/10.1103/Phys...
2. doi.org/10.1039/D4SC...
3. www.lifeatcanva.com/en/jobs/6000...
More info below
November 29, 2024 at 11:41 PM
Some steps from my life:
1. Quantum dynamics
2. ML in chemistry
3. ML for graphic design
Reflected in some publications and an advert from this year:
1. doi.org/10.1103/Phys...
2. doi.org/10.1039/D4SC...
3. www.lifeatcanva.com/en/jobs/6000...
More info below
1. Quantum dynamics
2. ML in chemistry
3. ML for graphic design
Reflected in some publications and an advert from this year:
1. doi.org/10.1103/Phys...
2. doi.org/10.1039/D4SC...
3. www.lifeatcanva.com/en/jobs/6000...
More info below