Will Decker
@jwilldecker.bsky.social
PhD student @ Georgia Tech 🐝 and LIT Lab. Interested in how brains and machines learn, organize, and use knowledge about the world.
https://w-decker.github.io/
https://w-decker.github.io/
Reposted by Will Decker
Our paper “The cost of thinking is similar between large reasoning models and humans” is now out in PNAS! 🤖🧠
w/ @fepdelia.bsky.social, @hopekean.bsky.social, @lampinen.bsky.social, and @evfedorenko.bsky.social
Link: www.pnas.org/doi/10.1073/... (1/6)
w/ @fepdelia.bsky.social, @hopekean.bsky.social, @lampinen.bsky.social, and @evfedorenko.bsky.social
Link: www.pnas.org/doi/10.1073/... (1/6)
PNAS
Proceedings of the National Academy of Sciences (PNAS), a peer reviewed journal of the National Academy of Sciences (NAS) - an authoritative source of high-impact, original research that broadly spans...
www.pnas.org
November 19, 2025 at 8:14 PM
Our paper “The cost of thinking is similar between large reasoning models and humans” is now out in PNAS! 🤖🧠
w/ @fepdelia.bsky.social, @hopekean.bsky.social, @lampinen.bsky.social, and @evfedorenko.bsky.social
Link: www.pnas.org/doi/10.1073/... (1/6)
w/ @fepdelia.bsky.social, @hopekean.bsky.social, @lampinen.bsky.social, and @evfedorenko.bsky.social
Link: www.pnas.org/doi/10.1073/... (1/6)
Reposted by Will Decker
🤖📈 How do LLMs use their depth?
Akshat Gupta led a fun project to find out! We leverage TunedLens (~linear decoding of tokens) to explore how LLMs' internal representations change from layer to layer.
Preprint: arxiv.org/abs/2510.18871
1/
Akshat Gupta led a fun project to find out! We leverage TunedLens (~linear decoding of tokens) to explore how LLMs' internal representations change from layer to layer.
Preprint: arxiv.org/abs/2510.18871
1/
October 23, 2025 at 4:17 PM
🤖📈 How do LLMs use their depth?
Akshat Gupta led a fun project to find out! We leverage TunedLens (~linear decoding of tokens) to explore how LLMs' internal representations change from layer to layer.
Preprint: arxiv.org/abs/2510.18871
1/
Akshat Gupta led a fun project to find out! We leverage TunedLens (~linear decoding of tokens) to explore how LLMs' internal representations change from layer to layer.
Preprint: arxiv.org/abs/2510.18871
1/
Reposted by Will Decker
The new #NSF #GRFP application now excludes 1000’s of students, including those who were told they could apply this year, or who planned to apply next year.
We have a new specific petition urging #NSFGRFP undo this harmful action. If you are impacted sign here: laurenkuehne.github.io/grfpChanges/
We have a new specific petition urging #NSFGRFP undo this harmful action. If you are impacted sign here: laurenkuehne.github.io/grfpChanges/
September 29, 2025 at 5:47 PM
The new #NSF #GRFP application now excludes 1000’s of students, including those who were told they could apply this year, or who planned to apply next year.
We have a new specific petition urging #NSFGRFP undo this harmful action. If you are impacted sign here: laurenkuehne.github.io/grfpChanges/
We have a new specific petition urging #NSFGRFP undo this harmful action. If you are impacted sign here: laurenkuehne.github.io/grfpChanges/
Reposted by Will Decker
Our young lab greatly appreciates open datasets and code provided by pioneers in the field! @alexanderhuth.bsky.social
@amandalebel.bsky.social
@rjantonello.bsky.social
@mtoneva.bsky.social
@samnastase.bsky.social
@jixingli.bsky.social
@mschrimpf.bsky.social
@evfedorenko.bsky.social
etc
3/end
@amandalebel.bsky.social
@rjantonello.bsky.social
@mtoneva.bsky.social
@samnastase.bsky.social
@jixingli.bsky.social
@mschrimpf.bsky.social
@evfedorenko.bsky.social
etc
3/end
September 29, 2025 at 5:39 PM
Our young lab greatly appreciates open datasets and code provided by pioneers in the field! @alexanderhuth.bsky.social
@amandalebel.bsky.social
@rjantonello.bsky.social
@mtoneva.bsky.social
@samnastase.bsky.social
@jixingli.bsky.social
@mschrimpf.bsky.social
@evfedorenko.bsky.social
etc
3/end
@amandalebel.bsky.social
@rjantonello.bsky.social
@mtoneva.bsky.social
@samnastase.bsky.social
@jixingli.bsky.social
@mschrimpf.bsky.social
@evfedorenko.bsky.social
etc
3/end
Check out @neurotaha.bsky.social new work!
🚨 Paper alert:
To appear in the DBM Neurips Workshop
LITcoder: A General-Purpose Library for Building and Comparing Encoding Models
📄 arxiv: arxiv.org/abs/2509.091...
🔗 project: litcoder-brain.github.io
To appear in the DBM Neurips Workshop
LITcoder: A General-Purpose Library for Building and Comparing Encoding Models
📄 arxiv: arxiv.org/abs/2509.091...
🔗 project: litcoder-brain.github.io
September 29, 2025 at 2:47 PM
Check out @neurotaha.bsky.social new work!
Reposted by Will Decker
The GRFP announcement from NSF cuts out an entire cohort of 2nd year students from consideration, without warning. This is so deeply unfair that it warrants a formal protest from the scientific community. If someone wants to work with me to craft an open letter and solicit signatures, LMK.
New: After a long wait, the GRFP solicitation is live! Deadlines have been extended to early November, so applicants have a bit over a month to submit. www.nsf.gov/funding/oppo...
NSF Graduate Research Fellowship Program (GRFP)
www.nsf.gov
September 26, 2025 at 5:10 PM
The GRFP announcement from NSF cuts out an entire cohort of 2nd year students from consideration, without warning. This is so deeply unfair that it warrants a formal protest from the scientific community. If someone wants to work with me to craft an open letter and solicit signatures, LMK.
Reposted by Will Decker
Here is our best thinking about how to make world models. I would apologize for it being a massive 40-page behemoth, but it's worth reading. arxiv.org/pdf/2509.09737
arxiv.org
September 15, 2025 at 11:47 PM
Here is our best thinking about how to make world models. I would apologize for it being a massive 40-page behemoth, but it's worth reading. arxiv.org/pdf/2509.09737
Reposted by Will Decker
I am hiring a post doc at UAlberta, affiliated with Amii! We study language processing in the brain using LLMs and neuroimaging. Looking for someone with experience with ideally both neuroimaging and LLMs, or a willingness to learn. Email me with Qs
apps.ualberta.ca/careers/post...
apps.ualberta.ca/careers/post...
Postdoctoral Fellow - Language Models and Neuroscience - Careers@UAlberta.ca
University of Alberta: Careers@UAlberta.ca
apps.ualberta.ca
September 15, 2025 at 9:43 PM
I am hiring a post doc at UAlberta, affiliated with Amii! We study language processing in the brain using LLMs and neuroimaging. Looking for someone with experience with ideally both neuroimaging and LLMs, or a willingness to learn. Email me with Qs
apps.ualberta.ca/careers/post...
apps.ualberta.ca/careers/post...
Reposted by Will Decker
::slowly stands while clapping::
September 10, 2025 at 11:07 PM
::slowly stands while clapping::
Reposted by Will Decker
I love the smell of consequences in the morning. Thank you, @dragoncon.org
September 1, 2025 at 6:01 PM
I love the smell of consequences in the morning. Thank you, @dragoncon.org
Reposted by Will Decker
In our new paper, we explore how we can build encoding models that are both powerful and understandable. Our model uses an LLM to answer 35 questions about a sentence's content. The answers linearly contribute to our prediction of how the brain will respond to that sentence. 1/6
August 18, 2025 at 9:42 AM
In our new paper, we explore how we can build encoding models that are both powerful and understandable. Our model uses an LLM to answer 35 questions about a sentence's content. The answers linearly contribute to our prediction of how the brain will respond to that sentence. 1/6
Reposted by Will Decker
Many thanks to the volunteer organizers and the flash talk presenters for making the CCN watch party at Georgia Tech a success! And thanks all attendees for coming and engaging in discussions!
August 16, 2025 at 4:02 PM
Many thanks to the volunteer organizers and the flash talk presenters for making the CCN watch party at Georgia Tech a success! And thanks all attendees for coming and engaging in discussions!
Reposted by Will Decker
Looking forward to #CogSci2025 ! Find us throughout the conference
July 28, 2025 at 3:24 PM
Looking forward to #CogSci2025 ! Find us throughout the conference
Reposted by Will Decker
Atlanta community: we are organizing a CCN @cogcompneuro.bsky.social watch party!
Watch the talks, engage in structured discussions, and (optionally) present your own work.
Register:
forms.gle/AWxVPbrgxkdd...
Schedule:
tinyurl.com/ccn2025atlanta
Watch the talks, engage in structured discussions, and (optionally) present your own work.
Register:
forms.gle/AWxVPbrgxkdd...
Schedule:
tinyurl.com/ccn2025atlanta
July 22, 2025 at 3:40 PM
Atlanta community: we are organizing a CCN @cogcompneuro.bsky.social watch party!
Watch the talks, engage in structured discussions, and (optionally) present your own work.
Register:
forms.gle/AWxVPbrgxkdd...
Schedule:
tinyurl.com/ccn2025atlanta
Watch the talks, engage in structured discussions, and (optionally) present your own work.
Register:
forms.gle/AWxVPbrgxkdd...
Schedule:
tinyurl.com/ccn2025atlanta
Reposted by Will Decker
Georgia Tech’s Neuro Next Initiative has officially evolved into the Institute for Neuroscience, Neurotechnology, and Society (INNS). A new era of interdisciplinary research begins today.
🔗 neuro.gatech.edu
#Neuroscience #Neurotech #GeorgiaTech
🔗 neuro.gatech.edu
#Neuroscience #Neurotech #GeorgiaTech
We're proud to introduce our two newest Interdisciplinary Research Institutes: @gt-neuro.bsky.social and the Space Research Institute. bit.ly/3G5HiW1
July 1, 2025 at 9:09 PM
Georgia Tech’s Neuro Next Initiative has officially evolved into the Institute for Neuroscience, Neurotechnology, and Society (INNS). A new era of interdisciplinary research begins today.
🔗 neuro.gatech.edu
#Neuroscience #Neurotech #GeorgiaTech
🔗 neuro.gatech.edu
#Neuroscience #Neurotech #GeorgiaTech
Reposted by Will Decker
So far we have CCN local meetups @ NYC @neurograce.bsky.social, Boston @lynnkasorensen.bsky.social, Atlanta @neuranna.bsky.social, Lyon and Bucharest!
Who will host one on even further away from Amsterdam (West Coast? East Asia? Africa?)
Who will host one on even further away from Amsterdam (West Coast? East Asia? Africa?)
June 20, 2025 at 10:34 AM
So far we have CCN local meetups @ NYC @neurograce.bsky.social, Boston @lynnkasorensen.bsky.social, Atlanta @neuranna.bsky.social, Lyon and Bucharest!
Who will host one on even further away from Amsterdam (West Coast? East Asia? Africa?)
Who will host one on even further away from Amsterdam (West Coast? East Asia? Africa?)
Reposted by Will Decker
What are the organizing dimensions of language processing?
We show that voxel responses during comprehension are organized along 2 main axes: processing difficulty & meaning abstractness—revealing an interpretable, topographic representational basis for language processing shared across individuals
We show that voxel responses during comprehension are organized along 2 main axes: processing difficulty & meaning abstractness—revealing an interpretable, topographic representational basis for language processing shared across individuals
May 23, 2025 at 5:00 PM
What are the organizing dimensions of language processing?
We show that voxel responses during comprehension are organized along 2 main axes: processing difficulty & meaning abstractness—revealing an interpretable, topographic representational basis for language processing shared across individuals
We show that voxel responses during comprehension are organized along 2 main axes: processing difficulty & meaning abstractness—revealing an interpretable, topographic representational basis for language processing shared across individuals
Reposted by Will Decker
A postdoc in my lab contacted a fellow scientist to get some advice and was told that, because he's currently working on the topic, he "can't give away all his secrets." I find this extremely lame, and yet I keep encountering it. Science is not a zero-sum game!!
April 16, 2025 at 9:14 PM
A postdoc in my lab contacted a fellow scientist to get some advice and was told that, because he's currently working on the topic, he "can't give away all his secrets." I find this extremely lame, and yet I keep encountering it. Science is not a zero-sum game!!
Reposted by Will Decker
New preprint ✨ w/ @olafhauk.bsky.social and Matt Lambon Ralph @mrccbu.bsky.social @cambridgeuni.bsky.social @gatescambridge.bsky.social
The anterior temporal lobe (ATL) seems crucial for both semantic memory and semantic composition. How to make sense of this neuroanatomical alignment? 🧠
🧵
The anterior temporal lobe (ATL) seems crucial for both semantic memory and semantic composition. How to make sense of this neuroanatomical alignment? 🧠
🧵
A common framework for semantic memory and semantic composition https://www.biorxiv.org/content/10.1101/2025.04.11.648390v1
April 15, 2025 at 2:31 PM
New preprint ✨ w/ @olafhauk.bsky.social and Matt Lambon Ralph @mrccbu.bsky.social @cambridgeuni.bsky.social @gatescambridge.bsky.social
The anterior temporal lobe (ATL) seems crucial for both semantic memory and semantic composition. How to make sense of this neuroanatomical alignment? 🧠
🧵
The anterior temporal lobe (ATL) seems crucial for both semantic memory and semantic composition. How to make sense of this neuroanatomical alignment? 🧠
🧵
Reposted by Will Decker
Very cool work by Qiawen (Ella) Liu, Paridon, @glupyan.bsky.social: e.g., Is white hot or cold?
LMs consider 'white' cold because it's mediated by 'snow'
Ablate the mediators, and you eliminate the associations.
www.nature.com/articles/s44...
LMs consider 'white' cold because it's mediated by 'snow'
Ablate the mediators, and you eliminate the associations.
www.nature.com/articles/s44...
Learning about color from language - Communications Psychology
Blind people show similar associations between adjectives (e.g. cold) and colours (e.g. blue) as sighted people; word embedding models trained on corpora of written and spoken language learn these ass...
www.nature.com
April 16, 2025 at 2:58 PM
Very cool work by Qiawen (Ella) Liu, Paridon, @glupyan.bsky.social: e.g., Is white hot or cold?
LMs consider 'white' cold because it's mediated by 'snow'
Ablate the mediators, and you eliminate the associations.
www.nature.com/articles/s44...
LMs consider 'white' cold because it's mediated by 'snow'
Ablate the mediators, and you eliminate the associations.
www.nature.com/articles/s44...
Reposted by Will Decker
New brain/language study w/ @evfedorenko.bsky.social! We applied task-agnostic individualized functional connectomics (iFC) to the entire history of fMRI scanning in the Fedorenko lab, parcellating nearly 1200 brains into networks based on activity fluctuations alone. doi.org/10.1101/2025... . 🧵
A language network in the individualized functional connectomes of over 1,000 human brains doing arbitrary tasks
A century and a half of neuroscience has yielded many divergent theories of the neurobiology of language. Two factors that likely contribute to this situation include (a) conceptual disagreement…
doi.org
March 31, 2025 at 3:19 PM
New brain/language study w/ @evfedorenko.bsky.social! We applied task-agnostic individualized functional connectomics (iFC) to the entire history of fMRI scanning in the Fedorenko lab, parcellating nearly 1200 brains into networks based on activity fluctuations alone. doi.org/10.1101/2025... . 🧵
Reposted by Will Decker
LIT lab is presenting 2 posters at #CNS2025!
Session D - Scaling Laws in Functional Region of Interest (fROI) Analyses by @ruimingao.bsky.social
Session E - Characterizing the effects of content, task and modality on task-driven semantic processing in the brain (sketchpad series) by Jin Li
Session D - Scaling Laws in Functional Region of Interest (fROI) Analyses by @ruimingao.bsky.social
Session E - Characterizing the effects of content, task and modality on task-driven semantic processing in the brain (sketchpad series) by Jin Li
March 31, 2025 at 3:18 AM
LIT lab is presenting 2 posters at #CNS2025!
Session D - Scaling Laws in Functional Region of Interest (fROI) Analyses by @ruimingao.bsky.social
Session E - Characterizing the effects of content, task and modality on task-driven semantic processing in the brain (sketchpad series) by Jin Li
Session D - Scaling Laws in Functional Region of Interest (fROI) Analyses by @ruimingao.bsky.social
Session E - Characterizing the effects of content, task and modality on task-driven semantic processing in the brain (sketchpad series) by Jin Li
Reposted by Will Decker
Our paper on key-value memory in the brain (updated from the preprint version) is now out in Neuron @cp-neuron.bsky.social
authors.elsevier.com/a/1kqIJ3BtfH...
authors.elsevier.com/a/1kqIJ3BtfH...
Key-value memory in the brain
Classical models of memory in psychology and neuroscience rely on similarity-based retrieval of stored patterns, where similarity is a function of ret…
www.sciencedirect.com
March 26, 2025 at 3:09 PM
Our paper on key-value memory in the brain (updated from the preprint version) is now out in Neuron @cp-neuron.bsky.social
authors.elsevier.com/a/1kqIJ3BtfH...
authors.elsevier.com/a/1kqIJ3BtfH...
Reposted by Will Decker
Excited to introduce funROI: A Python package for functional ROI analyses of fMRI data!
funroi.readthedocs.io/en/latest/
#fMRI #Neuroimaging #Python #OpenScience
Work w @neuranna.bsky.social
🧵👇
funroi.readthedocs.io/en/latest/
#fMRI #Neuroimaging #Python #OpenScience
Work w @neuranna.bsky.social
🧵👇
March 18, 2025 at 2:48 PM
Excited to introduce funROI: A Python package for functional ROI analyses of fMRI data!
funroi.readthedocs.io/en/latest/
#fMRI #Neuroimaging #Python #OpenScience
Work w @neuranna.bsky.social
🧵👇
funroi.readthedocs.io/en/latest/
#fMRI #Neuroimaging #Python #OpenScience
Work w @neuranna.bsky.social
🧵👇