Reposted by arXiv loVe
Excited to share TWO new preprints:
1) Breaking Mechanical Holography in Combinatorial Metamaterials - building blocks that don't induce holographic order allow to build lots of mechanially compatible metamaterials and to design the texture (almost) on their entire boundary
arxiv.org/abs/2411.15760
1) Breaking Mechanical Holography in Combinatorial Metamaterials - building blocks that don't induce holographic order allow to build lots of mechanially compatible metamaterials and to design the texture (almost) on their entire boundary
arxiv.org/abs/2411.15760
December 3, 2024 at 7:32 PM
Excited to share TWO new preprints:
1) Breaking Mechanical Holography in Combinatorial Metamaterials - building blocks that don't induce holographic order allow to build lots of mechanially compatible metamaterials and to design the texture (almost) on their entire boundary
arxiv.org/abs/2411.15760
1) Breaking Mechanical Holography in Combinatorial Metamaterials - building blocks that don't induce holographic order allow to build lots of mechanially compatible metamaterials and to design the texture (almost) on their entire boundary
arxiv.org/abs/2411.15760
Reposted by arXiv loVe
(1/5) Fun fact: Several classic results in the stat. mech. of learning can be derived in a couple lines of simple algebra!
In this paper with Haim Sompolinsky, we simplify and unify derivations for high-dimensional convex learning problems using a bipartite cavity method.
arxiv.org/abs/2412.01110
In this paper with Haim Sompolinsky, we simplify and unify derivations for high-dimensional convex learning problems using a bipartite cavity method.
arxiv.org/abs/2412.01110
Simplified derivations for high-dimensional convex learning problems
Statistical physics provides tools for analyzing high-dimensional problems in machine learning and theoretical neuroscience. These calculations, particularly those using the replica method, often invo...
arxiv.org
December 3, 2024 at 7:34 PM
(1/5) Fun fact: Several classic results in the stat. mech. of learning can be derived in a couple lines of simple algebra!
In this paper with Haim Sompolinsky, we simplify and unify derivations for high-dimensional convex learning problems using a bipartite cavity method.
arxiv.org/abs/2412.01110
In this paper with Haim Sompolinsky, we simplify and unify derivations for high-dimensional convex learning problems using a bipartite cavity method.
arxiv.org/abs/2412.01110
Reposted by arXiv loVe
Very cool! I'm happy our paths crossed.
Link: doi.org/10.1103/Phys...
Alt: arxiv.org/abs/0712.0835
I didn't focus much about graphene in the above post, but I will always love it especially for the charming "scotch tape" method.
Link: doi.org/10.1103/Phys...
Alt: arxiv.org/abs/0712.0835
I didn't focus much about graphene in the above post, but I will always love it especially for the charming "scotch tape" method.
Universal Optical Conductance of Graphite
We find experimentally that the optical sheet conductance of graphite per graphene layer is very close to $(\ensuremath{\pi}/2){e}^{2}/h$, which is the theoretically expected value of dynamical conduc...
doi.org
December 3, 2024 at 7:44 PM
Very cool! I'm happy our paths crossed.
Link: doi.org/10.1103/Phys...
Alt: arxiv.org/abs/0712.0835
I didn't focus much about graphene in the above post, but I will always love it especially for the charming "scotch tape" method.
Link: doi.org/10.1103/Phys...
Alt: arxiv.org/abs/0712.0835
I didn't focus much about graphene in the above post, but I will always love it especially for the charming "scotch tape" method.
Reposted by arXiv loVe
Thrilled to share that our paper “Flo: A Semantic Foundation for Progressive Stream Processing” (with @mpmilano.bsky.social, Alvin Cheung, and @joehellerstein.bsky.social) will appear at POPL 2025! Check out the preprint at arxiv.org/abs/2411.08274, and read on for more!
Flo: a Semantic Foundation for Progressive Stream Processing
Streaming systems are present throughout modern applications, processing continuous data in real-time. Existing streaming languages have a variety of semantic models and guarantees that are often inco...
arxiv.org
December 3, 2024 at 8:26 PM
Thrilled to share that our paper “Flo: A Semantic Foundation for Progressive Stream Processing” (with @mpmilano.bsky.social, Alvin Cheung, and @joehellerstein.bsky.social) will appear at POPL 2025! Check out the preprint at arxiv.org/abs/2411.08274, and read on for more!
Reposted by arXiv loVe
Strongly agree that "clearer, more specific language to describe the phenomena labelled as tipping points" is needed, due to all the existing definitions
Here is an illustration (for a bistable model) of "my definition" so far:
References: arxiv.org/abs/1103.0169
www.nature.com/articles/350...
Here is an illustration (for a bistable model) of "my definition" so far:
References: arxiv.org/abs/1103.0169
www.nature.com/articles/350...
December 3, 2024 at 8:31 PM
Strongly agree that "clearer, more specific language to describe the phenomena labelled as tipping points" is needed, due to all the existing definitions
Here is an illustration (for a bistable model) of "my definition" so far:
References: arxiv.org/abs/1103.0169
www.nature.com/articles/350...
Here is an illustration (for a bistable model) of "my definition" so far:
References: arxiv.org/abs/1103.0169
www.nature.com/articles/350...
Reposted by arXiv loVe
The "LLM World of Words" (n=12,000 cue words) finds humans generate more unique word associations than Llama3, Mistral, or Haiku. Semantic networks of LLMs overlap to varying degrees with humans and show similar semantic priming effects.
arxiv.org/abs/2412.01330
arxiv.org/abs/2412.01330
The "LLM World of Words" English free association norms generated by large language models
Free associations have been extensively used in cognitive psychology and linguistics for studying how conceptual knowledge is organized. Recently, the potential of applying a similar approach for inve...
arxiv.org
December 3, 2024 at 9:47 PM
The "LLM World of Words" (n=12,000 cue words) finds humans generate more unique word associations than Llama3, Mistral, or Haiku. Semantic networks of LLMs overlap to varying degrees with humans and show similar semantic priming effects.
arxiv.org/abs/2412.01330
arxiv.org/abs/2412.01330
Reposted by arXiv loVe
Who: Nikos Athanasiou, Alpár Cseke, Markos Diomataris, me, @gulvarol.bsky.social
MotionFix: Text-Driven 3D Human Motion Editing
arXiv: arxiv.org/abs/2408.00712
Demo: huggingface.co/spaces/atnik...
Project: motionfix.is.tue.mpg.de
Dataset: motionfix.is.tue.mpg.de/explore.php pic.x.com/70CoLIlSxM
MotionFix: Text-Driven 3D Human Motion Editing
arXiv: arxiv.org/abs/2408.00712
Demo: huggingface.co/spaces/atnik...
Project: motionfix.is.tue.mpg.de
Dataset: motionfix.is.tue.mpg.de/explore.php pic.x.com/70CoLIlSxM
December 3, 2024 at 10:49 PM
Who: Nikos Athanasiou, Alpár Cseke, Markos Diomataris, me, @gulvarol.bsky.social
MotionFix: Text-Driven 3D Human Motion Editing
arXiv: arxiv.org/abs/2408.00712
Demo: huggingface.co/spaces/atnik...
Project: motionfix.is.tue.mpg.de
Dataset: motionfix.is.tue.mpg.de/explore.php pic.x.com/70CoLIlSxM
MotionFix: Text-Driven 3D Human Motion Editing
arXiv: arxiv.org/abs/2408.00712
Demo: huggingface.co/spaces/atnik...
Project: motionfix.is.tue.mpg.de
Dataset: motionfix.is.tue.mpg.de/explore.php pic.x.com/70CoLIlSxM
Reposted by arXiv loVe
Finally catching up on this Test Time Training paper. Thoughts
- very nice results
- interesting that TTT helps smaller models more (sign that scale is disproportionately useful for more memorization?)
- What a pain, I really hope we don't have to do this 😅
arxiv.org/abs/2411.07279
- very nice results
- interesting that TTT helps smaller models more (sign that scale is disproportionately useful for more memorization?)
- What a pain, I really hope we don't have to do this 😅
arxiv.org/abs/2411.07279
The Surprising Effectiveness of Test-Time Training for Abstract Reasoning
Language models have shown impressive performance on tasks within their training distribution, but often struggle with novel problems requiring complex reasoning. We investigate the effectiveness of t...
arxiv.org
December 3, 2024 at 10:50 PM
Finally catching up on this Test Time Training paper. Thoughts
- very nice results
- interesting that TTT helps smaller models more (sign that scale is disproportionately useful for more memorization?)
- What a pain, I really hope we don't have to do this 😅
arxiv.org/abs/2411.07279
- very nice results
- interesting that TTT helps smaller models more (sign that scale is disproportionately useful for more memorization?)
- What a pain, I really hope we don't have to do this 😅
arxiv.org/abs/2411.07279
Reposted by arXiv loVe
Chemical Complexity and Prevalence of Life in the Universe: A New Method for the Estimation of Key Terms of Drake Equation
arxiv.org/abs/2412.01001
arxiv.org/abs/2412.01001
Chemical Complexity and Prevalence of Life in the Universe: A New Method for the Estimation of Key Terms of Drake Equation
I describe a new method of estimating the prevalence of life in the Universe, based on the fact that more chemically complex environments are more rare. The paper makes three main claims: (1) There is...
arxiv.org
December 3, 2024 at 11:06 PM
Chemical Complexity and Prevalence of Life in the Universe: A New Method for the Estimation of Key Terms of Drake Equation
arxiv.org/abs/2412.01001
arxiv.org/abs/2412.01001
Reposted by arXiv loVe
Our first 2025 paper is out:
Official copy: ieeexplore.ieee.org/abstract/doc...
ArXiv version: arxiv.org/pdf/2407.05478
Official copy: ieeexplore.ieee.org/abstract/doc...
ArXiv version: arxiv.org/pdf/2407.05478
December 3, 2024 at 11:11 PM
Our first 2025 paper is out:
Official copy: ieeexplore.ieee.org/abstract/doc...
ArXiv version: arxiv.org/pdf/2407.05478
Official copy: ieeexplore.ieee.org/abstract/doc...
ArXiv version: arxiv.org/pdf/2407.05478
Reposted by arXiv loVe
This work from Mila/FAIR peeps is actually very cool. arxiv.org/abs/2405.00740
Modeling Caption Diversity in Contrastive Vision-Language Pretraining
There are a thousand ways to caption an image. Contrastive Language Pretraining (CLIP) on the other hand, works by mapping an image and its caption to a single vector -- limiting how well CLIP-like mo...
arxiv.org
December 4, 2024 at 12:04 AM
This work from Mila/FAIR peeps is actually very cool. arxiv.org/abs/2405.00740
Reposted by arXiv loVe
#EMNLP2024 Accepted to EMNLP 2024 Main Conference 🎉
”A Bayesian Approach to Harnessing the Power of LLMs in Authorship Attribution“
It works so well that a 7B open-source LLM outperforms GPT-4 with prompting at identifying unique writing style. 🔍
👉 Learn more: arxiv.org/pdf/2410.217...
”A Bayesian Approach to Harnessing the Power of LLMs in Authorship Attribution“
It works so well that a 7B open-source LLM outperforms GPT-4 with prompting at identifying unique writing style. 🔍
👉 Learn more: arxiv.org/pdf/2410.217...
December 4, 2024 at 12:19 AM
#EMNLP2024 Accepted to EMNLP 2024 Main Conference 🎉
”A Bayesian Approach to Harnessing the Power of LLMs in Authorship Attribution“
It works so well that a 7B open-source LLM outperforms GPT-4 with prompting at identifying unique writing style. 🔍
👉 Learn more: arxiv.org/pdf/2410.217...
”A Bayesian Approach to Harnessing the Power of LLMs in Authorship Attribution“
It works so well that a 7B open-source LLM outperforms GPT-4 with prompting at identifying unique writing style. 🔍
👉 Learn more: arxiv.org/pdf/2410.217...
Reposted by arXiv loVe
Excited to announce that our paper on training-data memorization and copyright has been accepted for presentation at ACM CSLaw '25!
The law review version is in press, forthcoming in early 2025.
arxiv.org/abs/2404.12590
The law review version is in press, forthcoming in early 2025.
arxiv.org/abs/2404.12590
December 4, 2024 at 12:31 AM
Excited to announce that our paper on training-data memorization and copyright has been accepted for presentation at ACM CSLaw '25!
The law review version is in press, forthcoming in early 2025.
arxiv.org/abs/2404.12590
The law review version is in press, forthcoming in early 2025.
arxiv.org/abs/2404.12590
Reposted by arXiv loVe
MRS was a wild ride and cleared up a lot of long-standing questions I had about chiral perovskites. On way home to teach tomorrow. Materials Science!
And the always remarkable Pete Sercel put this paper on arxiv to accompany his talk: arxiv.org/abs/2412.00602
Absolutely worth a read.
And the always remarkable Pete Sercel put this paper on arxiv to accompany his talk: arxiv.org/abs/2412.00602
Absolutely worth a read.
Optical activity of chiral excitons
Recent activity in the area of chiroptical phenomena has been focused on the connection between structural asymmetry, electron spin configuration and light matter interactions in chiral semiconductors...
urldefense.com
December 4, 2024 at 12:46 AM
MRS was a wild ride and cleared up a lot of long-standing questions I had about chiral perovskites. On way home to teach tomorrow. Materials Science!
And the always remarkable Pete Sercel put this paper on arxiv to accompany his talk: arxiv.org/abs/2412.00602
Absolutely worth a read.
And the always remarkable Pete Sercel put this paper on arxiv to accompany his talk: arxiv.org/abs/2412.00602
Absolutely worth a read.
Reposted by arXiv loVe
Hi all 👋. For my first post here, check out our recent paper using adaptive noise cancellation for removing instrumental noise lines for LIGO arxiv.org/abs/2412.01058
Adaptive cancellation of mains power interference in continuous gravitational wave searches with a hidden Markov model
Continuous gravitational wave searches with terrestrial, long-baseline interferometers are hampered by long-lived, narrowband features in the power spectral density of the detector noise, known as lin...
arxiv.org
December 4, 2024 at 1:30 AM
Hi all 👋. For my first post here, check out our recent paper using adaptive noise cancellation for removing instrumental noise lines for LIGO arxiv.org/abs/2412.01058
Reposted by arXiv loVe
This work was led by Amir Barda, in collaboration with Matheus
@gadelha.bsky.social
, Noam Aigerman
@noamiko.bsky.social
, Vova Kim @vovakim.bsky.social and Amit Bermano.
Check out our paper for more details 📜: arxiv.org/abs/2412.00518
7/end
@gadelha.bsky.social
, Noam Aigerman
@noamiko.bsky.social
, Vova Kim @vovakim.bsky.social and Amit Bermano.
Check out our paper for more details 📜: arxiv.org/abs/2412.00518
7/end
December 4, 2024 at 1:55 AM
This work was led by Amir Barda, in collaboration with Matheus
@gadelha.bsky.social
, Noam Aigerman
@noamiko.bsky.social
, Vova Kim @vovakim.bsky.social and Amit Bermano.
Check out our paper for more details 📜: arxiv.org/abs/2412.00518
7/end
@gadelha.bsky.social
, Noam Aigerman
@noamiko.bsky.social
, Vova Kim @vovakim.bsky.social and Amit Bermano.
Check out our paper for more details 📜: arxiv.org/abs/2412.00518
7/end
Reposted by arXiv loVe
This. Is. AWESOME. A resource for people who know women interested in careers in astronomy. I've had the great honor of having many of the authors in my group as part of their journey. So proud of all of them! Please circulate widely! arxiv.org/abs/2312.04041 #astronomy #astrophysics
December 4, 2024 at 2:00 AM
This. Is. AWESOME. A resource for people who know women interested in careers in astronomy. I've had the great honor of having many of the authors in my group as part of their journey. So proud of all of them! Please circulate widely! arxiv.org/abs/2312.04041 #astronomy #astrophysics
Reposted by arXiv loVe
Just dropped our new origami paper with Erik and Ryuhei on arXiv! arxiv.org/abs/2412.02174
I was the main person pushing this out, though I'm not the best at the technical stuff so the coauthors helped a lot with honing in the language and proofs, couldn't have done it without them!
I was the main person pushing this out, though I'm not the best at the technical stuff so the coauthors helped a lot with honing in the language and proofs, couldn't have done it without them!
All Polyhedral Manifolds are Connected by a 2-Step Refolding
We prove that, for any two polyhedral manifolds P, Q, there is a polyhedral manifold I such that P, I share a common unfolding and I, Q share a common unfolding. In other words, we can unfold P, refol...
arxiv.org
December 4, 2024 at 2:29 AM
Just dropped our new origami paper with Erik and Ryuhei on arXiv! arxiv.org/abs/2412.02174
I was the main person pushing this out, though I'm not the best at the technical stuff so the coauthors helped a lot with honing in the language and proofs, couldn't have done it without them!
I was the main person pushing this out, though I'm not the best at the technical stuff so the coauthors helped a lot with honing in the language and proofs, couldn't have done it without them!
Reposted by arXiv loVe
arxiv.org/abs/2412.02050
/An illustrated introduction to the arithmetic of Apollonian circle packings, continued fractions, and other thin orbits/
Katherine E. Stange
/An illustrated introduction to the arithmetic of Apollonian circle packings, continued fractions, and other thin orbits/
Katherine E. Stange
An illustrated introduction to the arithmetic of Apollonian circle packings, continued fractions, and other thin orbits
These notes cover and expand upon the material for two summer schools: The first, which was held at CIRM, Marseille, France, July 10-14, 2023, as part of "Renormalization and Visualization for packing...
arxiv.org
December 4, 2024 at 3:13 AM
arxiv.org/abs/2412.02050
/An illustrated introduction to the arithmetic of Apollonian circle packings, continued fractions, and other thin orbits/
Katherine E. Stange
/An illustrated introduction to the arithmetic of Apollonian circle packings, continued fractions, and other thin orbits/
Katherine E. Stange
Reposted by arXiv loVe
PROMYS folks will love Exercise 2.3!
cc #iTeachMath ♾️
bcc #MathSky 🧮
s/o @jlanier.bsky.social, too
arxiv.org/abs/2412.02050
cc #iTeachMath ♾️
bcc #MathSky 🧮
s/o @jlanier.bsky.social, too
arxiv.org/abs/2412.02050
An illustrated introduction to the arithmetic of Apollonian circle packings, continued fractions, and other thin orbits
These notes cover and expand upon the material for two summer schools: The first, which was held at CIRM, Marseille, France, July 10-14, 2023, as part of "Renormalization and Visualization for packing...
arxiv.org
December 4, 2024 at 3:23 AM
PROMYS folks will love Exercise 2.3!
cc #iTeachMath ♾️
bcc #MathSky 🧮
s/o @jlanier.bsky.social, too
arxiv.org/abs/2412.02050
cc #iTeachMath ♾️
bcc #MathSky 🧮
s/o @jlanier.bsky.social, too
arxiv.org/abs/2412.02050
Reposted by arXiv loVe
Ah, but have you seen stochastic calculus for Rosenblatt processes? arxiv.org/abs/1908.00296
A Stochastic Calculus for Rosenblatt Processes
A stochastic calculus is given for processes described by stochastic integrals with respect to fractional Brownian motions and Rosenblatt processes somewhat analogous to the stochastic calculus for It...
arxiv.org
December 4, 2024 at 3:45 AM
Ah, but have you seen stochastic calculus for Rosenblatt processes? arxiv.org/abs/1908.00296
Reposted by arXiv loVe
Thanks to the reviewers, we have a much improved revised version of our manuscript on predicting scale up of particle coating by #ALDep. Basically #ALDep may be slow, but when you reach large enough batches, it is not. arxiv.org/abs/2408.13116
Modeling scale-up of particle coating by atomic layer deposition
Atomic layer deposition (ALD) is a promising technique to functionalize particle surfaces for energy applications including energy storage, catalysis, and decarbonization. In this work, we present a s...
arxiv.org
December 4, 2024 at 3:47 AM
Thanks to the reviewers, we have a much improved revised version of our manuscript on predicting scale up of particle coating by #ALDep. Basically #ALDep may be slow, but when you reach large enough batches, it is not. arxiv.org/abs/2408.13116
Reposted by arXiv loVe
Looking forward to reading this!
"The Landscape of Causal Discovery Data: Grounding Causal Discovery in Real-World Applications"
arxiv.org/abs/2412.01953
#stats #mlsky #blueskai 📈📉
"The Landscape of Causal Discovery Data: Grounding Causal Discovery in Real-World Applications"
arxiv.org/abs/2412.01953
#stats #mlsky #blueskai 📈📉
December 4, 2024 at 3:50 AM
Looking forward to reading this!
"The Landscape of Causal Discovery Data: Grounding Causal Discovery in Real-World Applications"
arxiv.org/abs/2412.01953
#stats #mlsky #blueskai 📈📉
"The Landscape of Causal Discovery Data: Grounding Causal Discovery in Real-World Applications"
arxiv.org/abs/2412.01953
#stats #mlsky #blueskai 📈📉
Reposted by arXiv loVe
LLMForecaster: Improving Seasonal Event Forecasts
with Unstructured Textual Data
Use a fine-tuned LLM embeddings to rescale forecasts using natural language prompts to capture seasonal nuances
arxiv.org/abs/2412.02525
with Unstructured Textual Data
Use a fine-tuned LLM embeddings to rescale forecasts using natural language prompts to capture seasonal nuances
arxiv.org/abs/2412.02525
December 4, 2024 at 3:55 AM
LLMForecaster: Improving Seasonal Event Forecasts
with Unstructured Textual Data
Use a fine-tuned LLM embeddings to rescale forecasts using natural language prompts to capture seasonal nuances
arxiv.org/abs/2412.02525
with Unstructured Textual Data
Use a fine-tuned LLM embeddings to rescale forecasts using natural language prompts to capture seasonal nuances
arxiv.org/abs/2412.02525