Tootfinder

Published another couple of new versions of https://thi.ng/genart-api, mainly involving updates to the Layer (layer.com) platform adapter, adding config options and minor quality of life improvements (e.g. handling of private [aka artist-only] params)
I also updated the readme, clarifying the current versionin…

thi.ng/genart-api

Federated Multi-Agent Reinforcement Learning for Privacy-Preserving and Energy-Aware Resource Management in 6G Edge Networks
Francisco Javier Esono Nkulu Andong, Qi Min
https://arxiv.org/abs/2509.10163

Federated Multi-Agent Reinforcement Learning for Privacy-Preserving and Energy-Aware Resource Management in 6G Edge Networks
As sixth-generation (6G) networks move toward ultra-dense, intelligent edge environments, efficient resource management under stringent privacy, mobility, and energy constraints becomes critical. This paper introduces a novel Federated Multi-Agent Reinforcement Learning (Fed-MARL) framework that incorporates cross-layer orchestration of both the MAC layer and application layer for energy-efficient, privacy-preserving, and real-time resource management across heterogeneous edge devices. Each age…

When I was walking down from the mountains recently I noticed the frost on the trees and branches -- and some single leafs that were still around.
I really liked this extra touch of color in the grey surrounding. And that it's just a little detail on the huge mountain.
#photography #winter

Dr.LLM: Dynamic Layer Routing in LLMs
Ahmed Heakl, Martin Gubri, Salman Khan, Sangdoo Yun, Seong Joon Oh
https://arxiv.org/abs/2510.12773 https://arxiv.org…

Dr.LLM: Dynamic Layer Routing in LLMs
Large Language Models (LLMs) process every token through all layers of a transformer stack, causing wasted computation on simple queries and insufficient flexibility for harder ones that need deeper reasoning. Adaptive-depth methods can improve efficiency, but prior approaches rely on costly inference-time search, architectural changes, or large-scale retraining, and in practice often degrade accuracy despite efficiency gains. We introduce Dr.LLM, Dynamic routing of Layers for LLMs, a retrofitt…

Collinear, incommensurate antiferromagnetism in van der Waals magnet alpha-UTe3
H. Sakai, C. Tabata, K. Kaneko, Y. Tokiwa, T. Kitazawa, S. Kambe, Y. Tokunaga, Y. Haga
https://arxiv.org/abs/2510.10999

Collinear, incommensurate antiferromagnetism in van der Waals magnet alpha-UTe3
alpha-UTe3, a van der Waals (vdW) actinide compound with a monoclinic ZrSe3-type structure, is a narrow-gap semiconductor with 5f moments. 125Te NMR reveals strongly anisotropic, layer-confined spin fluctuations below about 20 K, with the a-axis component enhanced, and a signal wipeout at the antiferromagnetic (AFM) transition at TN = 5 K. Single-crystal neutron diffraction finds q approx. (0.17, 0.5, 0) and a longitudinal sinusoidal modulation of a-axis moments (amplitude about 0.8 muB) with A…

Would you be happy with HDDs packaged for delivery like this?
(Box of 6 hard disk drives; first photo on its side showing no filler around the individual drive boxes; second photo showing single piece of paper filler layer right at the bottom of the now-empty box.)

Global Convergence of Four-Layer Matrix Factorization under Random Initialization
Minrui Luo, Weihang Xu, Xiang Gao, Maryam Fazel, Simon Shaolei Du
https://arxiv.org/abs/2511.09925 https://arxiv.org/pdf/2511.09925 https://arxiv.org/html/2511.09925
arXiv:2511.09925v1 Announce Type: new
Abstract: Gradient descent dynamics on the deep matrix factorization problem is extensively studied as a simplified theoretical model for deep neural networks. Although the convergence theory for two-layer matrix factorization is well-established, no global convergence guarantee for general deep matrix factorization under random initialization has been established to date. To address this gap, we provide a polynomial-time global convergence guarantee for randomly initialized gradient descent on four-layer matrix factorization, given certain conditions on the target matrix and a standard balanced regularization term. Our analysis employs new techniques to show saddle-avoidance properties of gradient decent dynamics, and extends previous theories to characterize the change in eigenvalues of layer weights.
toXiv_bot_toot

Anybody know of a Linux image editing tool that lets you dynamically mesh-warp a layer (i.e. click a point in the image and move it to align with a feature on another layer, adding new mesh key points each time you do this)?

Virtual Agent Economies
Nenad Tomasev, Matija Franklin, Joel Z. Leibo, Julian Jacobs, William A. Cunningham, Iason Gabriel, Simon Osindero
https://arxiv.org/abs/2509.10147 https…

Virtual Agent Economies
The rapid adoption of autonomous AI agents is giving rise to a new economic layer where agents transact and coordinate at scales and speeds beyond direct human oversight. We propose the "sandbox economy" as a framework for analyzing this emergent system, characterizing it along two key dimensions: its origins (emergent vs. intentional) and its degree of separateness from the established human economy (permeable vs. impermeable). Our current trajectory points toward a spontaneous emergence of a …

Simple Projection Variants Improve ColBERT Performance
Benjamin Clavi\'e, Sean Lee, Rikiya Takehi, Aamir Shakir, Makoto P. Kato
https://arxiv.org/abs/2510.12327 https://

Simple Projection Variants Improve ColBERT Performance
Multi-vector dense retrieval methods like ColBERT systematically use a single-layer linear projection to reduce the dimensionality of individual vectors. In this study, we explore the implications of the MaxSim operator on the gradient flows of the training of multi-vector models and show that such a simple linear projection has inherent, if non-critical, limitations in this setting. We then discuss the theoretical improvements that could result from replacing this single-layer projection with …

LayerLock: Non-collapsing Representation Learning with Progressive Freezing
Goker Erdogan, Nikhil Parthasarathy, Catalin Ionescu, Drew Hudson, Alexander Lerchner, Andrew Zisserman, Mehdi Sajjadi, Joao Carreira
https://arxiv.org/abs/2509.10156

LayerLock: Non-collapsing Representation Learning with Progressive Freezing
We introduce LayerLock, a simple yet effective approach for self-supervised visual representation learning, that gradually transitions from pixel to latent prediction through progressive layer freezing. First, we make the observation that during training of video masked-autoencoding (MAE) models, ViT layers converge in the order of their depth: shallower layers converge early, deeper layers converge late. We then show that this observation can be exploited to accelerate standard MAE by progress…

Modeling Epidemics on Multiplex Networks: Epidemic Threshold and Basic Reproduction Number
Eric Alejandro Rozan, Mario Ignacio Simoy, Sebastian Bouzat, Marcelo Nestor Kuperman
https://arxiv.org/abs/2510.12614

Modeling Epidemics on Multiplex Networks: Epidemic Threshold and Basic Reproduction Number
Accurate epidemic forecasting requires models that account for the layered and heterogeneous nature of real social interactions. The basic reproduction number $\mathcal R_0$ calculated from models that assume homogeneous mixing or single-layer contact structures have limited applicability to complex social systems. Here, we propose an expression of $\mathcal R_0$ in the context of multiplex networks, enabling the analysis of disease transmission across multiple social layers. We adapt the Deg…

OpenAI's blockbuster deals with Nvidia and AMD add a new layer to its complicated ownership structure and will dilute existing shareholders like Microsoft (George Hammond/Financial Times)

RAID-0e: A Resilient Striping Array Architecture for Balanced Performance and Availability
Yanzhao Jia, Zhaobo Wu, Zheyi Cao, Shihao Ji, Xu Tianhao, Zihui Song
https://arxiv.org/abs/2510.12139

RAID-0e: A Resilient Striping Array Architecture for Balanced Performance and Availability
This paper introduces a novel disk array architecture, designated RAID-0e (Resilient Striping Array), designed to superimpose a low-overhead fault tolerance layer upon traditional RAID 0 (striping). By employing a logically and physically separate parity domain to protect a primary data domain, RAID-0e mitigates the risk of array-wide data loss from common, non-catastrophic media failures, such as isolated bad blocks, transient read errors, or sector-level corruption. The architecture is engine…

Dimension-Free Minimax Rates for Learning Pairwise Interactions in Attention-Style Models
Shai Zucker, Xiong Wang, Fei Lu, Inbar Seroussi
https://arxiv.org/abs/2510.11789 https:…

Dimension-Free Minimax Rates for Learning Pairwise Interactions in Attention-Style Models
We study the convergence rate of learning pairwise interactions in single-layer attention-style models, where tokens interact through a weight matrix and a non-linear activation function. We prove that the minimax rate is $M^{-\frac{2β}{2β+1}}$ with $M$ being the sample size, depending only on the smoothness $β$ of the activation, and crucially independent of token count, ambient dimension, or rank of the weight matrix. These results highlight a fundamental dimension-free statistical efficie…

Coordinated Motion Planning of a Wearable Multi-Limb System for Enhanced Human-Robot Interaction
Chaerim Moon, Joohyung Kim
https://arxiv.org/abs/2509.10444 https://

Coordinated Motion Planning of a Wearable Multi-Limb System for Enhanced Human-Robot Interaction
Supernumerary Robotic Limbs (SRLs) can enhance human capability within close proximity. However, as a wearable device, the generated moment from its operation acts on the human body as an external torque. When the moments increase, more muscle units are activated for balancing, and it can result in reduced muscular null space. Therefore, this paper suggests a concept of a motion planning layer that reduces the generated moment for enhanced Human-Robot Interaction. It modifies given trajectories…

Partial Self-Correction in Layer Codes
Dominic J. Williamson
https://arxiv.org/abs/2510.09218 https://arxiv.org/pdf/2510.09218

Partial Self-Correction in Layer Codes
The storage of large-scale quantum information at finite temperature requires an autonomous and reliable quantum hard drive, also known as a self-correcting quantum memory. It is a long-standing open problem to find a self-correcting quantum memory in three dimensions. The recently introduced Layer Codes achieve the best possible scaling of code parameters and logical energy barrier in three dimensions, these are tantalizing features for the purposes of self-correction. In this work we show tha…

Model of deep zonal flows in giant planets
Laura K. Currie, Chris A. Jones
https://arxiv.org/abs/2510.12236 https://arxiv.org/pdf/2510.12236

Model of deep zonal flows in giant planets
A mechanism by which the surface zonal flows of giant planets can be gradually attenuated with depth is explored. The zonal flow is driven by an imposed forcing in a thin layer near the surface. A meridional circulation is set up, analogous to the Ferrel-like cells observed in Jupiter's atmosphere. Acting on a stably stratified thin surface layer, the meridional flow induces a horizontal temperature anomaly which leads to a gradual reduction of the zonal winds with depth, governed by the therma…

'Ghost job' postings are adding another layer of uncertainty to the stalling jobs picture (Jeff Cox/CNBC)
https://www.cnbc.com/2025/11/11/ghost-job-postings-add-another-layer-of-uncertainty-to-stalled-jobs-picture.html
http://www.memeorandum.com/251111/p79#a251111p79

Good morning! For this #silentSunday I want to share this #foggy path with you.
I was on my way down and when I entered the fog, temperature really dropped. Actually I didn't want to stop or slow down in order to avoid cooling out.
But at this spot I had to take a quick break for…

Intrinsic Dimension Estimating Autoencoder (IDEA) Using CancelOut Layer and a Projected Loss
Antoine Orioua, Philipp Krah, Julian Koellermeier
https://arxiv.org/abs/2509.10011 h…

Intrinsic Dimension Estimating Autoencoder (IDEA) Using CancelOut Layer and a Projected Loss
This paper introduces the Intrinsic Dimension Estimating Autoencoder (IDEA), which identifies the underlying intrinsic dimension of a wide range of datasets whose samples lie on either linear or nonlinear manifolds. Beyond estimating the intrinsic dimension, IDEA is also able to reconstruct the original dataset after projecting it onto the corresponding latent space, which is structured using re-weighted double CancelOut layers. Our key contribution is the introduction of the projected reconstr…

Bluetooth Fingerprint Identification Under Domain Shift Through Transient Phase Derivative
Haytham Albousayri, Bechir Hamdaoui, Weng-Keen Wong, Nora Basha
https://arxiv.org/abs/2510.09940

Bluetooth Fingerprint Identification Under Domain Shift Through Transient Phase Derivative
Deep learning-based radio frequency fingerprinting (RFFP) has become an enabling physical-layer security technology, allowing device identification and authentication through received RF signals. This technology, however, faces significant challenges when it comes to adapting to domain variations, such as time, location, environment, receiver and channel. For Bluetooth Low Energy (BLE) devices, addressing these challenges is particularly crucial due to the BLE protocol's frequency-hopping natur…

Tuning Layer Orbital Hall Effect via Spin Rotation in Ferromagnetic Transition Metal Dichalcogenides
Shilei Ji, Jianping Yang, Li Gao, Xing'ao Li
https://arxiv.org/abs/2510.11180

Tuning Layer Orbital Hall Effect via Spin Rotation in Ferromagnetic Transition Metal Dichalcogenides
Orbitronics, which leverages the angular momentum of atomic orbitals for information transmission, provides a novel strategy to overcome the limitations of electronic devices. Unlike electron spin, orbital angular momentum (OAM) is strongly influenced by crystal field effects and band topology, making its orientation difficult to manipulate with external fields. In this work, by using first principle calculations, we investigate quantum anomalous Hall insulators (QAHIs) as a model system to stu…

An Encoder-Integrated PhoBERT with Graph Attention for Vietnamese Token-Level Classification
Ba-Quang Nguyen
https://arxiv.org/abs/2510.11537 https://arxiv…

An Encoder-Integrated PhoBERT with Graph Attention for Vietnamese Token-Level Classification
We propose a novel neural architecture named TextGraphFuseGAT, which integrates a pretrained transformer encoder (PhoBERT) with Graph Attention Networks for token-level classification tasks. The proposed model constructs a fully connected graph over the token embeddings produced by PhoBERT, enabling the GAT layer to capture rich inter-token dependencies beyond those modeled by sequential context alone. To further enhance contextualization, a Transformer-style self-attention layer is applied on …

Trajectory control of a suspended load with non-stopping flying carriers
Sofia Girardello, Giulia Michieletto, Angelo Cenedese, Antonio Franchi, Chiara Gabellieri
https://arxiv.org/abs/2510.11413

Trajectory control of a suspended load with non-stopping flying carriers
This paper presents the first closed-loop control framework for cooperative payload transportation with non-stopping flying carriers. Building upon grasp-matrix formulations and internal force redundancy, we propose a feedback wrench controller that actively regulates the payload's pose while an optimization layer dynamically shapes internal-force oscillations to guarantee persistent carrier motion. Preliminary experimental results on multirotor UAVs validate the model assumptions, and numerica…

Superconducting spin valve effect in Fe/Si$_3$N$_4$/Pb/Si$_3$N$_4$/Fe heterostructures
A. A. Kamashev, N. N. Garif'yanov, A. A. Validov, A. S. Osin, Ya. V. Fominov, I. A. Garifullin
https://arxiv.org/abs/2510.11333

Superconducting spin valve effect in Fe/Si$_3$N$_4$/Pb/Si$_3$N$_4$/Fe heterostructures
The structures of the superconducting spin valve (SSV) Fe/\allowbreak Si$_3$N$_4$/\allowbreak Pb/\allowbreak Si$_3$N$_4$/\allowbreak Fe (where Si$_3$N$_4$ is a dielectric insulating layer of controlled thickness) were investigated. The dependence of the magnitude of the SSV effect on the thicknesses of the superconducting (S) and insulating (I) layers was studied. Optimization of the S and I layer thicknesses enabled a complete switching between the normal and superconducting states when the mu…

A thread dumping some (non-obvious?) notes about #firewalld - I think I've got it doing what I want, but it's been a bit of a fight, and I made a bunch of wrong turns.
It's a layer on top of nftables or iptables; is XML based but you can just drive it from a command line tool firewall-cmd and avoid the XML.
It's main idea is a graph of 'zones' where a 'zo…

Magnetohydrodynamic shallow water equations with the alpha effect: Rossby-dynamo waves in solar--stellar tachoclines
T. V. Zaqarashvili, M. Dikpati, P. A. Gilman
https://arxiv.org/abs/2510.09130

Magnetohydrodynamic shallow water equations with the alpha effect: Rossby-dynamo waves in solar--stellar tachoclines
The activity of Sun-like stars is governed by the magnetic field, which is believed to be generated in a thin layer between convective and radiative envelopes. The dynamo layer, also called the tachocline, permits the existence of Rossby waves (r-modes) described by magnetohydrodynamic shallow water models, which may lead to short-term cycles in stellar activity. Convective cells penetrate into the layer creating an overshoot upper part, where they transport an additional energy for vigorous ac…

Hybrid MAC Protocol with Integrated Multi-Layered Security for Resource-Constrained UAV Swarm Communications
Dhrumil Bhatt, Siddharth Penumatsa, Vidushi Kumar
https://arxiv.org/abs/2510.10236

Hybrid MAC Protocol with Integrated Multi-Layered Security for Resource-Constrained UAV Swarm Communications
Flying Ad Hoc Networks (FANETs) present unique challenges due to high node mobility, dynamic topologies, and strict resource constraints. Existing routing protocols often optimize for a single metric, such as path length or energy, while neglecting the complex dependencies between network performance, security, and MAC layer efficiency. This paper introduces a novel hardware software co design framework for secure and adaptive UAV swarm communications, featuring an energy aware protocol stack. …

Near room temperature magnetoelectric response and tunable magnetic anisotropy in the two-dimensional magnet 1T-CrTe2
Fengping Li, Bheema Lingam Chittari, Chao Lei, Jeil Jung
https://arxiv.org/abs/2510.10547

Near room temperature magnetoelectric response and tunable magnetic anisotropy in the two-dimensional magnet 1T-CrTe2
Magnets with controllable magnetization and high critical temperature are essential for practical spintronics devices, among which the two-dimensional 1T-CrTe2 stands out because of its high experimental critical temperature up to about 300K down to the single layer limit. By using ab initio density functional theory, we investigate the magnetic properties of monolayer and bilayer 1T-CrTe2 and demonstrate that the magnetic properties, such as the magnetocrystalline anisotropy, critical Curie te…

Polymer Topology and the Depletion Interaction
Mauro L. Mugnai
https://arxiv.org/abs/2510.09890 https://arxiv.org/pdf/2510.09890

Polymer Topology and the Depletion Interaction
Using a theoretical model we show that ideal ring polymers are stronger depletants than ideal linear polymers of equal radii of gyration, but not of equal hydrodynamic radii. The difference in the depletion-induced force profile is largely controlled by the thickness of the depletion layer. Theory suggests that this thickness is equal to the average extent of a polymer along the direction perpendicular to the surfaces of the colloids. Within the limits of finite-size effects, Molecular Dynamics…

Dipole-quadrupole model and multipole analysis of resonant membrane metasurfaces
Izzatjon Allayarov, Andrey B. Evlyukhin, Antonio Cal\`a Lesina
https://arxiv.org/abs/2510.11864 …

Dipole-quadrupole model and multipole analysis of resonant membrane metasurfaces
Membrane-metasurfaces, formed by periodic arrangements of holes in a dielectric layer, are gaining attention for their easier manufacturing via subtractive techniques, unnecessity of substrates, and access to resonant near-fields. Despite their practical relevance, their theoretical description remains elusive. Here, we present a semi-analytical dipole-quadrupole model for the multipole analysis of numerically-obtained reflection and transmission spectra in metasurfaces excited at arbitrary ang…

Towards fully predictive gyrokinetic full-f simulations
A. C. D. Hoffmann, T. N. Bernard, M. Francisquez, G. W. Hammett, A. Hakim, J. Boedo, R. Rizkallah, C. K. Tsui, the TCV team
https://arxiv.org/abs/2510.11874

Towards fully predictive gyrokinetic full-f simulations
Designing economical magnetic confinement fusion power plants motivates computational tools that can estimate plasma behavior from engineering parameters without direct reliance on experimental measurement of the plasma profiles. In this work, we present full-$f$ global gyrokinetic (GK) turbulence simulations of edge and scrape-off layer turbulence in tokamaks that use only magnetic geometry, heating power, and particle inventory as inputs. Unlike many modeling approaches that employ free param…

Discrete non-abelian X-ray transforms
Pranav Gupta, Roman Novikov
https://arxiv.org/abs/2510.09611 https://arxiv.org/pdf/2510.09611

Discrete non-abelian X-ray transforms
We define a discrete version of the non-abelian X-ray transform, going back in particular to Manakov, Zakharov (1981) and Strichartz (1982). We extend to this transform non-overdetermined reconstruction results obtained for the abelian case in the recent article by Novikov, Sharma (2025). In addition, we establish relations with the continuous non-abelian X-ray transform. In this respect, our results include an explicit and exact non-overdetermined layer-stripping reconstruction procedure for p…

Empowering LLM Agents with Geospatial Awareness: Toward Grounded Reasoning for Wildfire Response
Yiheng Chen, Lingyao Li, Zihui Ma, Qikai Hu, Yilun Zhu, Min Deng, Runlong Yu
https://arxiv.org/abs/2510.12061

Empowering LLM Agents with Geospatial Awareness: Toward Grounded Reasoning for Wildfire Response
Effective disaster response is essential for safeguarding lives and property. Existing statistical approaches often lack semantic context, generalize poorly across events, and offer limited interpretability. While Large language models (LLMs) provide few-shot generalization, they remain text-bound and blind to geography. To bridge this gap, we introduce a Geospatial Awareness Layer (GAL) that grounds LLM agents in structured earth data. Starting from raw wildfire detections, GAL automatically r…

IPbus extension for the ALFRED framework
Krystian Roslon, Maciej Czarnynoga, Sebastian Koryciak, Monika Kutyla, Ignacy Mermer, Jacek Tomasz Otwinowski, Wiktor Pierozak, Pawel Grzegorz Russek, Wladyslaw Trzaska, Franciszek Urbanski
https://arxiv.org/abs/2510.11165

IPbus extension for the ALFRED framework
We present a development of the ALFRED framework that includes support for the IPbus protocol, created for the Fast Interaction Trigger (FIT) detector in the ALICE experiment at CERN. This modification resolves the incompatibility between the current GBT-based slow-control protocols and the FIT electronics, which is based on the IPbus. A compatibility layer, named ALF IPbus, was created in C++ and integrated with DIM services, allowing for the transparent conversion of SWT frames into IPbus tra…

Wiedemann-Franz behavior at the Weyl points in compressively strained HgTe
Abu Alex Aravindnath, Yi-Ju Ho, Fabian Schmitt, Dongyun Chen, Johannes Kleinlein, Wouter Beugeling, Hartmut Buhmann, Stanislau U. Piatrusha, Laurens W. Molenkamp
https://arxiv.org/abs/2510.12339

Wiedemann-Franz behavior at the Weyl points in compressively strained HgTe
Weyl semimetals, with their unique electronic band structure, have drawn significant interest for their potential to explore quantum anomalies in condensed matter systems. In this study, we investigate the large positive magneto-thermal conductance associated with the gravitational anomaly -- one of the predicted anomalies -- for a Weyl semimetal based on a compressively strained HgTe layer. We clearly identify the Weyl regime in our device and accurately extract the thermal conductance by perf…

Quantum criticality at the end of a pseudogap phase in superconducting infinite-layer nickelates
C. Iorio-Duval, E. Beauchesne-Blanchet, F. Perreault, J. L. Santana Gonz\'alez, W. Sun, Y. F. Nie, A. Gourgout, G. Grissonnanche
https://arxiv.org/abs/2510.12786

Quantum criticality at the end of a pseudogap phase in superconducting infinite-layer nickelates
In many unconventional superconductors, the strange-metal regime is thought to emerge from quantum criticality, yet in cuprates this link is obscured by the enigmatic pseudogap. Superconducting infinite-layer nickelates provide a new arena to test this paradigm but are constrained to thin films, precluding calorimetry. We use the Seebeck coefficient as a low-temperature proxy for entropy per carrier and uncover a clear quantum-critical thermodynamic signature: in La$_{1-x}$Sr$_x$NiO$_2$ at the …

🇺🇦 #NowPlaying on KEXP's #MiddayShow
The Smashing Pumpkins:
🎵 1979
#TheSmashingPumpkins
https://soundwavesofficial.bandcamp.com/album/the-smashing-pumpkins-1979-layer-by-layer
https://open.spotify.com/track/3h5zik31hTTat9jmpCZZNC

I needed to replace* my lovely extra long yoga mat and I’ve been angry for a week that I can only buy longer mats (I’m 178 cm tall, most mats are about 180) that are also wider. My new cute blue mat got delivered today in time for my class and … actually it was great. The extra width means less shuffling between poses for long limbs.
* the top layer unstuck from the bottom layer and stuck to itself instead to create a sort of curly lettuce effect, after years of loyal service!

Time to harness the sun! ☀️🛰️
We are LIVE for today's 3D printing stream, and we're adding more solar arrays to the Tiangong space station model. Powering the space race, one layer at a time (and still running smooth on Linux, naturally 🐧).
Tune in to watch the model grow: 👇 https://www.twitch.tv/tuxramus

Remember when Ethernet switches (with spanning tree loop avoidance) came out and companies started building nation-wide layer-2 networks?
And then those networks collapsed under broadcast storms?
I am starting to get the impression that those lessons are being forgotten.

On Thompson Sampling and Bilateral Uncertainty in Additive Bayesian Optimization
Nathan Wycoff
https://arxiv.org/abs/2510.11792 https://arxiv.org/pdf/2510.…

On Thompson Sampling and Bilateral Uncertainty in Additive Bayesian Optimization
In Bayesian Optimization (BO), additive assumptions can mitigate the twin difficulties of modeling and searching a complex function in high dimension. However, common acquisition functions, like the Additive Lower Confidence Bound, ignore pairwise covariances between dimensions, which we'll call \textit{bilateral uncertainty} (BU), imposing a second layer of approximations. While theoretical results indicate that asymptotically not much is lost in doing so, little is known about the practical e…

Phase Aware Ear-Conditioned Learning for Multi-Channel Binaural Speaker Separation
Ruben Johnson Robert Jeremiah, Peyman Goli, Steven van de Par
https://arxiv.org/abs/2510.11366

Phase Aware Ear-Conditioned Learning for Multi-Channel Binaural Speaker Separation
Separating competing speech in reverberant environments requires models that preserve spatial cues while maintaining separation efficiency. We present a Phase-aware Ear-conditioned speaker Separation network using eight microphones (PEASE-8) that consumes complex STFTs and directly introduces a raw-STFT input to the early decoder layer, bypassing the entire encoder pathway to improve reconstruction. The model is trained end-to-end with an SI-SDR-based objective against direct-path ear targets, …

PIC12F683 sample 2H6, after 14 HF etch cycles and a clean.
This sample has now been imaged 30 times, ranging from fresh after decap to the current state, after each etch, after cleaning passes to remove loose debris, and sometimes at both 20x and 100x in the same state.
This sample is now mostly at via1, but there's still upper layer visible in the seal ring and in the analog IPs.
It's now imaging at 100x. Things go slow because a full focus-stacked 100x image is 10…

MTMD: A Multi-Task Multi-Domain Framework for Unified Ad Lightweight Ranking at Pinterest
Xiao Yang, Peifeng Yin, Abe Engle, Jinfeng Zhuang, Ling Leng
https://arxiv.org/abs/2510.09857

MTMD: A Multi-Task Multi-Domain Framework for Unified Ad Lightweight Ranking at Pinterest
The lightweight ad ranking layer, living after the retrieval stage and before the fine ranker, plays a critical role in the success of a cascaded ad recommendation system. Due to the fact that there are multiple optimization tasks depending on the ad domain, e.g., Click Through Rate (CTR) for click ads and Conversion Rate (CVR) for conversion ads, as well as multiple surfaces where an ad is served (home feed, search, or related item recommendation) with diverse ad products (shopping or standard…

Ein internationales #Forschungsteam hat die Langzeitstabilität von #Perowskit-#Solarzellen deutlich verbessert.
Durch eine neuartige Fluorverbindung als Zwischenschicht bleibt ein

Long-term stability for perovskite solar cells: a big step forward
A Teflon-like layer between the perovskite and the top contact ensures exceptional stability: even after 1,200 hours under a standard sun, the high efficiency does not decrease.

I'm so grateful for my S.A.D light as someone rudely deployed a thick, moist cloud layer today and the sun has been out of view.
#SAD #SeasonalAffectiveDisorder

62.6 GHz ScAlN Solidly Mounted Acoustic Resonators
Yinan Wang, Byeongjin Kim, Nishanth Ravi, Kapil Saha, Supratik Dasgupta, Vakhtang Chulukhadze, Eugene Kwon, Lezli Matto, Pietro Simeoni, Omar Barrera, Ian Anderson, Tzu-Hsuan Hsu, Jue Hou, Matteo Rinaldi, Mark S. Goorsky, Ruochen Lu
https://arxiv.org/abs/2510.11994

62.6 GHz ScAlN Solidly Mounted Acoustic Resonators
We demonstrate a record-high 62.6 GHz solidly mounted acoustic resonator (SMR) incorporating a 67.6 nm scandium aluminum nitride (Sc0.3Al0.7N) piezoelectric layer on a 40 nm buried platinum (Pt) bottom electrode, positioned above an acoustic Bragg reflector composed of alternating SiO2 (28.2 nm) and Ta2O5 (24.3 nm) layers in 8.5 pairs. The Bragg reflector and piezoelectric stack above are designed to confine a third-order thickness-extensional (TE) bulk acoustic wave (BAW) mode, while efficient…

I think today's worker is the owner of the company, he certainly is the assessor.
He worked a lot later than those without the vested interest there. 😆 I finished work at my job before he did! 🤭
All the wood was sanded down and remaining nude wood given some paint. We have a test plank in the foreground of the first picture here which is painted with a second coat of paint. Seems likely we lose all the wood grain when doing that, and so will prefer the paler look where it's obviously made of wood not paint.
Won't really know for sure till it's dry. Prefer the colour a bit darker like that but if we're hiding the wood grain we might as well have used MDF instead of pine. We're after something clearly made of wood.
Another area is test-painted with just the clearcoat top varnish as a second layer. That's likely to be right, just a bit more shiny and protected.
The carpenter proper is back from holiday and starts tomorrow. He has a lot of drawers and doors to build and edging to attach to make the door panels. Still hoping at least the carpentry will be pretty much all done by the end of the week but likely some painting and touching up still to do next week. Hopefully by the end of Tuesday because I'm not really able to be here all day each day for most of the two weeks after that.

Keep Calm and Avoid Harmful Content: Concept Alignment and Latent Manipulation Towards Safer Answers
Ruben Belo, Claudia Soares, Marta Guimaraes
https://arxiv.org/abs/2510.12672

Keep Calm and Avoid Harmful Content: Concept Alignment and Latent Manipulation Towards Safer Answers
Large Language Models are susceptible to jailbreak attacks that bypass built-in safety guardrails (e.g., by tricking the model with adversarial prompts). We propose Concept Alignment and Concept Manipulation \textbf{CALM}, an inference-time method that suppresses harmful concepts by modifying latent representations of the last layer of the model, without retraining. Leveraging \gls*{cw} technique from Computer Vision combined with orthogonal projection, CALM removes unwanted latent directions a…

Automated Glaucoma Report Generation via Dual-Attention Semantic Parallel-LSTM and Multimodal Clinical Data Integration
Cheng Huang, Weizheng Xie, Zeyu Han, Tsengdar Lee, Karanjit Kooner, Jui-Ka Wang, Ning Zhang, Jia Zhang
https://arxiv.org/abs/2510.10037

Automated Glaucoma Report Generation via Dual-Attention Semantic Parallel-LSTM and Multimodal Clinical Data Integration
Generative AI for automated glaucoma diagnostic report generation faces two predominant challenges: content redundancy in narrative outputs and inadequate highlighting of pathologically significant features including optic disc cupping, retinal nerve fiber layer defects, and visual field abnormalities. These limitations primarily stem from current multimodal architectures' insufficient capacity to extract discriminative structural-textural patterns from fundus imaging data while maintaining pre…

ResearStudio: A Human-Intervenable Framework for Building Controllable Deep-Research Agents
Linyi Yang, Yixuan Weng
https://arxiv.org/abs/2510.12194 https://

ResearStudio: A Human-Intervenable Framework for Building Controllable Deep-Research Agents
Current deep-research agents run in a ''fire-and-forget'' mode: once started, they give users no way to fix errors or add expert knowledge during execution. We present ResearStudio, the first open-source framework that places real-time human control at its core. The system follows a Collaborative Workshop design. A hierarchical Planner-Executor writes every step to a live ''plan-as-document,'' a fast communication layer streams each action, file change, and tool call to a web interface. At any …

Lighter, which operates an Ethereum Layer 2 network and a decentralized exchange to support perpetual futures contracts, raised $68M at a ~$1.5B valuation (Naga Avan-Nomayo/The Block)
https://www.theblock.co/post/378358/lighte

Credal Transformer: A Principled Approach for Quantifying and Mitigating Hallucinations in Large Language Models
Shihao Ji, Zihui Song, Jiajie Huang
https://arxiv.org/abs/2510.12137

Credal Transformer: A Principled Approach for Quantifying and Mitigating Hallucinations in Large Language Models
Large Language Models (LLMs) hallucinate, generating factually incorrect yet confident assertions. We argue this stems from the Transformer's Softmax function, which creates "Artificial Certainty" by collapsing ambiguous attention scores into a single probability distribution, discarding uncertainty information at each layer. To fix this, we introduce the Credal Transformer, which replaces standard attention with a Credal Attention Mechanism (CAM) based on evidential theory. CAM produces a "cre…

🖖 Oh No, Someone Built A Game-Crashing Death Star In No Man’s Sky
#games

"Ozone layer recovery shows what global cooperation can achieve"
#OzoneLayer #Emissions

Ozone layer recovery shows what global cooperation can achieve | The Optimist Daily
The UN’s latest ozone bulletin finds steady recovery under the Montreal Protocol, a proof point that science-based policy can repair the planet.

Low Reynolds number flow in a packed bed of rotated bars
Wojciech Sadowski, Christin Velten, Maximilian Br\"ommer, Hakan Demir, Kerstin H\"ulz, Francesca di Mare, Katharina Z\"ahringer, Viktor Scherer
https://arxiv.org/abs/2510.12571

Low Reynolds number flow in a packed bed of rotated bars
The present study focuses on the gas flow through an experiment-scale modular packed bed reactor consisting of square bars, arranged in layers. Each layer is rotated by $30^\circ$ resulting in a complex shape of the void spaces between the bars. Particle Image Velocimetry measurement results inside and on top of the studied system are presented for particle-based Reynolds numbers of 100 and 200, and used as validation data for two sets of particle-resolved numerical simulations, using boundary …

HANDO: Hierarchical Autonomous Navigation and Dexterous Omni-loco-manipulation
Jingyuan Sun, Chaoran Wang, Mingyu Zhang, Cui Miao, Hongyu Ji, Zihan Qu, Han Sun, Bing Wang, Qingyi Si
https://arxiv.org/abs/2510.09221

HANDO: Hierarchical Autonomous Navigation and Dexterous Omni-loco-manipulation
Seamless loco-manipulation in unstructured environments requires robots to leverage autonomous exploration alongside whole-body control for physical interaction. In this work, we introduce HANDO (Hierarchical Autonomous Navigation and Dexterous Omni-loco-manipulation), a two-layer framework designed for legged robots equipped with manipulators to perform human-centered mobile manipulation tasks. The first layer utilizes a goal-conditioned autonomous exploration policy to guide the robot to sema…

Host-atom-driven transformation of a honeycomb oxide into a dodecagonal quasicrystal
Martin Haller, Julia Hewelt, V. Y. M. Rajesh Chirala, Loi Vinh Tran, Ankur Bhide, Muriel Wegner, Stefan F\"orster, Wolf Widdra
https://arxiv.org/abs/2510.11426

Host-atom-driven transformation of a honeycomb oxide into a dodecagonal quasicrystal
Dodecagonal oxide quasicrystals (OQCs) have so far been limited to a few elemental systems, with no general formation mechanism established. Here, we demonstrate a versatile approach to OQC formation via a host-atom-induced transformation of a metal-oxide honeycomb (HC) network. Adsorption of Ba, Sr, or Eu onto the HC layer triggers its reorganization into a dodecagonal tiling, as revealed by low-energy electron diffraction and scanning tunneling microscopy. Full conversion occurs when 73% of t…

Multiplexons: Limits of Multiplex Networks
Ankan Ganguly, Bhaswar B. Bhattacharya
https://arxiv.org/abs/2510.08639 https://arxiv.org/pdf/2510.08639

Multiplexons: Limits of Multiplex Networks
In a multiplex network, a set of nodes is connected by different types of interactions, each represented as a separate layer within the network. Multiplexes have emerged as a key instrument for modeling large-scale complex systems, due to the widespread coexistence of diverse interactions in social, industrial, and biological domains. This motivates the development of a rigorous and readily applicable framework for studying properties of large multiplex networks. In this article, we provide a s…

paris_transportation: Paris Transportation (2016)
A multiplex transportation network of Ile-de-France, jointly representing the road network, train network, and subway network. Nodes are intersections, rail stations or metro stations. Different modes of transportation are given by different layers. Within-layer edges represent direct transference by a particular mode of transportation, and between-layer edges represent the conjunction of those layers at that node.
This network ha…

Fluidity and morphological stability of an amorphous thin film with radiation-induced defect kinetics
Tyler P. Evans, Eden Heyen
https://arxiv.org/abs/2510.10745 https://…

Fluidity and morphological stability of an amorphous thin film with radiation-induced defect kinetics
It is common to model ion-irradiated amorphous thin films as if they were highly viscous fluids. In such models, one is frequently concerned with the ion-enhanced fluidity, a measure of the ability of the free interface to relax surface energy. Motivated by usual fluid dynamics problems, the ion-enhanced fluidity is near-universally treated as a constant throughout the amorphous layer. However, for an irradiated thin film, the fluidity is ultimately caused by radiation-induced defect kinetics w…

Exploring phase sensitivity and limit of detection near the critical coupling of metasurfaces and its phase singularity
L. Berguiga, T. Girerd, X. Letartre, T. Benyattou, C. Jamois, F. Mandorlo, L. Ferrier
https://arxiv.org/abs/2510.10153

Exploring phase sensitivity and limit of detection near the critical coupling of metasurfaces and its phase singularity
It is commonly accepted that phase singularities in refractive index sensors can provide highly sensitive detection. To address this issue, we studied the phase sensitivity and the limit of detection of Tamm photonic crystals used as temperature sensors, taken here as a model system by exploring critical coupling and its associated phase singularity. To finely tune the optical Tamm mode to critical coupling, the top metal layer is periodically nanostructured and controlled, enabling the investi…

Here's a photo that's a bit different from my usual style.
I took it with the tele lens on last weekend's hike. The inversion caused such a great sea of clouds with just some sole islands.
This kind of weather really offers such unique motives. And again: it was worth carrying the zoom lens up there :-D
Enjoy!
#photography

The Hidden Width of Deep ResNets: Tight Error Bounds and Phase Diagrams
L\'ena\"ic Chizat
https://arxiv.org/abs/2509.10167 https://arxiv.org/pdf/2…

The Hidden Width of Deep ResNets: Tight Error Bounds and Phase Diagrams
We study the gradient-based training of large-depth residual networks (ResNets) from standard random initializations. We show that with a diverging depth $L$, a fixed embedding dimension $D$, and an arbitrary hidden width $M$, the training dynamics converges to a Neural Mean ODE training dynamics. Remarkably, the limit is independent of the scaling of $M$, covering practical cases of, say, Transformers, where $M$ (the number of hidden units or attention heads per layer) is typically of the orde…

Characterisation of the first wafer-scale prototype for the ALICE ITS3 upgrade: the monolithic stitched sensor (MOSS)
Omar Abdelrahman, Gianluca Aglieri Rinella, Luca Aglietta, Giacomo Alocco, Matias Antonelli, Roberto Baccomi, Francesco Barile, Pascal Becht, Franco Benotto, Stefania Maria Beol\`e, Marcello Borri, Daniela Bortoletto, Naseem Bouchhar, Giuseppe Eugenio Bruno, Matthew Daniel Buckland, Szymon Bugiel, Paolo Camerini, Francesca Carnesecchi, Marielle Chartier, Domenico Colell…

Characterisation of the first wafer-scale prototype for the ALICE ITS3 upgrade: the monolithic stitched sensor (MOSS)
This paper presents the characterisation and testing of the first wafer-scale monolithic stitched sensor (MOSS) prototype developed for the ALICE ITS3 upgrade that is to be installed during the LHC Long Shutdown 3 (2026-2030). The MOSS chip design is driven by the truly cylindrical detector geometry that imposes that each layer is built out of two wafer-sized, bent silicon chips. The stitching technique is employed to fabricate sensors with dimensions of 1.4 $\times$ 25.9 cm, thinned to 50 $μ$…

Analysing Moral Bias in Finetuned LLMs through Mechanistic Interpretability
Bianca Raimondi, Daniela Dalbagno, Maurizio Gabbrielli
https://arxiv.org/abs/2510.12229 https://

Analysing Moral Bias in Finetuned LLMs through Mechanistic Interpretability
Large language models (LLMs) have been shown to internalize human-like biases during finetuning, yet the mechanisms by which these biases manifest remain unclear. In this work, we investigated whether the well-known Knobe effect, a moral bias in intentionality judgements, emerges in finetuned LLMs and whether it can be traced back to specific components of the model. We conducted a Layer-Patching analysis across 3 open-weights LLMs and demonstrated that the bias is not only learned during finet…

The spin Hall conductivity in the hole-doped bilayer Haldane-Hubbard model with odd-parity ALM
Minghuan Zeng, Ling Qin, Shiping Feng, Dong-Hui Xu, Rui Wang
https://arxiv.org/abs/2510.12602

The spin Hall conductivity in the hole-doped bilayer Haldane-Hubbard model with odd-parity ALM
Spin current generated electrically is among the core phenomena of spintronics for driving high-performance spin device applications. Here, on the basis of systematic investigations for the hole doped single-layer Haldane-Hubbard(HH) model, we propose a new bilayer HH model to realize the compensated odd-parity spin splitting and the $T$-even spin Hall conductivity where the two layers are connected by the time reversal transformation. Our results show that the vanishing layer-dependent electri…

Wireless Channel Modeling for Machine Learning -- A Critical View on Standardized Channel Models
Benedikt B\"ock, Amar Kasibovic, Wolfgang Utschick
https://arxiv.org/abs/2510.12279

Wireless Channel Modeling for Machine Learning -- A Critical View on Standardized Channel Models
Standardized (link-level) channel models such as the 3GPP TDL and CDL models are frequently used to evaluate machine learning (ML)-based physical-layer methods. However, in this work, we argue that a link-level perspective incorporates limiting assumptions, causing unwanted distributional shifts or necessitating impractical online training. An additional drawback is that this perspective leads to (near-)Gaussian channel characteristics. Thus, ML-based models, trained on link-level channel data,…

Evaluating Earth-Observing Satellite Sampling Effectiveness Using Kullback-Leibler Divergence
Negin Esmaeili, Paul T. Grogan
https://arxiv.org/abs/2510.10859 https://

Evaluating Earth-Observing Satellite Sampling Effectiveness Using Kullback-Leibler Divergence
This work presents an objective, repeatable, automatic, and fast methodology for assessing the representativeness of geophysical variables sampled by Earth-observing satellites. The primary goal is to identify and mitigate potential sampling biases attributed to orbit selection during pre-Phase A mission studies. This methodology supports current incubation activities for a future Planetary Boundary Layer observing system by incorporating a sampling effectiveness measure into a broader architec…

URLLC for 6G Enabled Industry 5.0: A Taxonomy of Architectures, Cross Layer Techniques, and Time Critical Applications
Abdikarim Mohamed Ibrahim, Rosdiadee Nordin, Yahya S. M. Khamayseh, Angela Amphawan, Muhammed Basheer Jasser
https://arxiv.org/abs/2510.08080

URLLC for 6G Enabled Industry 5.0: A Taxonomy of Architectures, Cross Layer Techniques, and Time Critical Applications
The evolution from Industry 4.0 to Industry 5.0 introduces stringent requirements for ultra reliable low latency communication (URLLC) to support human centric, intelligent, and resilient industrial systems. Sixth-generation (6G) wireless networks aim to meet these requirements through sub-millisecond end-to-end delays, microsecond level jitter, and near perfect reliability, enabled by advances such as terahertz (THz) communication, reconfigurable intelligent surfaces (RIS), multi-access edge c…

Impact of Wind Direction on Flow Over a Realistic Urban Area: A Large-Eddy Simulation Study
Ivette Rodr\'iguez, Josep Maria Dur\'o, Ernest Mestres, Ming Teng, Oriol Lehmkuhl
https://arxiv.org/abs/2510.11247

Impact of Wind Direction on Flow Over a Realistic Urban Area: A Large-Eddy Simulation Study
We conducted high-resolution large-eddy simulations over a real urban district in Barcelona to examine the impact of wind direction on near-ground flow. The computational mesh resolves over 500 million degrees of freedom, with a spatial resolution on the order of 1 m at pedestrian level. This allows a detailed analysis of mean velocity and turbulence patterns within the canopy layer. Although instantaneous flow fields differ significantly between cases, double-averaged profiles of velocity and …

Ambient-Stable Transfer-Free Graphdiyne Wafers with Superhigh Hole Mobility at Room Temperature
Beining Ma, Jianyuan Qi, Xinghai Shen
https://arxiv.org/abs/2510.09998 https://…

Ambient-Stable Transfer-Free Graphdiyne Wafers with Superhigh Hole Mobility at Room Temperature
Graphdiyne (GDY) is recognized as a compelling candidate for the fabrication of next-generation high-speed low-energy electronic devices due to its inherent p-type semiconductor characteristics. However, the development of GDY for applications in field-effect transistors (FETs), complementary metal-oxide-semiconductor (CMOS), and logic devices remains constrained by the relatively low carrier mobility reported in current experimental studies. Herein, the synthesis of layer-controlled hydrogen-s…

Mapping the moir\'e potential in multi-layer rhombohedral graphene
Eric Seewald, Sanat Ghosh, Nishchhal Verma, John Cenker, Yinan Dong, Birui Yang, Amit Basu, Takashi Taniguchi, Kenji Watanabe, Mandar M. Deshmukh, Dmitri N. Basov, Raquel Queiroz, Cory Dean, Abhay N. Pasupathy
https://arxiv.org/abs/2510.09548

Mapping the moiré potential in multi-layer rhombohedral graphene
Rhombohedral graphene (rG) aligned with hexagonal boron nitride (hBN) has been shown to host flat bands that stabilize various strongly correlated quantum phases, including Mott insulators, integer, and fractional quantum anomalous Hall phases. In this work, we use scanning tunneling microscopy/spectroscopy (STM/STS) to visualize the dispersion of flat bands with doping and applied displacement fields in a hBN-aligned rhombohedral trilayer graphene (rtG)/hBN moiré superlattice. In addition to …

Deconstructing Attention: Investigating Design Principles for Effective Language Modeling
Huiyin Xue, Nafise Sadat Moosavi, Nikolaos Aletras
https://arxiv.org/abs/2510.11602 htt…

Deconstructing Attention: Investigating Design Principles for Effective Language Modeling
The success of Transformer language models is widely credited to their dot-product attention mechanism, which interweaves a set of key design principles: mixing information across positions (enabling multi-token interactions), sequence-dependent activations (where attention weights adapt to each input), a specific mathematical form (dot-product similarities plus softmax weighting), and coupling of queries and keys to evolving hidden states (grounding attention in the current layer). However, th…

Checking on the Ceres FPGA board sims for @… that I had running while I was away. We're looking at a PCIe differential pair layer transition plus AC coupling capacitor.
With no cutout (green trace) response is awful: -3 dB insertion loss at 9.1 GHz, and -10 dB return loss at only 2.8 GHz.
Adding the cutout in the gerbers (blue trace) ma…

Dual-Waveguide Pinching Antennas for PLS: Parallel Placement or Orthogonal Placement?
Yang Lu, Xinke Xie, Yanqing Xu, Bo Ai, Octavia A. Dobre, Dusit Niyato
https://arxiv.org/abs/2510.11044

Dual-Waveguide Pinching Antennas for PLS: Parallel Placement or Orthogonal Placement?
Pinching antennas (PAs), as an emerging flexible-antenna technology, enables movable PAs deployed along waveguides to customize channel conditions over a large scale. This paper investigates an application of PAs to enable physical-layer security (PLS) by enlarging the channel condition diversity between legitimate users (LUs) and eavesdroppers (Eves). Particularly, we focus on the dual-waveguide scenario, where the two waveguides employs multiple PAs to serve multiple LUs in the presence of an…

BaldWhisper: Faster Whisper with Head Shearing and Layer Merging
Yaya Sy, Christophe Cerisara, Irina Illina
https://arxiv.org/abs/2510.08599 https://arxiv.…

BaldWhisper: Faster Whisper with Head Shearing and Layer Merging
Pruning large pre-trained transformers for low-resource languages is challenging, as it often requires massive retraining data to recover performance. For instance, Distill-Whisper prunes Whisper by 40% and retrains on 21,000 hours of speech, far beyond what is available for most languages. Can Whisper be made lighter and faster for edge devices in data-scarce settings? Focusing on Bambara with only 32h of speech-to-text data, we propose a new pruning recipe. Instead of vocabulary pruning, whic…

The impact of plasma turbulence on atomic reaction rates in detached divertors
Konrad Eder, Wladimir Zholobenko, Andreas Stegmeir, Kaiyu Zhang, Frank Jenko
https://arxiv.org/abs/2510.09579

The impact of plasma turbulence on atomic reaction rates in detached divertors
Numerical plasma models of the edge and scrape-off layer rely on reaction rates to describe key atomic processes such as ionization, recombination, charge-exchange, and line radiation. These rates depend non-linearly on local density and temperature, and are thus sensitive to turbulent fluctuations. We present an analysis of atomic rates and their dependence on turbulent fluctuations, obtained from GRILLIX edge turbulence simulations in detached divertor conditions. Ionization, recombination an…

Based on Deep Neural Networks: A Machine Learning-Assisted Channel Estimation Method for MIMO Systems
Haoran He
https://arxiv.org/abs/2510.11891 https://ar…

Based on Deep Neural Networks: A Machine Learning-Assisted Channel Estimation Method for MIMO Systems
This paper proposes a machine learning-assisted channel estimation approach for massive MIMO systems, leveraging DNNs to outperform traditional LS and MMSE methods. In 5G and beyond, accurate channel estimation mitigates pilot contamination and high mobility issues that harm system reliability. The proposed DNN architecture includes multi-layer perceptrons with ReLU activation, 3 hidden layers (256, 128, 64 neurons respectively), uses Adam optimizer (learning rate 1e-4) and MSE loss function. I…

"Hole in Antarctic ozone layer shrinks to smallest since 2019, scientists say"
#Antarctic #OzoneLayer

Spinon Mediation of Witness-Spin Dynamics and Ground State in Herbertsmithite
Hiroto Takahashi, Jack Murphy, Mitikorn Wood-Thanan, Pascal Puphal, Miguel Angel Sanchez-Martinez, Fabian Jerzembeck, Chun-Chih Hsu, Jonathan Ward, Masahiko Isobe, Yosuke Matsumoto, Hidenori Takagi, Stephen J. Blundell, Michael R. Norman, Felix Flicker, J. C. S\'eamus Davis
https:…

Spinon Mediation of Witness-Spin Dynamics and Ground State in Herbertsmithite
The kagome lattice of spin-1/2 Cu atoms in herbertsmithite (ZnCu3(OH)6Cl2) may sustain a quantum spin liquid (QSL) state with spinon quasiparticles. Each kagome plane is separated from its homologues by a layer of spinless Zn atoms. Providentially, however, some spin-1/2 Cu atoms substitute randomly onto these inter-kagome Zn sites. We reconceptualize these 'impurity' atoms as quantum 'witness-spins', an exceptional new interrogative of the conjectured Z2-gauge-symmetric QSL state. Thus we intr…

Tigris, which is building a network of localized data storage centers that it claims can meet the compute needs of AI workloads, raised a $25M Series A (Rebecca Bellan/TechCrunch)
https://techcrunch.com/2025/10/09/this-distributed-data-stor…

This distributed data storage startup wants to take on Big Cloud | TechCrunch
Tigris has raised $25 million to expand its growing network of localized data storage centers — the storage layer for decentralized computing infrastructure.

paris_transportation: Paris Transportation (2016)
A multiplex transportation network of Ile-de-France, jointly representing the road network, train network, and subway network. Nodes are intersections, rail stations or metro stations. Different modes of transportation are given by different layers. Within-layer edges represent direct transference by a particular mode of transportation, and between-layer edges represent the conjunction of those layers at that node.
This network ha…

Intelligent backpropagated neural networks application on Couette-Poiseuille flow of variable viscosity in a composite porous channel filled with an anisotropic porous layer
Timir Karmakar, Amrita Mandal
https://arxiv.org/abs/2510.08745

Intelligent backpropagated neural networks application on Couette-Poiseuille flow of variable viscosity in a composite porous channel filled with an anisotropic porous layer
This study examines Couette-Poiseuille flow of variable viscosity within a channel that is partially filled with a porous medium. To enhance its practical relevance, we assume that the porous medium is anisotropic with permeability varying in all directions, making it a positive semidefinite matrix in the momentum equation. We assume the Navier-Stokes equations govern the flow in the free flow region, while the Brinkman-Forchheimer-extended Darcy's equation governs the flow inside the porous me…

My take for #silentSunday . Seen yesterday from the summit that I was #hiking to. I had to zoom in quite a bit to capture this distant scene.
Whenever I see a trailrunner passing me with his ultralight / small backpack, I'm wondering why I'm carrying that heavy gear up those mounta…

Characterizing 5G User Throughput via Uncertainty Modeling and Crowdsourced Measurements
Javier Albert-Smet, Zoraida Frias, Luis Mendo, Sergio Melones, Eduardo Yraola
https://arxiv.org/abs/2510.09239

Characterizing 5G User Throughput via Uncertainty Modeling and Crowdsourced Measurements
Characterizing application-layer user throughput in next-generation networks is increasingly challenging as the higher capacity of the 5G Radio Access Network (RAN) shifts connectivity bottlenecks towards deeper parts of the network. Traditional methods, such as drive tests and operator equipment counters, are costly, limited, or fail to capture end-to-end (E2E) Quality of Service (QoS) and its variability. In this work, we leverage large-scale crowdsourced measurements-including E2E, radio, co…

Co-TAP: Three-Layer Agent Interaction Protocol Technical Report
Shunyu An, Miao Wang, Yongchao Li, Dong Wan, Lina Wang, Ling Qin, Liqin Gao, Congyao Fan, Zhiyong Mao, Jiange Pu, Wenji Xia, Dong Zhao, Rui Hu, Ji Lu, Guiyue Zhou, Baoyu Tang, Yanqin Gao, Yongsheng Du, Daigang Xu, Lingjun Huang, Baoli Wang, Xiwen Zhang, Luyao Wang, Shilong Liu
https://

Co-TAP: Three-Layer Agent Interaction Protocol Technical Report
This paper proposes Co-TAP (T: Triple, A: Agent, P: Protocol), a three-layer agent interaction protocol designed to address the challenges faced by multi-agent systems across the three core dimensions of Interoperability, Interaction and Collaboration, and Knowledge Sharing. We have designed and proposed a layered solution composed of three core protocols: the Human-Agent Interaction Protocol (HAI), the Unified Agent Protocol (UAP), and the Memory-Extraction-Knowledge Protocol (MEK). HAI focuse…

PIC12F683 full die image after wet chemical delayer to (approximately) polysilicon layer. There's a few areas in the right and bottom right that are overetched to substrate and a few spots in the analog and power rings that still have some M1 and even M2 remaining.
These are both full resolution scans with 20x/0.42 and 100x/0.90 objectives.
It's interesting to note how some features (most notably the silicon substrate in the overetched areas) are *more* visible in the low…

Similarweb: generative AI platforms were averaging ~7B monthly web visits as of September 2025, up 76% YoY, rivaling the traffic of major social networks (Matthias Bastian/The Decoder)
https://the-decoder.com/chatbots-are-now-riva…

Chatbots are now rivaling social networks as a core layer of internet infrastructure
New data from Similarweb reveals massive spikes in traffic and app downloads, while the user base expands to include older generations.

Single layer tiny Co$^4$ outpaces GPT-2 and GPT-BERT
Noor Ul Zain, Mohsin Raza, Ahsan Adeel
https://arxiv.org/abs/2510.08404 https://arxiv.org/pdf/2510.084…

Single layer tiny Co$^4$ outpaces GPT-2 and GPT-BERT
We show that a tiny Co$^4$ machine(Adeel,2025) with a single layer, two heads, and 8M parameters, operating at an approximate cost of $O(N)$ (where $N$ is the number of input tokens), outpaces the BabyLM Challenge baselines GPT-2 (124M, 12 layers, $O(N^2))$ and GPT-BERT (30M, 12 layers, $O(N^2))$ in just two epochs, while both are trained for ten. Co$^4$ achieves orders-of-magnitude greater training efficiency on 10M tokens, demonstrating highly sample efficient pretraining. Using the BabyLM ch…

Plan C turned out to be a nice day in the mountains. But Woah was there a chill wind. Good that I was prepared.
#hiking #wandern #mountains

ELMUR: External Layer Memory with Update/Rewrite for Long-Horizon RL
Egor Cherepanov, Alexey K. Kovalev, Aleksandr I. Panov
https://arxiv.org/abs/2510.07151 https://

ELMUR: External Layer Memory with Update/Rewrite for Long-Horizon RL
Real-world robotic agents must act under partial observability and long horizons, where key cues may appear long before they affect decision making. However, most modern approaches rely solely on instantaneous information, without incorporating insights from the past. Standard recurrent or transformer models struggle with retaining and leveraging long-term dependencies: context windows truncate history, while naive memory extensions fail under scale and sparsity. We propose ELMUR (External Laye…

Guided Star-Shaped Masked Diffusion
Viacheslav Meshchaninov, Egor Shibaev, Artem Makoian, Ivan Klimov, Danil Sheshenya, Andrei Malinin, Nikita Balagansky, Daniil Gavrilov, Aibek Alanov, Dmitry Vetrov
https://arxiv.org/abs/2510.08369

Guided Star-Shaped Masked Diffusion
The performance of pre-trained masked diffusion models is often constrained by their sampling procedure, which makes decisions irreversible and struggles in low-step generation regimes. We introduce a novel sampling algorithm that works with pre-trained models and, after a lightweight fine-tuning of a single layer, significantly improves sample quality and efficiency. Our method reformulates the generation process using a star-shaped paradigm, which inherently allows for error correction. To ma…

The Geometry of Truth: Layer-wise Semantic Dynamics for Hallucination Detection in Large Language Models
Amir Hameed Mir
https://arxiv.org/abs/2510.04933 https://

The Geometry of Truth: Layer-wise Semantic Dynamics for Hallucination Detection in Large Language Models
Large Language Models (LLMs) often produce fluent yet factually incorrect statements-a phenomenon known as hallucination-posing serious risks in high-stakes domains. We present Layer-wise Semantic Dynamics (LSD), a geometric framework for hallucination detection that analyzes the evolution of hidden-state semantics across transformer layers. Unlike prior methods that rely on multiple sampling passes or external verification sources, LSD operates intrinsically within the model's representational…

Learning What's Missing: Attention Dispersion and EMA Stabilization in Length Generalization
P\'al Zs\'amboki, Benjamin Levi, David Ansel Josef Smith, Mitansh Kagalwala, Arlington Kell, Samuel Liechty, Cong Wang
https://arxiv.org/abs/2510.08341

Learning What's Missing: Attention Dispersion and EMA Stabilization in Length Generalization
We study length generalization in transformers through the set complement task, where a model must predict a uniform distribution over tokens absent from an input sequence -- an ability central to board-game style reasoning. Our main theoretical result establishes two statements. First, we prove tight bounds on embedding and value dimensions for single-layer attention-only transformers. Second, we show that if such a model achieves balanced logit displacement at lengths 1 and 2, then it must ge…

Native Hybrid Attention for Efficient Sequence Modeling
Jusen Du, Jiaxi Hu, Tao Zhang, Weigao Sun, Yu Cheng
https://arxiv.org/abs/2510.07019 https://arxiv.…

Native Hybrid Attention for Efficient Sequence Modeling
Transformers excel at sequence modeling but face quadratic complexity, while linear attention offers improved efficiency but often compromises recall accuracy over long contexts. In this work, we introduce Native Hybrid Attention (NHA), a novel hybrid architecture of linear and full attention that integrates both intra \& inter-layer hybridization into a unified layer design. NHA maintains long-term context in key-value slots updated by a linear RNN, and augments them with short-term tokens fro…