Tootfinder

Bring the Apple, Not the Sofa: Impact of Irrelevant Context in Embodied AI Commands on VLA Models
Daria Pugacheva, Andrey Moskalenko, Denis Shepelev, Andrey Kuznetsov, Vlad Shakhuro, Elena Tutubalina
https://arxiv.org/abs/2510.07067

Bring the Apple, Not the Sofa: Impact of Irrelevant Context in Embodied AI Commands on VLA Models
Vision Language Action (VLA) models are widely used in Embodied AI, enabling robots to interpret and execute language instructions. However, their robustness to natural language variability in real-world scenarios has not been thoroughly investigated. In this work, we present a novel systematic study of the robustness of state-of-the-art VLA models under linguistic perturbations. Specifically, we evaluate model performance under two types of instruction noise: (1) human-generated paraphrasing a…

Calibration of Quantum Devices via Robust Statistical Methods
Alexandra Ram\^oa, Raffaele Santagati, Nathan Wiebe
https://arxiv.org/abs/2507.06941 https://…

Calibration of Quantum Devices via Robust Statistical Methods
Bayesian inference is a widely used technique for real-time characterization of quantum systems. It excels in experimental characterization in the low data regime, and when the measurements have degrees of freedom. A decisive factor for its performance is the numerical representation of the Bayesian probability distributions. In this work, we explore advanced statistical methods for this purpose, and numerically analyze their performance against the state-of-the-art in quantum parameter learnin…

Reverse Browser: Vector-Image-to-Code Generator
Zoltan Toth-Czifra
https://arxiv.org/abs/2509.05394 https://arxiv.org/pdf/2509.05394

Reverse Browser: Vector-Image-to-Code Generator
Automating the conversion of user interface design into code (image-to-code or image-to-UI) is an active area of software engineering research. However, the state-of-the-art solutions do not achieve high fidelity to the original design, as evidenced by benchmarks. In this work, I approach the problem differently: I use vector images instead of bitmaps as model input. I create several large datasets for training machine learning models. I evaluate the available array of Image Quality Assessment …

NEW ART ALERT!
Hey Garden State Social! Thanks to everyone on the contributing to the cost of running the server I have commissioned more art for the server.
This new piece is by @… and be sure to boost and give them some huge thanks for this work!
This will be our new server header image and can be found on our about page here:

Learning Global Representation from Queries for Vectorized HD Map Construction
Shoumeng Qiu, Xinrun Li, Yang Long, Xiangyang Xue, Varun Ojha, Jian Pu
https://arxiv.org/abs/2510.06969

Learning Global Representation from Queries for Vectorized HD Map Construction
The online construction of vectorized high-definition (HD) maps is a cornerstone of modern autonomous driving systems. State-of-the-art approaches, particularly those based on the DETR framework, formulate this as an instance detection problem. However, their reliance on independent, learnable object queries results in a predominantly local query perspective, neglecting the inherent global representation within HD maps. In this work, we propose \textbf{MapGR} (\textbf{G}lobal \textbf{R}epresent…

Among college graduates ages 22-27, computer science and computer engineering majors face unemployment rates of 6.1% and 7.5%, respectively, over double the unemployment rate among recent biology and art history graduates, which is just 3%.
Goodbye, $165,000 Tech Jobs. Student Coders Seek Work at Chipotle
https://www.…

Computer Science Grads Struggle to Find Jobs in the A.I. Age
As companies like Amazon and Microsoft lay off workers and embrace A.I. coding tools, computer science graduates say they’re struggling to land tech jobs.

FASL-Seg: Anatomy and Tool Segmentation of Surgical Scenes
Muraam Abdel-Ghani, Mahmoud Ali, Mohamed Ali, Fatmaelzahraa Ahmed, Mohamed Arsalan, Abdulaziz Al-Ali, Shidin Balakrishnan
https://arxiv.org/abs/2509.06159

FASL-Seg: Anatomy and Tool Segmentation of Surgical Scenes
The growing popularity of robotic minimally invasive surgeries has made deep learning-based surgical training a key area of research. A thorough understanding of the surgical scene components is crucial, which semantic segmentation models can help achieve. However, most existing work focuses on surgical tools and overlooks anatomical objects. Additionally, current state-of-the-art (SOTA) models struggle to balance capturing high-level contextual features and low-level edge features. We propose …

Greener Deep Reinforcement Learning: Analysis of Energy and Carbon Efficiency Across Atari Benchmarks
Jason Gardner, Ayan Dutta, Swapnoneel Roy, O. Patrick Kreidl, Ladislau Boloni
https://arxiv.org/abs/2509.05273

Greener Deep Reinforcement Learning: Analysis of Energy and Carbon Efficiency Across Atari Benchmarks
The growing computational demands of deep reinforcement learning (DRL) have raised concerns about the environmental and economic costs of training large-scale models. While algorithmic efficiency in terms of learning performance has been extensively studied, the energy requirements, greenhouse gas emissions, and monetary costs of DRL algorithms remain largely unexplored. In this work, we present a systematic benchmarking study of the energy consumption of seven state-of-the-art DRL algorithms, …

"How a Teaching-Focused Institution Approaches AI Literacy"
https://katinamagazine.org/content/article/future-of-work/2025/ai-literacy-at-a-teaching-focused-institution
"At the Institute of Inner-City …

End-to-End Efficiency in Keyword Spotting: A System-Level Approach for Embedded Microcontrollers
Pietro Bartoli, Tommaso Bondini, Christian Veronesi, Andrea Giudici, Niccol\`o Antonello, Franco Zappa
https://arxiv.org/abs/2509.07051

End-to-End Efficiency in Keyword Spotting: A System-Level Approach for Embedded Microcontrollers
Keyword spotting (KWS) is a key enabling technology for hands-free interaction in embedded and IoT devices, where stringent memory and energy constraints challenge the deployment of AI-enabeld devices. In this work, we systematically evaluate and compare several state-of-the-art lightweight neural network architectures, including DS-CNN, LiCoNet, and TENet, alongside our proposed Typman-KWS (TKWS) architecture built upon MobileNet, specifically designed for efficient KWS on microcontroller unit…

Reusable Surrogate Models for Distillation Columns
Martin Bubel, Tobias Seidel, Michael Bortz
https://arxiv.org/abs/2509.06638 https://arxiv.org/pdf/2509.0…

Reusable Surrogate Models for Distillation Columns
Surrogate modeling is a powerful methodology in chemical process engineering, frequently employed to accelerate optimization tasks where traditional flowsheet simulators are computationally prohibitive. However, the state-of-the-art is dominated by surrogate models trained for a narrow range of fixed chemical systems and operating conditions, limiting their reusability. This work introduces a paradigm shift towards reusable surrogates by developing a single model for distillation columns that g…

Exploring System Adaptations For Minimum Latency Real-Time Piano Transcription
Patricia Hu, Silvan David Peter, Jan Schl\"uter, Gerhard Widmer
https://arxiv.org/abs/2509.07586

Exploring System Adaptations For Minimum Latency Real-Time Piano Transcription
Advances in neural network design and the availability of large-scale labeled datasets have driven major improvements in piano transcription. Existing approaches target either offline applications, with no restrictions on computational demands, or online transcription, with delays of 128-320 ms. However, most real-time musical applications require latencies below 30 ms. In this work, we investigate whether and how the current state-of-the-art online transcription model can be adapted for real-t…

Revisiting Node Affinity Prediction in Temporal Graphs
Krishna Sri Ipsit Mantri, Or Feldman, Moshe Eliasof, Chaim Baskin
https://arxiv.org/abs/2510.06940 https://

Revisiting Node Affinity Prediction in Temporal Graphs
Node affinity prediction is a common task that is widely used in temporal graph learning with applications in social and financial networks, recommender systems, and more. Recent works have addressed this task by adapting state-of-the-art dynamic link property prediction models to node affinity prediction. However, simple heuristics, such as Persistent Forecast or Moving Average, outperform these models. In this work, we analyze the challenges in training current Temporal Graph Neural Networks …

Distribution estimation via Flow Matching with Lipschitz guarantees
Lea Kunkel
https://arxiv.org/abs/2509.02337 https://arxiv.org/pdf/2509.02337

Distribution estimation via Flow Matching with Lipschitz guarantees
Flow Matching, a promising approach in generative modeling, has recently gained popularity. Relying on ordinary differential equations, it offers a simple and flexible alternative to diffusion models, which are currently the state-of-the-art. Despite its empirical success, the mathematical understanding of its statistical power so far is very limited. This is largely due to the sensitivity of theoretical bounds to the Lipschitz constant of the vector field which drives the ODE. In this work, we…

👋 Hello! I’m Łukasz. Thirty-something, living in Warsaw, Poland. On fedi since 2021 or so. Right now, I'm merging my two accounts into one and giving this profile a fresh new start.
I live a few parallel lives.
Most of the time I’m a web developer. I work with TypeScript, React and all that jazz. But I still enjoy static site generators and good ol’ HTML/CSS.
In another life, I’m a documentary photographer and photojournalist. I’m interested in activism, art and culture, politics, and pretty much anything interesting happening on city streets. A big part of my photography journey involves documenting protests and demonstrations in Poland, especially those related to human rights and social issues. I’m also drawn to street art, particularly when it carries strong messages.
I occasionally blog and translate. Many years ago I wrote two books about WordPress. However, these days I’m more likely to go outside and touch grass than spend even more time in front of a screen.
Outside of that, expect me to post about mental health and politics. I promise to use CWs when appropriate.
I post in both English and Polish, depending on the topic.
Aaaand that’s it. Nice to meet you! ☺️
#introduction

Artists once worried about people downloading images. Today, AI has them so rattled that hackers are extorting them by threatening to submit their work to training datasets.
https://it.slashdot.org/story/25/09/02

Faster Distributed $\Delta$-Coloring via a Reduction to MIS
Yann Bourreau, Sebastian Brandt, Alexandre Nolin
https://arxiv.org/abs/2508.01762 https://arxiv…

Faster Distributed $Δ$-Coloring via a Reduction to MIS
Recent improvements on the deterministic complexities of fundamental graph problems in the LOCAL model of distributed computing have yielded state-of-the-art upper bounds of $\tilde{O}(\log^{5/3} n)$ rounds for maximal independent set (MIS) and $(Δ+ 1)$-coloring [Ghaffari, Grunau, FOCS'24] and $\tilde{O}(\log^{19/9} n)$ rounds for the more restrictive $Δ$-coloring problem [Ghaffari, Kuhn, FOCS'21; Ghaffari, Grunau, FOCS'24; Bourreau, Brandt, Nolin, STOC'25]. In our work, we show that $Δ$-col…

'Archivist fired by Trump launches a national effort to strengthen democracy': 'On Constitution Day, Shogan launched a national bipartisan effort called "More Perfect" to work on strengthening democracy' https://www.pbs.org/newsho…

Archivist fired by Trump launches a national effort to strengthen democracy
Colleen Shogan made history when she became the first woman to serve as Archivist of the United States in 2023. In February, President Trump fired her with no reason given. On Constitution Day, Shogan launched a national bipartisan effort called "More Perfect" to work on strengthening democracy. Amna Nawaz reports for our series, Art in Action, and our arts and culture coverage, CANVAS.

On my usual runs I pass this piece which is my favorite work of street art in Berlin.
Now there's construction on the parking lot next to it and this artwork might soon disappear :(

Hate in Plain Sight: On the Risks of Moderating AI-Generated Hateful Illusions
Yiting Qu, Ziqing Yang, Yihan Ma, Michael Backes, Savvas Zannettou, Yang Zhang
https://arxiv.org/abs/2507.22617

Hate in Plain Sight: On the Risks of Moderating AI-Generated Hateful Illusions
Recent advances in text-to-image diffusion models have enabled the creation of a new form of digital art: optical illusions--visual tricks that create different perceptions of reality. However, adversaries may misuse such techniques to generate hateful illusions, which embed specific hate messages into harmless scenes and disseminate them across web communities. In this work, we take the first step toward investigating the risks of scalable hateful illusion generation and the potential for bypa…

Every three years, we work with curator Didi Dunphy to put together an exhibition of work by Georgia artists focused on a theme that then travels the state. The newest one is "Seams to Be: New Approaches to Textile Techniques," which opens at the Lyndon House Arts Center October 2 before moving on to the LaGrange Art Museum, the Morris Museum of Art in Augusta, the Museum of Arts and Sciences in Macon, the Albany Museum Of Art and the Columbus Museum.

Variation-matching sensitivity-based virtual fields for hyperelastic material model calibration
Denislav P. Nikolov, Zhiren Zhu, Jonathan B. Estrada
https://arxiv.org/abs/2509.03363

Variation-matching sensitivity-based virtual fields for hyperelastic material model calibration
Accurate identification of nonlinear material parameters from three-dimensional full-field deformation data remains a challenge in experimental mechanics. The virtual fields method (VFM) provides a powerful, computationally efficient approach for material model calibration, however, its success depends critically on the choice of virtual fields and the informativeness of available kinematic data. In this work, we advance the state-of-the-art discrete formulation of the sensitivity-based virtual…

When I do web dev, I don’t just work in the DOM, I put on a pencil moustache and work in the MANDOM.
Pat didn’t know it when he made the shirt, but it’s a guys’ web dev shirt: https://www.teepublic.com/t-shirt/78574049-mandom
(Or a shifter smut lady book club tank, as it t…

Disc-Cover Complexity Trends in Music Illustrations from Sinatra to Swift
Nicolas Fracaro, Stefano Cecconello, Mauro Conti, Niccol\`o Di Marco, Alessandro Galeazzi
https://arxiv.org/abs/2510.00990

Disc-Cover Complexity Trends in Music Illustrations from Sinatra to Swift
The study of art evolution has provided valuable insights into societal change, often revealing long-term patterns of simplification and transformation. Album covers represent a distinctive yet understudied form of visual art that has both shaped and been shaped by cultural, technological, and commercial dynamics over the past century. As highly visible artifacts at the intersection of art and commerce, they offer a unique lens through which to study cultural evolution. In this work, we examine…

Speech LLMs in Low-Resource Scenarios: Data Volume Requirements and the Impact of Pretraining on High-Resource Languages
Seraphina Fong, Marco Matassoni, Alessio Brutti
https://arxiv.org/abs/2508.05149

Speech LLMs in Low-Resource Scenarios: Data Volume Requirements and the Impact of Pretraining on High-Resource Languages
Large language models (LLMs) have demonstrated potential in handling spoken inputs for high-resource languages, reaching state-of-the-art performance in various tasks. However, their applicability is still less explored in low-resource settings. This work investigates the use of Speech LLMs for low-resource Automatic Speech Recognition using the SLAM-ASR framework, where a trainable lightweight projector connects a speech encoder and a LLM. Firstly, we assess training data volume requirements t…

A Lazy Resynthesis Approach for Simultaneous T Gate and Two-Qubit Gate Optimization of Quantum Circuits
Mu-Te Lau (Ric), Hsiang-Chun Yang (Ric), Hsin-Yu Chen (Ric), Chung-Yang (Ric), Huang
https://arxiv.org/abs/2508.04092

A Lazy Resynthesis Approach for Simultaneous T Gate and Two-Qubit Gate Optimization of Quantum Circuits
State-of-the-art quantum circuit optimization (QCO) algorithms for T-count reduction often lead to a substantial increase in two-qubit gate count (2Q-count) -- a drawback that existing 2Q-count optimization techniques struggle to address effectively. In this work, we propose a novel lazy resynthesis approach for modern tableau-based QCO flows that significantly mitigates the 2Q-gate surges commonly introduced during T-count optimization in Clifford+T circuits. Experimental results show that our…

A year or so after I finished this book, I realized that the Jones character in 'Liar, Cheater, Sinner, Saint' was Kevin from the Navy Trent Series. And, of course, Kevin wouldn't star in his own book. He prefers to work in the background. Art in this reel is by @empyranni_illustrates, cover art for the book is by @gargoyle.pastures. https://www.

"The Art of Solving Impossible Problems" @ Katina Magazine: https://katinamagazine.org/content/article/future-of-work/2025/the-art-of-solving-impossible-problems
"Sometimes the best way to solve a problem is by ad…

Higgs production via vector-boson fusion at the LHC
Gaetano Barone, Jiayi Chen, Stephane Cooperstein, Nikita Dolganov, Silvia Ferrario Ravasio, Yacine Haddad, Stefan H\"oche, Barbara J\"ager, Alexander Karlberg, Alexander M\"uck, Mathieu Pellen, Christian T. Preuss, Daniel Reichelt, Simon Reinhardt, Marco Zaro
https://arxiv.…

Higgs production via vector-boson fusion at the LHC
In this article, we summarise the recent experimental measurements and theoretical work on Higgs boson production via vector-boson fusion at the LHC. Along with this, we provide state-of-the-art predictions at fixed order as well as with parton-shower corrections within the Standard Model at 13.6 TeV. The results are presented in the form of multi-differential distributions as well as in the Simplified Template Cross Section bins. All materials and outputs of this study are available on public …

Product-Quantised Image Representation for High-Quality Image Synthesis
Denis Zavadski, Nikita Philip Tatsch, Carsten Rother
https://arxiv.org/abs/2510.03191 https://

Product-Quantised Image Representation for High-Quality Image Synthesis
Product quantisation (PQ) is a classical method for scalable vector encoding, yet it has seen limited usage for latent representations in high-fidelity image generation. In this work, we introduce PQGAN, a quantised image autoencoder that integrates PQ into the well-known vector quantisation (VQ) framework of VQGAN. PQGAN achieves a noticeable improvement over state-of-the-art methods in terms of reconstruction performance, including both quantisation methods and their continuous counterparts. …

Day 6: Kamome Shirahama
Before I wander much father afield, I'd be remiss not to include at least one Mangaka (I've got 8 on my planning list; if you think Manga is pushing it just wait until you see what the next few days have in store).
I'm currently following "Witch Hat Atelier," and it's absolutely amazing in several dimensions: first class world-building, deep philosophical themes, nuanced diverse cast, tightly-constructed interwoven plots, deep mysteries that keep everything churning and show up in unexpected places, absolutely stellar art both in terms of in-panel depictions and page layouts (some are Watchmen-quality), especially if you are sartorially inclined, and general kindness of its core messages. This is a series I wish every programmer would read, because it includes excellent advice about software design in multiple ways (did I mention there's an intricate and logical magic system within which the main character innovates in legible-to-the-reader-as-innovation ways?). Also, I bet I would have enjoyed this just a much as a 10-year-old as I'm enjoying it in my 30's, which is something that takes well-honed skill to pull off.
Shirahama is a master of her craft, and I'm honestly kinda surprised to see Witch Hat is only her second series. Definitely thinking how I can get my hands on her earlier work in English.
#20AuthorsNoMen

Evaluation of Black-Box XAI Approaches for Predictors of Values of Boolean Formulae
Stav Armoni-Friedmann, Hana Chockler, David A. Kelly
https://arxiv.org/abs/2509.09982 https:/…

Evaluation of Black-Box XAI Approaches for Predictors of Values of Boolean Formulae
Evaluating explainable AI (XAI) approaches is a challenging task in general, due to the subjectivity of explanations. In this paper, we focus on tabular data and the specific use case of AI models predicting the values of Boolean functions. We extend the previous work in this domain by proposing a formal and precise measure of importance of variables based on actual causality, and we evaluate state-of-the-art XAI tools against this measure. We also present a novel XAI tool B-ReX, based on the e…

1/2 Thanks to @… for this interesting article. It speaks to me. :)
I’ve been weather blogging @… since 2005. It is interesting how it has changed, and how I have changed.
My website used to be just data from the (expensive) station I bought when I moved back to Port Alberni. It was a hobby and a side project to practice web/coding skills I use at work. My focus was on creating useful data for people that was more local/relevant than the official EC station outside of the city.
Then I put up a webcam and learned how to make timelapses. This got the attention of local media… because pictures. :)
Then I added a blog and started to write about the weather almost daily. This was before Facebook. There was a popular local online forum where I would post things. The media would also follow my website and they started to call me when there was extreme weather (usually very hot or very wet/stormy).
Then Facebook started to get big and I made a page that eventually had a few thousand followers. I would blog often. Lots of traffic from Facebook… this was 2010 and on. I blogged about climate and weather pretty equally.
Like anyone in Port Alberni, I was/am obsessed with the Martin Mars and got wrapped up in that issue along with others which combined with the weather following probably gave me just enough exposure to have me elected as a councillor in 2014.
I continued through that 4 years, blogging often in addition to councillor duties and work, heavily on facebook, then it all went sideways on my own poor judgement (go ahead and google it, it’s ok :)) and I was not reelected, but Facebook by 2018 had also changed. Cambridge Analytica, etc.
….Continued…
https://www.theglobeandmail.com/canada/article-weather-apps-data-wildfires-storms-preparation-obsession-social-media/

Author Adelaide Faith on the fear and joy of total artistic freedom
https://thecreativeindependent.com/people/author-adelaide-fai…

Author Adelaide Faith on the fear and joy of total artistic freedom – The Creative Independent
Author Adelaide Faith discusses managing work, parenting and creative time, sticking to a schedule, and choosing a day job that makes a difference in the world.

Optimal Impulsive Control of Cislunar Relative Motion using Reachable Set Theory
Matthew Hunter, Walter J. Manuel, Simone D'Amico
https://arxiv.org/abs/2507.21425 https://…

Optimal Impulsive Control of Cislunar Relative Motion using Reachable Set Theory
This work presents the first application of the state-of-the-art Koenig-D'Amico reachable set theory solver to cislunar, chaotic relative motion in the Circular-Restricted Three-Body Problem (CR3BP). The relative motion dynamics of two spacecraft, a chief and a deputy, in the CR3BP are formulated as a Linear Time-Variant (LTV) system, allowing the solver to find an optimal impulsive control maneuver plan. This methodology demonstrates robust and accurate control performance for both small and l…

The Cat No Banana

🎨by Marc P.I.

11Plus-Bench: Demystifying Multimodal LLM Spatial Reasoning with Cognitive-Inspired Analysis
Chengzu Li, Wenshan Wu, Huanyu Zhang, Qingtao Li, Zeyu Gao, Yan Xia, Jos\'e Hern\'andez-Orallo, Ivan Vuli\'c, Furu Wei
https://arxiv.org/abs/2508.20068

11Plus-Bench: Demystifying Multimodal LLM Spatial Reasoning with Cognitive-Inspired Analysis
For human cognitive process, spatial reasoning and perception are closely entangled, yet the nature of this interplay remains underexplored in the evaluation of multimodal large language models (MLLMs). While recent MLLM advancements show impressive performance on reasoning, their capacity for human-like spatial cognition remains an open question. In this work, we introduce a systematic evaluation framework to assess the spatial reasoning abilities of state-of-the-art MLLMs relative to human pe…

Exactly 20 years ago I published a blog post (which I've edited to adapt to my most recent incarnation of a blog) , with a comparison of tools and frameworks for quite a few technology stacks: Java, .NET, PHP, etc.
This was right at the very early stages of the Cloud, there was no "Big Data", there was certainly no AI, there was no iPhone or Android, Mac OS X releases were still big cats, Ubuntu had just appeared, and social media was just Orkut or MySpace. Enjoy.

State of the Art
This article is a copy of a research work I did today, to draw a map of today’s development technologies.

The Rhythm of Tai Chi: Revitalizing Cultural Heritage in Virtual Reality through Interactive Visuals
Xianghan Wang
https://arxiv.org/abs/2508.16605 https://

The Rhythm of Tai Chi: Revitalizing Cultural Heritage in Virtual Reality through Interactive Visuals
The Rhythm of Tai Chi reinterprets the ancient Chinese martial art as a dynamic, interactive virtual reality (VR) experience. By leveraging computer vision and multimedia technologies, the project transforms Tai Chi's philosophy and movements into an immersive digital form. Real-time motion tracking captures user gestures, while visual feedback systems simulate the flow of Qi, enabling an intuitive and engaging practice environment. Beyond technological innovation, this work bridges traditional…

Surface diffusion: The intermediate scattering function seen as a characteristic function of probability theory
E. E. Torres-Miyares, S. Miret-Art\'es
https://arxiv.org/abs/2509.15240

Surface diffusion: The intermediate scattering function seen as a characteristic function of probability theory
In surface diffusion, one of the key observables is the so-called intermediate scattering function which is measured directly from the surface technique called Helium spin echo. In this work, we show that this function can be seen as a characteristic function of probability theory. From the characteristic function, the moments and cumulants of the probability distribution function of the position of the adsorbate are straightforward obtained in an analytical way; in particular, the second order…

Observational Constraints on Evolution of Dust Disc Properties in Upper Scorpius
Paola Pinilla, Anibal Sierra, Nicolas T. Kurtovic, Rossella Anania, Sean Andrews, John Carpenter, Osmar Guerra-Alvarado, Feng Long, Sebastian Marino, Miguel Vioque, Ke Zhang
https://arxiv.org/abs/2509.19442

Observational Constraints on Evolution of Dust Disc Properties in Upper Scorpius
Protoplanetary discs in the Upper Scorpius star-forming region are excellent laboratories to investigate late stages of planet formation. In this work, we analyse the morphology of the dust continuum emission of 121 discs from an ALMA Band 7 survey of the Upper Scorpius region. This analysis is done in the visibility plane, to measure the flux, geometry and characterise potential structures. We compare the results with state-of-the art gas and dust evolution models that include external photoev…

Hyper-spectral Unmixing algorithms for remote compositional surface mapping: a review of the state of the art
Alfredo Gimenez Zapiola, Andrea Boselli, Alessandra Menafoglio, Simone Vantini
https://arxiv.org/abs/2507.14260

Hyper-spectral Unmixing algorithms for remote compositional surface mapping: a review of the state of the art
This work concerns a detailed review of data analysis methods used for remotely sensed images of large areas of the Earth and of other solid astronomical objects. In detail, it focuses on the problem of inferring the materials that cover the surfaces captured by hyper-spectral images and estimating their abundances and spatial distributions within the region. The most successful and relevant hyper-spectral unmixing methods are reported as well as compared, as an addition to analysing the most r…

Can General-Purpose Omnimodels Compete with Specialists? A Case Study in Medical Image Segmentation
Yizhe Zhang, Qiang Chen, Tao Zhou
https://arxiv.org/abs/2509.00866 https://…

Can General-Purpose Omnimodels Compete with Specialists? A Case Study in Medical Image Segmentation
The emergence of powerful, general-purpose omnimodels capable of processing diverse data modalities has raised a critical question: can these ``jack-of-all-trades'' systems perform on par with highly specialized models in knowledge-intensive domains? This work investigates this question within the high-stakes field of medical image segmentation. We conduct a comparative study analyzing the zero-shot performance of a state-of-the-art omnimodel (Gemini 2.5 Pro, the ``Nano Banana'' model) against …

Btw. I know there hasn't been much new work on #ThingUmbrella over the past month. That's because I've been spending _a lot_ more time on my multi-faceted toolchain for note taking, asset management (images/videos, incl. auto-creation of different versions) and also aimed as generator for blogs/websites, feeds, visualizations (e.g. incl. those on the current

Open source building blocks for computational design. Est. 2006
thi.ng is a set of ~350 complementing code libraries & projects, a long-term effort devoted to Computational Design and Art in its many forms.

Fold-transversal surface code cultivation
Kaavya Sahay, Pei-Kai Tsai, Kathleen Chang, Qile Su, Thomas B. Smith, Shraddha Singh, Shruti Puri
https://arxiv.org/abs/2509.05212 http…

Fold-transversal surface code cultivation
Magic state cultivation is a state-of-the-art protocol to prepare ultra-high fidelity non-Clifford resource states for universal quantum computation. It offers a significant reduction in spacetime overhead compared to traditional magic state distillation techniques. Cultivation protocols involve measuring a transversal logical Clifford operator on an initial small-distance code and then rapidly growing to a larger-distance code. In this work, we present a new cultivation scheme in which we meas…

Robust Recursive Query Parallelism in Graph Database Management Systems
Anurag Chakraborty, Semih Saliho\u{g}lu
https://arxiv.org/abs/2508.19379 https://ar…

Robust Recursive Query Parallelism in Graph Database Management Systems
Efficient multi-core parallel processing of recursive join queries is critical for achieving good performance in graph database management systems (GDBMSs). Prior work adopts two broad approaches. First is the state of the art morsel-driven parallelism, whose vanilla application in GDBMSs parallelizes computations at the source node level. Second is to parallelize each iteration of the computation at the frontier level. We show that these approaches can be seen as part of a design space of mors…

Directed Grammar-Based Test Generation
Lukas Kirschner (Saarland University,University of Luxembourg), Ezekiel Soremekun (Singapore University of Technology,Design)
https://arxiv.org/abs/2508.01472

Directed Grammar-Based Test Generation
To effectively test complex software, it is important to generate goal-specific inputs, i.e., inputs that achieve a specific testing goal. However, most state-of-the-art test generators are not designed to target specific goals. Notably, grammar-based test generators, which (randomly) produce syntactically valid inputs via an input specification (i.e., grammar) have a low probability of achieving an arbitrary testing goal. This work addresses this challenge by proposing an automated test genera…

Adaptive Control of Heterogeneous Platoons with Guaranteed Collision Avoidance
Ashutosh Chandra Pandey, Sayan Basu Roy, Simone Baldi
https://arxiv.org/abs/2508.20493 https://

Adaptive Control of Heterogeneous Platoons with Guaranteed Collision Avoidance
This work proposes a framework for Cooperative Adaptive Cruise Control of a vehicular platoon characterized by unidirectional communication and heterogeneous parameters. In the proposed framework, the actual (heterogeneous) platoon is made to converge to a reference (homogeneous) platoon via adaptive laws designed using of set-theoretic model reference adaptive control. Yet, in contrast to the state-of-art that is based on ensuring collision avoidance on the reference platoon dynamics only, the…

"But to me the most extraordinary thing about it is that body. The entire back of the car from the front of the doors rearward including the roof is a single piece of carbon because, as Mallock puts it, ‘I hate shut lines’. It is a vast work of art yet weighs just 19kg.
‘What if you damage it?’ I ask. ‘On balance it’s probably better if you don’t,’ he says."
via:

Rain stopped play - The Intercooler
A 920bhp prototype worth millions on track day tyres on streaming wet roads. What could possibly go wrong? Andrew Frankel tries hard not to find out

Multi-robot LiDAR SLAM: a practical case study in underground tunnel environments
Federica Di Lauro, Domenico G. Sorrenti, Miguel Angel Sotelo
https://arxiv.org/abs/2507.21553 h…

Multi-robot LiDAR SLAM: a practical case study in underground tunnel environments
Multi-robot SLAM aims at localizing and building a map with multiple robots, interacting with each other. In the work described in this article, we analyze the pipeline of a decentralized LiDAR SLAM system to study the current limitations of the state of the art, and we discover a significant source of failures, i.e., that the loop detection is the source of too many false positives. We therefore develop and propose a new heuristic to overcome these limitations. The environment taken as referen…

"The Bib Record in the Age of Digital Reproduction" https://journal.radicallibrarianship.org/index.php/journal/article/view/135
"Taking both structural and thematic inspiration from Walter Benjamin’s “The Work of Art in the Age of Mechanic…

CapsBeam: Accelerating Capsule Network based Beamformer for Ultrasound Non-Steered Plane Wave Imaging on Field Programmable Gate Array
Abdul Rahoof, Vivek Chaturvedi, Mahesh Raveendranatha Panicker, Muhammad Shafique
https://arxiv.org/abs/2509.03201

CapsBeam: Accelerating Capsule Network based Beamformer for Ultrasound Non-Steered Plane Wave Imaging on Field Programmable Gate Array
In recent years, there has been a growing trend in accelerating computationally complex non-real-time beamforming algorithms in ultrasound imaging using deep learning models. However, due to the large size and complexity these state-of-the-art deep learning techniques poses significant challenges when deploying on resource-constrained edge devices. In this work, we propose a novel capsule network based beamformer called CapsBeam, designed to operate on raw radio-frequency data and provide an en…

I missed today's protest because I had client work but clients paying me for work helps fuel the efforts in put into the community.
I got a bunch of supplies recently to make more unhoused person care packages and I purchased art from marginalized people that will be made available in the Free Little Art Gallery.
We all gotta do what we can, when we can, in the ways we can.
#resist

A Bag of Tricks for Efficient Implicit Neural Point Clouds
Florian Hahlbohm, Linus Franke, Leon Overk\"amping, Paula Wespe, Susana Castillo, Martin Eisemann, Marcus Magnor
https://arxiv.org/abs/2508.19140

A Bag of Tricks for Efficient Implicit Neural Point Clouds
Implicit Neural Point Cloud (INPC) is a recent hybrid representation that combines the expressiveness of neural fields with the efficiency of point-based rendering, achieving state-of-the-art image quality in novel view synthesis. However, as with other high-quality approaches that query neural networks during rendering, the practical usability of INPC is limited by comparatively slow rendering. In this work, we present a collection of optimizations that significantly improve both the training …

A Scene is Worth a Thousand Features: Feed-Forward Camera Localization from a Collection of Image Features
Axel Barroso-Laguna, Tommaso Cavallari, Victor Adrian Prisacariu, Eric Brachmann
https://arxiv.org/abs/2510.00978

A Scene is Worth a Thousand Features: Feed-Forward Camera Localization from a Collection of Image Features
Visually localizing an image, i.e., estimating its camera pose, requires building a scene representation that serves as a visual map. The representation we choose has direct consequences towards the practicability of our system. Even when starting from mapping images with known camera poses, state-of-the-art approaches still require hours of mapping time in the worst case, and several minutes in the best. This work raises the question whether we can achieve competitive accuracy much faster. We …

Diffusion Models and the Manifold Hypothesis: Log-Domain Smoothing is Geometry Adaptive
Tyler Farghly, Peter Potaptchik, Samuel Howard, George Deligiannidis, Jakiw Pidstrigach
https://arxiv.org/abs/2510.02305

Diffusion Models and the Manifold Hypothesis: Log-Domain Smoothing is Geometry Adaptive
Diffusion models have achieved state-of-the-art performance, demonstrating remarkable generalisation capabilities across diverse domains. However, the mechanisms underpinning these strong capabilities remain only partially understood. A leading conjecture, based on the manifold hypothesis, attributes this success to their ability to adapt to low-dimensional geometric structure within the data. This work provides evidence for this conjecture, focusing on how such phenomena could result from the …

Good Morning #Canada
#HappyBirthday to Kenojuak Ashevak, one of Canada’s most famous Inuk artists. She was born in 1927 at Camp Kerrasak on southern Baffin Island and died on January 8, 2013, in Cape Dorset, Nunavut. Known primarily for her drawings as a graphic artist, she had a diverse artistic experience, making sculpture and engraving and working with textiles and also on stained glass. She is celebrated as a leading figure of modern Inuit art and one of Canada's preeminent artists and cultural icons. She was the first Inuk artist inducted into Canada's Walk of Fame (2001), was made an Officer of the Order of Canada (1967), and promoted to Companion in 1982. She received the Governor General's Award in Visual and Media Arts (2008) and the Order of Nunavut (2012). Her work, with its superb design qualities, was used for Canadian stamps, coins and banknotes, including her best-known work, The Enchanted Owl
#CanadaIsAwesome #IndigenousArt
https://youtu.be/u4a2V_jQU98?si=s80cCelwfP8tR8nV

Competing $s$-wave pairing in overdoped $t$-$J$ model
Wayne Zheng, Tao Cheng, Zheng-Yuan Yue, Fu-Chun Zhang, Wei-Qiang Chen, Zheng-Cheng Gu
https://arxiv.org/abs/2509.22473 http…

Competing $s$-wave pairing in overdoped $t$-$J$ model
The $d$-wave pairing symmetry has long been considered a defining feature of high-temperature superconductivity in cuprates. In this work, we reveal that $s$-wave pairing states exhibit variational energies comparable to the $d$-wave state in a square $t$-$J$ model, particularly at high doping levels ($δ\gtrsim 15\%$) by using the state-of-the-art tensor network simulation. This surprising result suggests that $s$-wave pairing may play an important role in the cuprate phase diagram, especially…

Gradient-based grand canonical optimization enabled by graph neural networks with fractional atomic existence
Mads-Peter Verner Christiansen, Bj{\o}rk Hammer
https://arxiv.org/abs/2507.19438

Gradient-based grand canonical optimization enabled by graph neural networks with fractional atomic existence
Machine learning interatomic potentials have become an indispensable tool for materials science, enabling the study of larger systems and longer timescales. State-of-the-art models are generally graph neural networks that employ message passing to iteratively update atomic embeddings that are ultimately used for predicting properties. In this work we extend the message passing formalism with the inclusion of a continuous variable that accounts for fractional atomic existence. This allows us to …

Supernova cooling from neutrino-devouring dark matter
Yugen Lin, Chih-Ting Lu, Ningqiang Song
https://arxiv.org/abs/2507.22124 https://arxiv.org/pdf/2507.2…

Supernova cooling from neutrino-devouring dark matter
Supernova cooling provides a powerful probe of physics beyond the Standard Model (SM), in particular for new, light states interacting feebly with SM particles. In this work, we investigate for the first time the production of fermionic dark matter (DM) via the neutrino-devouring process inside a core-collapse supernova, which contributes to the excessive cooling. By incorporating state-of-the-art supernova simulation data and the full time evolution information, we derive stringent and robust …

Source Separation for A Cappella Music
Luca A. Lanzend\"orfer, Constantin Pinkl, Florian Gr\"otschla
https://arxiv.org/abs/2509.26580 https://arx…

Source Separation for A Cappella Music
In this work, we study the task of multi-singer separation in a cappella music, where the number of active singers varies across mixtures. To address this, we use a power set-based data augmentation strategy that expands limited multi-singer datasets into exponentially more training samples. To separate singers, we introduce SepACap, an adaptation of SepReformer, a state-of-the-art speaker separation model architecture. We adapt the model with periodic activations and a composite loss function …

LiLa-Net: Lightweight Latent LiDAR Autoencoder for 3D Point Cloud Reconstruction
Mario Resino, Borja P\'erez, Jaime Godoy, Abdulla Al-Kaff, Fernando Garc\'ia
https://arxiv.org/abs/2510.02028

LiLa-Net: Lightweight Latent LiDAR Autoencoder for 3D Point Cloud Reconstruction
This work proposed a 3D autoencoder architecture, named LiLa-Net, which encodes efficient features from real traffic environments, employing only the LiDAR's point clouds. For this purpose, we have real semi-autonomous vehicle, equipped with Velodyne LiDAR. The system leverage skip connections concept to improve the performance without using extensive resources as the state-of-the-art architectures. Key changes include reducing the number of encoder layers and simplifying the skip connections, …

Towards interpretable emotion recognition: Identifying key features with machine learning
Yacouba Kaloga, Ina Kodrasi
https://arxiv.org/abs/2508.04230 https://

Towards interpretable emotion recognition: Identifying key features with machine learning
Unsupervised methods, such as wav2vec2 and HuBERT, have achieved state-of-the-art performance in audio tasks, leading to a shift away from research on interpretable features. However, the lack of interpretability in these methods limits their applicability in critical domains like medicine, where understanding feature relevance is crucial. To better understand the features of unsupervised models, it remains critical to identify the interpretable features relevant to a given task. In this work, …

I have the priviledge of taking a whole semester for my research this fall, and intend to blog / think out loud as I work my way through #Pedagogy, #Philosophy, #Art,

Random Forest Classification of MBTA Gravitational-Wave Triggers for Low-Latency Detection
Lorenzo Mobilia, Gianluca Maria Guidi
https://arxiv.org/abs/2509.12882 https://…

Random Forest Classification of MBTA Gravitational-Wave Triggers for Low-Latency Detection
In this work, we explore a possible application of a machine learning classifier for candidate events in a template-based search for gravitational-wave (GW) signals from various compact system sources. We analyze data from the O3a and O3b data acquisition campaign, during which the sensitivity of ground-based detectors is limited by real non-Gaussian noise transient. The state-of-the-art searches for such signals tipically rely on the signal-to-noise ratio (SNR) and a chi-square test to assess …

Fast Percolation Centrality Approximation with Importance Sampling
Antonio Cruciani, Leonardo Pellegrina
https://arxiv.org/abs/2509.11454 https://arxiv.org…

Fast Percolation Centrality Approximation with Importance Sampling
In this work we present PercIS, an algorithm based on Importance Sampling to approximate the percolation centrality of all the nodes of a graph. Percolation centrality is a generalization of betweenness centrality to attributed graphs, and is a useful measure to quantify the importance of the vertices in a contagious process or to diffuse information. However, it is impractical to compute it exactly on modern-sized networks. First, we highlight key limitations of state-of-the-art sampling-bas…

Art Adventures, our free summer program for camps, community centers and day cares, is chugging along. Want to learn more about it and what makes this year's program special? https://georgiamuseum.org/gmoa_blog/this-summers-art-adventures-imm…

FastVGGT: Training-Free Acceleration of Visual Geometry Transformer
You Shen, Zhipeng Zhang, Yansong Qu, Liujuan Cao
https://arxiv.org/abs/2509.02560 https://

FastVGGT: Training-Free Acceleration of Visual Geometry Transformer
Foundation models for 3D vision have recently demonstrated remarkable capabilities in 3D perception. However, scaling these models to long-sequence image inputs remains a significant challenge due to inference-time inefficiency. In this work, we present a detailed analysis of VGGT, a state-of-the-art feed-forward visual geometry model and identify its primary bottleneck. Visualization further reveals a token collapse phenomenon in the attention maps. Motivated by these findings, we explore the …

SecInfer: Preventing Prompt Injection via Inference-time Scaling
Yupei Liu, Yanting Wang, Yuqi Jia, Jinyuan Jia, Neil Zhenqiang Gong
https://arxiv.org/abs/2509.24967 https://

SecInfer: Preventing Prompt Injection via Inference-time Scaling
Prompt injection attacks pose a pervasive threat to the security of Large Language Models (LLMs). State-of-the-art prevention-based defenses typically rely on fine-tuning an LLM to enhance its security, but they achieve limited effectiveness against strong attacks. In this work, we propose \emph{SecInfer}, a novel defense against prompt injection attacks built on \emph{inference-time scaling}, an emerging paradigm that boosts LLM capability by allocating more compute resources for reasoning dur…

Benchmarking Deep Learning Convolutions on Energy-constrained CPUs
Enrique Galvez (ALSOC), Adrien Cassagne (ALSOC), Alix Munier (ALSOC), Manuel Bouyer
https://arxiv.org/abs/2509.26217

Benchmarking Deep Learning Convolutions on Energy-constrained CPUs
This work evaluates state-of-the-art convolution algorithms for CPU-based deep learning inference. While most prior studies focus on GPUs or NPUs, CPU implementations remain relatively underoptimized. We benchmark direct, GEMM-based, and Winograd convolutions across modern CPUs from ARM __ , Intel __ , AMD __ , Apple __ , and Nvidia __ , considering both latency and energy efficiency. Our results highlight the key architectural factors that govern CPU efficiency for convolution operations, prov…

Musician Ela Minus on having more power over your work
#creativity #artist

Musician Ela Minus on having more power over your work – The Creative Independent
Musician Ela Minus discusses demystifying your industry, questioning yourself and the systems around you, and the importance of teaching others

Overlap-Adaptive Regularization for Conditional Average Treatment Effect Estimation
Valentyn Melnychuk, Dennis Frauen, Jonas Schweisthal, Stefan Feuerriegel
https://arxiv.org/abs/2509.24962

Overlap-Adaptive Regularization for Conditional Average Treatment Effect Estimation
The conditional average treatment effect (CATE) is widely used in personalized medicine to inform therapeutic decisions. However, state-of-the-art methods for CATE estimation (so-called meta-learners) often perform poorly in the presence of low overlap. In this work, we introduce a new approach to tackle this issue and improve the performance of existing meta-learners in the low-overlap regions. Specifically, we introduce Overlap-Adaptive Regularization (OAR) that regularizes target models prop…

Resource-efficient universal photonic processor based on time-multiplexed hybrid architectures
Jonas Lammers, Laura Ares, Federico Pegoraro, Philip Held, Benjamin Brecht, Jan Sperling, Christine Silberhorn
https://arxiv.org/abs/2509.22521

Resource-efficient universal photonic processor based on time-multiplexed hybrid architectures
For the ever-growing field of quantum information processing, large-scale, efficient multi-port interferometers serving as photonic processors are required. In this context, the suitability of quantum walks as the interferometric base for universal computation has been theoretically proven. In this work, we bridge the gap between theoretical proposals and state-of-the-art experimental capabilities by providing the recipe for the implementation of a universal photonic processor in discrete-time …

PIPer: On-Device Environment Setup via Online Reinforcement Learning
Alexander Kovrigin, Aleksandra Eliseeva, Konstantin Grotov, Egor Bogomolov, Yaroslav Zharov
https://arxiv.org/abs/2509.25455

PIPer: On-Device Environment Setup via Online Reinforcement Learning
Environment setup-the process of configuring the system to work with a specific software project-represents a persistent challenge in Software Engineering (SE). Automated environment setup methods could assist developers by providing fully configured environments for arbitrary repositories without manual effort. This also helps SE researchers to scale execution-based benchmarks. However, recent studies reveal that even state-of-the-art Large Language Models (LLMs) achieve limited success in aut…

Simulating Image Coaddition with the Nancy Grace Roman Space Telescope. IV. Hyperparameter Optimization and Experimental Features
Kaili Cao, Christopher M. Hirata, Katherine Laliotis, Masaya Yamamoto, Emily Macbeth, M. A. Troxel
https://arxiv.org/abs/2509.18286

"What My Library Degree Taught Me About the Resilience of Libraries" by Renee Kepert: https://katinamagazine.org/content/article/future-of-work/2025/what-my-library-degree-taught-me-how-to-learn
"I [= a for…

Ensemble of Pre-Trained Models for Long-Tailed Trajectory Prediction
Divya Thuremella, Yi Yang, Simon Wanna, Lars Kunze, Daniele De Martini
https://arxiv.org/abs/2509.13914 http…

Ensemble of Pre-Trained Models for Long-Tailed Trajectory Prediction
This work explores the application of ensemble modeling to the multidimensional regression problem of trajectory prediction for vehicles in urban environments. As newer and bigger state-of-the-art prediction models for autonomous driving continue to emerge, an important open challenge is the problem of how to combine the strengths of these big models without the need for costly re-training. We show how, perhaps surprisingly, combining state-of-the-art deep learning models out-of-the-box (withou…

Author and editor Callie Collins on finding spaces to think and work creatively https://thecreativeindependent.com/peo

Author and editor Callie Collins on finding spaces to think and work creatively – The Creative Independent
Author and editor Callie Collins discusses having a stable job, the ethics of art about living people, and finding the perfect zone to create

LLM Ensemble for RAG: Role of Context Length in Zero-Shot Question Answering for BioASQ Challenge
Dima Galat, Diego Molla-Aliod
https://arxiv.org/abs/2509.08596 https://

LLM Ensemble for RAG: Role of Context Length in Zero-Shot Question Answering for BioASQ Challenge
Biomedical question answering (QA) poses significant challenges due to the need for precise interpretation of specialized knowledge drawn from a vast, complex, and rapidly evolving corpus. In this work, we explore how large language models (LLMs) can be used for information retrieval (IR), and an ensemble of zero-shot models can accomplish state-of-the-art performance on a domain-specific Yes/No QA task. Evaluating our approach on the BioASQ challenge tasks, we show that ensembles can outperfor…

Good Morning #Canada
#HappyBirthday to my eldest daughter and, coincidentally, one of my favourite painters, Tom Thomson. Born in 1877 and raised in rural Ontario, Thomson displayed no immediate artistic talent and worked several jobs in sign making and graphic design. He didn't start painting seriously until he was almost 30, and during his short career, he produced roughly 400 oil sketches on small wood panels and approximately 50 larger works on canvas. His art consists almost entirely of landscapes, depicting trees, skies, lakes, and rivers. An excellent collection of his work can be seen at the McMichael Canadian Art Collection in Kleinburg.
#CanadaIsAwesome ##CanadianArtist
https://thegroupofseven.ca/tom-thomson/

"How to Become an Integrity Sleuth in the Library"
https://katinamagazine.org/content/article/future-of-work/2025/how-to-become-an-integrity-sleuth-in-the-library
"Open access agreement management c…

SatDINO: A Deep Dive into Self-Supervised Pretraining for Remote Sensing
Jakub Straka, Ivan Gruber
https://arxiv.org/abs/2508.21402 https://arxiv.org/pdf/2…

SatDINO: A Deep Dive into Self-Supervised Pretraining for Remote Sensing
Self-supervised learning has emerged as a powerful tool for remote sensing, where large amounts of unlabeled data are available. In this work, we investigate the use of DINO, a contrastive self-supervised method, for pretraining on remote sensing imagery. We introduce SatDINO, a model tailored for representation learning in satellite imagery. Through extensive experiments on multiple datasets in multiple testing setups, we demonstrate that SatDINO outperforms other state-of-the-art methods base…

Drop-Muon: Update Less, Converge Faster
Kaja Gruntkowska, Yassine Maziane, Zheng Qu, Peter Richt\'arik
https://arxiv.org/abs/2510.02239 https://arxiv.o…

Drop-Muon: Update Less, Converge Faster
Conventional wisdom in deep learning optimization dictates updating all layers at every step-a principle followed by all recent state-of-the-art optimizers such as Muon. In this work, we challenge this assumption, showing that full-network updates can be fundamentally suboptimal, both in theory and in practice. We introduce a non-Euclidean Randomized Progressive Training method-Drop-Muon-a simple yet powerful framework that updates only a subset of layers per step according to a randomized sche…

Cross-Platform Scaling of Vision-Language-Action Models from Edge to Cloud GPUs
Amir Taherin, Juyi Lin, Arash Akbari, Arman Akbari, Pu Zhao, Weiwei Chen, David Kaeli, Yanzhi Wang
https://arxiv.org/abs/2509.11480

Cross-Platform Scaling of Vision-Language-Action Models from Edge to Cloud GPUs
Vision-Language-Action (VLA) models have emerged as powerful generalist policies for robotic control, yet their performance scaling across model architectures and hardware platforms, as well as their associated power budgets, remain poorly understood. This work presents an evaluation of five representative VLA models -- spanning state-of-the-art baselines and two newly proposed architectures -- targeting edge and datacenter GPU platforms. Using the LIBERO benchmark, we measure accuracy alongsid…

Impact of a Sharpness Based Loss Function for Removing Out-of-Focus Blur
Uditangshu Aurangabadkar, Darren Ramsook, Anil Kokaram
https://arxiv.org/abs/2509.11735 https://

Impact of a Sharpness Based Loss Function for Removing Out-of-Focus Blur
Recent research has explored complex loss functions for deblurring. In this work, we explore the impact of a previously introduced loss function - Q which explicitly addresses sharpness and employ it to fine-tune State-of-the-Art (SOTA) deblurring models. Standard image quality metrics such as PSNR or SSIM do not distinguish sharpness from ringing. Therefore, we propose a novel full-reference image quality metric Omega that combines PSNR with Q. This metric is sensitive to ringing artefacts, bu…

Transport Based Mean Flows for Generative Modeling
Elaheh Akbari, Ping He, Ahmadreza Moradipari, Yikun Bai, Soheil Kolouri
https://arxiv.org/abs/2509.22592 https://

Transport Based Mean Flows for Generative Modeling
Flow-matching generative models have emerged as a powerful paradigm for continuous data generation, achieving state-of-the-art results across domains such as images, 3D shapes, and point clouds. Despite their success, these models suffer from slow inference due to the requirement of numerous sequential sampling steps. Recent work has sought to accelerate inference by reducing the number of sampling steps. In particular, Mean Flows offer a one-step generation approach that delivers substantial s…

Characterization of Speech Similarity Between Australian Aboriginal and High-Resource Languages: A Case Study on Dharawal
Ting Dang, Trini Manoj Jeyaseelan, Eliathamby Ambikairajah, Vidhyasaharan Sethu
https://arxiv.org/abs/2509.01419

Characterization of Speech Similarity Between Australian Aboriginal and High-Resource Languages: A Case Study on Dharawal
Australian Aboriginal languages are of significant cultural and linguistic value but remain severely underrepresented in modern speech AI systems. While state-of-the-art speech foundation models and automatic speech recognition excel in high-resource settings, they often struggle to generalize to low-resource languages, especially those lacking clean, annotated speech data. In this work, we collect and clean a speech dataset for Dharawal, a low-resource Australian Aboriginal language, by carefu…

Principled Curriculum Learning using Parameter Continuation Methods
Harsh Nilesh Pathak, Randy Paffenroth
https://arxiv.org/abs/2507.22089 https://arxiv.or…

Principled Curriculum Learning using Parameter Continuation Methods
In this work, we propose a parameter continuation method for the optimization of neural networks. There is a close connection between parameter continuation, homotopies, and curriculum learning. The methods we propose here are theoretically justified and practically effective for several problems in deep neural networks. In particular, we demonstrate better generalization performance than state-of-the-art optimization techniques such as ADAM for supervised and unsupervised learning tasks.

Quantum Portfolio Optimization: An Extensive Benchmark
Eric Stopfer, Friedrich Wagner
https://arxiv.org/abs/2509.17876 https://arxiv.org/pdf/2509.17876

Quantum Portfolio Optimization: An Extensive Benchmark
Recently, several researchers proposed portfolio optimization as a potential use case for quantum optimization. However, the literature is lacking an extensive benchmark quantifying the potential of quantum computers for portfolio optimization. In this work, we fill this gap. We provide a computational study, comparing quantum approaches against state-of-the-art classical methods on a meaningful, real-world instance set. In particular, we compare quantum annealing and the quantum approximate op…

"How Should Diamond Open Access Work for Books?" @ Katina Magazine: https://katinamagazine.org/content/article/open-knowledge/2025/how-should-diamond-oa-work-for-books

Comic book artist Ian Bertram on encouraging your inner artist https://thecreativeindependent.com/people/comic…

Comic book artist Ian Bertram on encouraging your inner artist – The Creative Independent
Comic book artist Ian Bertram discusses making time and space for creativity, finding an overall balance, and the difficulty of describing your own work.

DistrAttention: An Efficient and Flexible Self-Attention Mechanism on Modern GPUs
Haolin Jin, Mengbai Xiao, Yuan Yuan, Xiao Zhang, Dongxiao Yu, Guanghui Zhang, Haoliang Wang
https://arxiv.org/abs/2507.17245

DistrAttention: An Efficient and Flexible Self-Attention Mechanism on Modern GPUs
The Transformer architecture has revolutionized deep learning, delivering the state-of-the-art performance in areas such as natural language processing, computer vision, and time series prediction. However, its core component, self-attention, has the quadratic time complexity relative to input sequence length, which hinders the scalability of Transformers. The exsiting approaches on optimizing self-attention either discard full-contextual information or lack of flexibility. In this work, we des…

ViT-FIQA: Assessing Face Image Quality using Vision Transformers
Andrea Atzori, Fadi Boutros, Naser Damer
https://arxiv.org/abs/2508.13957 https://arxiv.or…

ViT-FIQA: Assessing Face Image Quality using Vision Transformers
Face Image Quality Assessment (FIQA) aims to predict the utility of a face image for face recognition (FR) systems. State-of-the-art FIQA methods mainly rely on convolutional neural networks (CNNs), leaving the potential of Vision Transformer (ViT) architectures underexplored. This work proposes ViT-FIQA, a novel approach that extends standard ViT backbones, originally optimized for FR, through a learnable quality token designed to predict a scalar utility score for any given face image. The le…

Most incompatible measurements and sum-of-squares optimisation
S\'ebastien Designolle
https://arxiv.org/abs/2509.10381 https://arxiv.org/pdf/2509.10381…

Most incompatible measurements and sum-of-squares optimisation
Measurement incompatibility, or joint measurability, is a cornerstone of quantum theory and a useful resource. For finite-dimensional systems, quantifying this resource and establishing universal bounds valid for all measurements is a long-standing problem. In this work, we exhibit analytical universal parent measurements giving access to bounds that beat the state of the art. In particular, we can show that, for relevant robustnesses, sets of anticommuting observables give rise to the most inc…

Risk In Context: Benchmarking Privacy Leakage of Foundation Models in Synthetic Tabular Data Generation
Jessup Byun, Xiaofeng Lin, Joshua Ward, Guang Cheng
https://arxiv.org/abs/2507.17066

Risk In Context: Benchmarking Privacy Leakage of Foundation Models in Synthetic Tabular Data Generation
Synthetic tabular data is essential for machine learning workflows, especially for expanding small or imbalanced datasets and enabling privacy-preserving data sharing. However, state-of-the-art generative models (GANs, VAEs, diffusion models) rely on large datasets with thousands of examples. In low-data settings, often the primary motivation for synthetic data, these models can overfit, leak sensitive records, and require frequent retraining. Recent work uses large pre-trained transformers to …

Warm-Starting PCE for Traveling Salesman Problem
Rafael S. do Carmo, Renato Gomes dos Reis, Samuel Fernando F Silva, Luiz Gustavo E. Arruda, Felipe F. Fanchini
https://arxiv.org/abs/2509.14414

Warm-Starting PCE for Traveling Salesman Problem
Variational quantum algorithms are promising for combinatorial optimization, but their scalability is often limited by qubit-intensive encoding schemes. To overcome this bottleneck, Pauli Correlation Encoding (PCE) has emerged as one of the most promising algorithms in this scenario. The method offers not only a polynomial reduction in qubit count and a suppression of barren plateaus but also demonstrates competitive performance with state-of-the-art methods on Maxcut. In this work, we propose …

Leveraging Multi-View Weak Supervision for Occlusion-Aware Multi-Human Parsing
Laura Bragagnolo, Matteo Terreran, Leonardo Barcellona, Stefano Ghidoni
https://arxiv.org/abs/2509.10093

Leveraging Multi-View Weak Supervision for Occlusion-Aware Multi-Human Parsing
Multi-human parsing is the task of segmenting human body parts while associating each part to the person it belongs to, combining instance-level and part-level information for fine-grained human understanding. In this work, we demonstrate that, while state-of-the-art approaches achieved notable results on public datasets, they struggle considerably in segmenting people with overlapping bodies. From the intuition that overlapping people may appear separated from a different point of view, we pro…

SplitMeanFlow: Interval Splitting Consistency in Few-Step Generative Modeling
Yi Guo, Wei Wang, Zhihang Yuan, Rong Cao, Kuan Chen, Zhengyang Chen, Yuanyuan Huo, Yang Zhang, Yuping Wang, Shouda Liu, Yuxuan Wang
https://arxiv.org/abs/2507.16884

SplitMeanFlow: Interval Splitting Consistency in Few-Step Generative Modeling
Generative models like Flow Matching have achieved state-of-the-art performance but are often hindered by a computationally expensive iterative sampling process. To address this, recent work has focused on few-step or one-step generation by learning the average velocity field, which directly maps noise to data. MeanFlow, a leading method in this area, learns this field by enforcing a differential identity that connects the average and instantaneous velocities. In this work, we argue that this d…

CountTRuCoLa: Rule Confidence Learning for Temporal Knowledge Graph Forecasting
Julia Gastinger, Christian Meilicke, Heiner Stuckenschmidt
https://arxiv.org/abs/2509.09474 https…

CountTRuCoLa: Rule Confidence Learning for Temporal Knowledge Graph Forecasting
We address the task of temporal knowledge graph (TKG) forecasting by introducing a fully explainable method based on temporal rules. Motivated by recent work proposing a strong baseline using recurrent facts, our approach learns four simple types of rules with a confidence function that considers both recency and frequency. Evaluated on nine datasets, our method matches or surpasses the performance of eight state-of-the-art models and two baselines, while providing fully interpretable predictio…