Tootfinder

Trump’s ‘gas-guzzling’ parade will produce planet-heating pollution costs, analysis says https://www.theguardian.com/us-news/2025/jun/13/trump-military-parade-pollution-emissions

«What's more, AI Overview will produce a response like this one even when you ask it about air purifiers that don't exist, like the "Levoit Core 5510," the "Winnix Airmega" and the "Coy Mega 700."»
On the faulty economics behind the #Google #AI search previews…

A Review of Privacy Metrics for Privacy-Preserving Synthetic Data Generation
Frederik Marinus Trudslev, Matteo Lissandrini, Juan Manuel Rodriguez, Martin B{\o}gsted, Daniele Dell'Aglio
https://arxiv.org/abs/2507.11324

A Review of Privacy Metrics for Privacy-Preserving Synthetic Data Generation
Privacy Preserving Synthetic Data Generation (PP-SDG) has emerged to produce synthetic datasets from personal data while maintaining privacy and utility. Differential privacy (DP) is the property of a PP-SDG mechanism that establishes how protected individuals are when sharing their sensitive data. It is however difficult to interpret the privacy loss ($\varepsilon$) expressed by DP. To make the actual risk associated with the privacy loss more transparent, multiple privacy metrics (PMs) have b…

Life on Different Paths 🚶🚶‍♀️🚶‍♂️
生活诸多路径 🚶🚶‍♀️🚶‍♂️
📷 Nikon FE
🎞️FOMAPAN Action 400
buy me ☕️ ?/请我喝杯☕️？
#filmphotography

Boosting Scientific Error-Bounded Lossy Compression through Optimized Synergistic Lossy-Lossless Orchestration
Shixun Wu, Jinwen Pan, Jinyang Liu, Jiannan Tian, Ziwei Qiu, Jiajun Huang, Kai Zhao, Xin Liang, Sheng Di, Zizhong Chen, Franck Cappello
https://arxiv.org/abs/2507.11165

Boosting Scientific Error-Bounded Lossy Compression through Optimized Synergistic Lossy-Lossless Orchestration
As high-performance computing architectures evolve, more scientific computing workflows are being deployed on advanced computing platforms such as GPUs. These workflows can produce raw data at extremely high throughputs, requiring urgent high-ratio and low-latency error-bounded data compression solutions. In this paper, we propose cuSZ-Hi, an optimized high-ratio GPU-based scientific error-bounded lossy compressor with a flexible, domain-irrelevant, and fully open-source framework design. Our n…

Bonkers, using leguminous crops to make petrol. Nearly as mad as using it to produce meat.
Biofuel demand to soak up more than half of US soyoil production next year, USDA says | Reuters https://www.reuters.com/sustainability…

LiLAC: A Lightweight Latent ControlNet for Musical Audio Generation
Tom Baker, Javier Nistal
https://arxiv.org/abs/2506.11476 https://

LiLAC: A Lightweight Latent ControlNet for Musical Audio Generation
Text-to-audio diffusion models produce high-quality and diverse music but many, if not most, of the SOTA models lack the fine-grained, time-varying controls essential for music production. ControlNet enables attaching external controls to a pre-trained generative model by cloning and fine-tuning its encoder on new conditionings. However, this approach incurs a large memory footprint and restricts users to a fixed set of controls. We propose a lightweight, modular architecture that considerably …

Feynman Integral Reduction using Syzygy-Constrained Symbolic Reduction Rules
Sid Smith, Mao Zeng
https://arxiv.org/abs/2507.11140 https://

Feynman Integral Reduction using Syzygy-Constrained Symbolic Reduction Rules
We present a new algorithm for integration-by-parts (IBP) reduction of Feynman integrals with high powers of numerators or propagators, a demanding computational step in evaluating multi-loop scattering amplitudes. The algorithm starts with solving syzygy equations in individual sectors to produce IBP operators that turn seed integrals into IBP equations without artificially raised propagator powers. The IBP operators are expressed in terms of index-shift operators and number operators. We perf…

Attorney General Bailey Fights To Expose Big Tech Censorship Of President Trump As AI Chatbots Produce Fake News (AG Bailey/Attorney General Office of Missouri)
https://ago.mo.gov/attorney-general-bailey-fights-to-expose-big-tech-censorship-of-president-trump-as-ai-chatbots-produce-fake-news/
http://www.memeorandum.com/250709/p153#a250709p153

Entropy production in active Rouse polymers
Sandipan Dutta
https://arxiv.org/abs/2506.12013 https://arxiv.org/pdf/2506.12013

Entropy production in active Rouse polymers
Active polymers are the archetype of nonequilibrium viscoelastic systems that constantly consume energy to produce motion. The activity of many biopolymers is essential to many life processes. The entropy production rate quantifies their nonequilibrium nature through the breaking of the time reversal symmetry. In this work we build an analytical model of active polymers as active Rouse polymers where the beads are active Ornstein Uhlenbeck particles and calculate their entropy production. The i…

Asymmetric muon-antimuon emission from $Z^0$ decays: a clear magnetometer in relativistic heavy-ion collisions
Alejandro Ayala, Ana Julia Mizher, Javier Rend\'on
https://arxiv.org/abs/2506.11370

Asymmetric muon-antimuon emission from $Z^0$ decays: a clear magnetometer in relativistic heavy-ion collisions
We show that a very clear signal of the presence of a strong magnetic field during the early stage of a high-energy heavy-ion collision is provided by the decay of the Z^0 into dimuon pairs. We find that the process is highly anisotropic, producing pairs mainly out of plane, as signaled by a negative value of v_2, and leads to a harder antimuon transverse momentum spectrum, compared with that of the muon transverse momentum spectrum. We also show that the process does not produce a significant …

Multivariate Time-series Transformer Embeddings for Light Curves
Gabriel Chiong, Ignacio Becker, Pavlos Protopapas
https://arxiv.org/abs/2506.11637 https:/…

Multivariate Time-series Transformer Embeddings for Light Curves
Astronomical surveys produce time-series data by observing stellar objects across multiple wavelength bands. Foundational transformer-based models, such as Astromer, encode each time-series as a sequence of embeddings of uniform dimensions. However, such models operate independently on each band at a single time and do not natively leverage information across telescope filters. We extend this framework by introducing a fusion mechanism that maps the collection of single-band embeddings to a uni…

The Potential Danger to Satellites due to Ejecta from a 2032 Lunar Impact by Asteroid 2024 YR4
Paul Wiegert, Peter Brown, Jack Lopes, Martin Connors
https://arxiv.org/abs/2506.11217

The Potential Danger to Satellites due to Ejecta from a 2032 Lunar Impact by Asteroid 2024 YR4
On 2032 December 22 the 60 m diameter asteroid 2024 YR4 has a 4% chance of impacting the Moon. Such an impact would release 6.5 MT TNT equivalent energy and produce a ~1 km diameter crater. We estimate that up to 10^8 kg of lunar material could be liberated in such an impact by exceeding lunar escape speed. Depending on the actual impact location on the Moon as much as 10% of this material may accrete to the Earth on timescales of a few days. The lunar ejecta-associated particle fluence at 0.1 …

🌳 InventWood is about to mass produce wood that’s stronger than steel
https://techcrunch.com/2025/05/12/inventwood-is-about-to-mass-produce-wood-thats-stronger-than-steel/

Exclusive: InventWood is about to mass-produce wood that’s stronger than steel | TechCrunch
The material has 50% more tensile strength than steel with a strength-to-weight ratio that’s ten-times better.

A visco-plastic constitutive model for accurate densification and shape predictions in powder metallurgy hot isostatic pressing
Subrato Sarkar, Jason R Mayeur, KPK Ajjarapu, Fred A List III, Soumya Nag, Ryan R Dehoff
https://arxiv.org/abs/2506.11946

A visco-plastic constitutive model for accurate densification and shape predictions in powder metallurgy hot isostatic pressing
Powder metallurgy hot isostatic pressing (PM-HIP) is an advanced manufacturing process that produces near net shape parts with high material utilization and uniform microstructures. Despite being used frequently to produce small-scale components, the application of PM-HIP to large-scale components is limited due to inadequate understanding of its complex mechanisms that cause unpredictable post-HIP shape distortions. A computational model can provide necessary information about the intermediate…

Optimal experiment design for practical parameter identifiability and model discrimination
Yue Liu (University of Oxford, Purdue University), Philip K. Maini (University of Oxford), Ruth E. Baker (University of Oxford)
https://arxiv.org/abs/2506.11311

Optimal experiment design for practical parameter identifiability and model discrimination
Mechanistic mathematical models of biological systems usually contain a number of unknown parameters whose values need to be estimated from available experimental data in order for the models to be validated and used to make quantitative predictions. This requires that the models are practically identifiable, that is, the values of the parameters can be confidently determined, given available data. A well-designed experiment can produce data that are much more informative for the purpose of inf…

Good discussion of Apple's AI paper ((without denying that GenAI can' t think or reason and produce bullshit)
#AI

Achieving Optimal Locomotion using Self-Generated Waves
D. O'Donovan, M. D. Bustamante, O. Devauchelle, G. P. Benham
https://arxiv.org/abs/2506.11961 h…

Achieving Optimal Locomotion using Self-Generated Waves
An oscillating body floating at the water surface produces a wave-field of self-generated waves. When the oscillation induces a difference in fore-aft wave amplitude squared, these self-generated waves can be used as a mechanism to propel the body horizontally across the surface (Longuet-Higgins and Stewart 1964). The optimisation of this wave-driven propulsion (WDP) is the interest of this work. To study the conditions necessary to produce optimal thrust we will utilise a shallow water set-up …

Trump Seeks to Cut Basic Scientific Research by Roughly One-Third,
An analysis by the American Association for the Advancement of Science
shows the impact of the administration’s budget plan on the kind of studies that produce the most breakthroughs.
https://…

Imagine you want to produce E-Fuels⚡⛽ but the power company does not want to sell you electricity⚡
That appears to be what potential E-Fuels producers faced in Iceland🇮🇸. The country uses 100% renewable electricity (hydro/geo) & could produce more.
Yet, Iceland's electricity producer Landsvirkjun does not have enough electricity⚡. Building new plants is delayed by slow permitting 📝. Landsvirkjun has therefore set out priorities.

Iceland's E-Fuels Controversy
Hydrogen and E-Fuels companies want to utilize Iceland's abundant renewable energy sources. But Iceland's largest energy company Landsvirkjun does not want to sell electricity to them. E-Fuels producers hope that European authorities will intervene, the EFTA Surveillane Authority investigates.

What Pareto-Efficiency Adjustments Cannot Fix
Josue Ortega, Gabriel Ziegler, R. Pablo Arribillaga, Geng Zhao
https://arxiv.org/abs/2506.11660 https://

What Pareto-Efficiency Adjustments Cannot Fix
The Deferred Acceptance (DA) algorithm is stable and strategy-proof, but can produce outcomes that are Pareto-inefficient for students, and thus several alternative mechanisms have been proposed to correct this inefficiency. However, we show that these mechanisms cannot correct DA's rank-inefficiency and inequality, because these shortcomings can arise even in cases where DA is Pareto-efficient. We also examine students' segregation in settings with advantaged and marginalized students. We pr…

Ukraine, US company to co-produce hundreds of thousands of drones in 2025, Zelensky announces: https://benborges.xyz/2025/07/03/ukraine-us-company-to-coproduce.html

NL in the Middle: Code Translation with LLMs and Intermediate Representations
Chi-en Amy Tai, Pengyu Nie, Lukasz Golab, Alexander Wong
https://arxiv.org/abs/2507.08627

NL in the Middle: Code Translation with LLMs and Intermediate Representations
Studies show that large language models (LLMs) produce buggy code translations. One avenue to improve translation accuracy is through intermediate representations, which could provide structured insights to guide the model's understanding. We explore whether code translation using LLMs can benefit from intermediate representations via natural language (NL) and abstract syntax trees (ASTs). Since prompt engineering greatly affects LLM performance, we consider several ways to integrate these repr…

Imitation Learning for Obstacle Avoidance Using End-to-End CNN-Based Sensor Fusion
Lamiaa H. Zain, Hossam H. Ammar, Raafat E. Shalaby
https://arxiv.org/abs/2507.08112

Imitation Learning for Obstacle Avoidance Using End-to-End CNN-Based Sensor Fusion
Obstacle avoidance is crucial for mobile robots' navigation in both known and unknown environments. This research designs, trains, and tests two custom Convolutional Neural Networks (CNNs), using color and depth images from a depth camera as inputs. Both networks adopt sensor fusion to produce an output: the mobile robot's angular velocity, which serves as the robot's steering command. A newly obtained visual dataset for navigation was collected in diverse environments with varying lighting con…

This does say a lot about how easy it is to trick our brains into thinking that something is a person, just because they produce words.. I am quite surprised that the friends and family don't push back more though..
"‘I felt pure, unconditional love’: the people who marry their AI chatbots"

Very excited about this! Code to access GRIN will help lots of Google Books partners, and the example might open other doors, as well as the obvious benefits of access to data!
'Institutional Books 1.0: A 242B token dataset from Harvard Library's collections, refined for accuracy and usability' https://arxiv.org/abs/2506…

Institutional Books 1.0: A 242B token dataset from Harvard Library's collections, refined for accuracy and usability
Large language models (LLMs) use data to learn about the world in order to produce meaningful correlations and predictions. As such, the nature, scale, quality, and diversity of the datasets used to train these models, or to support their work at inference time, have a direct impact on their quality. The rapid development and adoption of LLMs of varying quality has brought into focus the scarcity of publicly available, high-quality training data and revealed an urgent need to ground the steward…

"Ultimately, engaging the public enriches our scholarship. It pushes us to think more carefully about clarity, ethics, and relevance. It also ensures that the knowledge we produce contributes to a wider societal good, affirming the role of the university not just as a place of knowledge production, but also of public service."
#OpenScience
https://social.edu.nl/@Bibliothecaris/114833640819292319

First images of Antares photosphere from spectropolarimetry
Quentin Pilate, Eric Josselin, Agn\`es L\`ebre, Arturo L\'opez Ariste, Philippe Mathias, Alexis Lavail
https://arxiv.org/abs/2507.08614

First images of Antares photosphere from spectropolarimetry
Antares is the closest red supergiant (RSG) to Earth. The discovery of linear polarization in the atomic lines of the star opens the path to produce direct images of the photosphere, hence probing the dynamics at the surface. We analyze this linear polarization signals following the same scheme as has been previously done for the RSG Betelgeuse, and find that they are comparable in all its details. This allows us to use the same models for the analysis of these polarization signals in both star…

It is very revealing that @… has received money from the fascist Chinese government to publish their propaganda article “explaining” why Taiwan allegedly belongs to #China and why recognising Taiwan's struggle for independence from the totalitarian-ruled Chinese empire…

Brain-wide interpolation and conditioning of gene expression in the human brain using Implicit Neural Representations
Xizheng Yu, Justin Torok, Sneha Pandya, Sourav Pal, Vikas Singh, Ashish Raj
https://arxiv.org/abs/2506.11158

Brain-wide interpolation and conditioning of gene expression in the human brain using Implicit Neural Representations
In this paper, we study the efficacy and utility of recent advances in non-local, non-linear image interpolation and extrapolation algorithms, specifically, ideas based on Implicit Neural Representations (INR), as a tool for analysis of spatial transcriptomics data. We seek to utilize the microarray gene expression data sparsely sampled in the healthy human brain, and produce fully resolved spatial maps of any given gene across the whole brain at a voxel-level resolution. To do so, we first obt…

FWIW, I’ve yet to see any indication that the use of LLMs (pseudo-AI) has improved the quality of phishing as measured by how much gets past technical defenses.
LLMs are a great leveler. They produce median texts to fit their prompts. They cannot produce anything that requires creativity. They cannot produce high-quality fakes because they cannot produce high-quality anything. They are a play on the fact that 50% of people are at or below median cognitive capacity.

A group of merging galaxies falling onto Abell 2142
Aashiya Anitha Shaji, Ana\"elle Hall\'e, Damir Gasymov, Anne-Laure Melchior, Fran\c{c}oise Combes, Andrea Cattaneo
https://arxiv.org/abs/2507.08172

A group of merging galaxies falling onto Abell 2142
Galaxy clusters produce a very hostile environment to galaxies, whose gas gets stripped by ram-pressure, suffer galaxy interactions and witness quenching of their star formation. Clusters, like Abell 2142, grow not only through galaxy accretion but also through group infall. Our goal is to study the physical and dynamical state of the most conspicuous infalling group, on a filament projected at 1.3 Mpc from the Abell 2142 centre. The galaxy group is the leading edge of a spectacular trailing 70…

Poro-viscoelastic tidal heating of Io
Hamish C. F. C. Hay, Ian Hewitt, Marc Rovira-Navarro, Richard F. Katz
https://arxiv.org/abs/2507.11266 https://

Poro-viscoelastic tidal heating of Io
Io's tidally-driven global volcanism indicates widespread partial melting in its mantle. How this melt participates in the interior dynamics, and, in particular, the role it plays in tidal dissipation, is poorly understood. We model Io's tidal deformation by treating its mantle as a two-phase (solid and melt) system. By combining poro-viscous and poro-elastic compaction theories in a Maxwell framework with a consistent model of tidal and self-gravitation, we produce the first self-consistent ev…

🇺🇦 #NowPlaying on #BBC6Music's #DonLettsCultureClashRadio
Fat Produce:
🎵 No Way
#FatProduce
https://fatproduce.bandcamp.com/album/no-way-54-46-was-my-number
https://open.spotify.com/track/655GfEtxXoADMMYxM0ZScy

Search for a Dark Gauge Boson Within Einstein-Cartan Theory at the ILC Using Multivariate Analysis
Hossam Taha, El-sayed A. El-dahshan, S. Elgammal
https://arxiv.org/abs/2507.08678

Search for a Dark Gauge Boson Within Einstein-Cartan Theory at the ILC Using Multivariate Analysis
Multivariate analysis (MVA) is employed to probe the dark matter candidate A$^{\prime}$, a gauge boson of a model rooted within Einstein-Cartan Theory, at the International Linear Collider (ILC). \texttt{WHIZARD} package is used as the event generator to simulate electron-positron interactions at the ILC, at a 500 GeV center-of-mass energy ($\sqrt s$) and a 500 fb$^{-1}$ detector's integrated luminosity ($\mathcal{L}$), to produce the A$^{\prime}$ signal and the expected standard model backgrou…

On a few pitfalls in KL divergence gradient estimation for RL
Yunhao Tang, R\'emi Munos
https://arxiv.org/abs/2506.09477 https://…

On a few pitfalls in KL divergence gradient estimation for RL
We point out a few pitfalls in implementing gradient estimation for KL divergence in RL training for LLM, as seen in a number of open source projects and papers. The first major pitfall is to differentiate through the KL estimate as loss functions to minimize KL divergence. We show that such implementations are generally incorrect and do not produce the desired KL gradient. Secondly, we show that some implementations do not account for the sequential nature of the estimation problem and produce…

Step-by-step Instructions and a Simple Tabular Output Format Improve the Dependency Parsing Accuracy of LLMs
Hiroshi Matsuda, Chunpeng Ma, Masayuki Asahara
https://arxiv.org/abs/2506.09983

Step-by-step Instructions and a Simple Tabular Output Format Improve the Dependency Parsing Accuracy of LLMs
Recent advances in large language models (LLMs) have enabled impressive performance in various tasks. However, standard prompting often struggles to produce structurally valid and accurate outputs, especially in dependency parsing. We propose a novel step-by-step instruction strategy, where universal part-of-speech tagging precedes the prediction of syntactic heads and dependency labels, and a simplified CoNLL-U like output format, our method achieves state-of-the-art accuracy on Universal Depe…

I wonder if AI’s effectiveness will speed up the inevitable death of some ecosystems.
If folks want the “best” (and I mean least shit) Agentic AI experience, they will probably gravitate towards more populated communities where the training data can produce a semblance of “working”.
This will likely flat-line some communities since many devs want something that works.

Plants produce more nectar when they ‘hear’ bees buzzing, scientists find;
Researchers suggest buzzing noises could be used as environmentally friendly way to enhance crop pollination
https://www.theguardian.com/environment/20

Assessing the Quality of Denoising Diffusion Models in Wasserstein Distance: Noisy Score and Optimal Bounds
Vahan Arsenyan, Elen Vardanyan, Arnak Dalalyan
https://arxiv.org/abs/2506.09681

Assessing the Quality of Denoising Diffusion Models in Wasserstein Distance: Noisy Score and Optimal Bounds
Generative modeling aims to produce new random examples from an unknown target distribution, given access to a finite collection of examples. Among the leading approaches, denoising diffusion probabilistic models (DDPMs) construct such examples by mapping a Brownian motion via a diffusion process driven by an estimated score function. In this work, we first provide empirical evidence that DDPMs are robust to constant-variance noise in the score evaluations. We then establish finite-sample guara…

Robust copula estimation for one-shot devices with correlated failure modes
E. Castilla, P. J. Chocano
https://arxiv.org/abs/2506.10152 https://

Robust copula estimation for one-shot devices with correlated failure modes
This paper presents a robust method for estimating copula models to evaluate dependence between failure modes in one-shot devices-systems designed for single use and destroyed upon activation. Traditional approaches, such as maximum likelihood estimation (MLE), often produce unreliable results when faced with outliers or model misspecification. To overcome these limitations, we introduce a divergence-based estimation technique that enhances robustness and provides a more reliable characterizati…

Multiple telescopes produce new view of Andromeda #space

You've never seen the Andromeda Galaxy like this before. Multiple telescopes produce new view of our spiral neighbour | BBC Sky at Night Magazine
Image of the Andromeda Galaxy in multiple wavelengths, produced by multiple space telescopes and ground-based observatories.

Urban Discoveries II 🌆
都市发现 II 🌆
📷 Pentax MX
🎞️Lucky SHD 400
buy me ☕️ ?/请我喝杯☕️？
#filmphotography

Yiding Cao Geld mit PayPal.Me senden
Folgen Sie dem Link PayPal.Me/dingsterdingster und geben Sie den Betrag ein. Es ist bequem und sicher. Sie haben noch kein PayPal-Konto? Kein Problem.

Last week I had the great honor to give a talk with my friend Malte about capitalism, (tech-)fascism and how it connects to "AI".
(Original title: “Fascist AI” talk at LOOPS)
https://tante.cc/2025/07/04/fascist-ai-talk-at-loops/

"Fascist AI" talk at LOOPS
The New Practice space at TU Berlin runs a series of talks called LOOPS. Together with my friend Malte I was invited to talk a bit about “Fascist AI”. How capitalism, fascism and AI narratives are very closely aligned and (re)produce one another. I enjoyed giving the talk (I rarely do those together with others) […]

Does Residuals-on-Residuals Regression Produce Representative Estimates of Causal Effects?
Apoorva Lal, Winston Chou
https://arxiv.org/abs/2506.07462 https…

Does Residuals-on-Residuals Regression Produce Representative Estimates of Causal Effects?
Double Machine Learning is commonly used to estimate causal effects in large observational datasets. The "residuals-on-residuals" regression estimator (RORR) is especially popular for its simplicity and computational tractability. However, when treatment effects are heterogeneous, the proper interpretation of RORR may not be well understood. We show that, for many-valued treatments with continuous dose-response functions, RORR converges to a conditional variance-weighted average of derivatives …

Computing Floating-Point Errors by Injecting Perturbations
Youshuai Tan, Zhanwei Zhang, Jinfu Chen, Zishuo Ding, Jifeng Xuan, Weiyi Shang
https://arxiv.org/abs/2507.08467

Computing Floating-Point Errors by Injecting Perturbations
Floating-point programs form the foundation of modern science and engineering, providing the essential computational framework for a wide range of applications, such as safety-critical systems, aerospace engineering, and financial analysis. Floating-point errors can lead to severe consequences. Although floating-point errors widely exist, only a subset of inputs may trigger significant errors in floating-point programs. Therefore, it is crucial to determine whether a given input could produce s…

Predictive control of wastewater treatment plants as energy-autonomous water resource recovery facilities
Otacilio B. L. Neto, Michela Mulas, Iiro Harjunkoski, Francesco Corona
https://arxiv.org/abs/2506.10490

Predictive control of wastewater treatment plants as energy-autonomous water resource recovery facilities
This work proposes an automatic control solution for the operation of conventional wastewater treatment plants (WWTPs) as energy-autonomous water resource recovery facilities. We first conceptualize a classification of the quality of treated water for three resource recovery applications (environmental, industrial, and agricultural water reuse). We then present an output-feedback model predictive controller (Output MPC) that operates a plant to produce water of specific quality class, while als…

This https://arxiv.org/abs/2505.24718 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csCV_…

Reinforcing Video Reasoning with Focused Thinking
Recent advancements in reinforcement learning, particularly through Group Relative Policy Optimization (GRPO), have significantly improved multimodal large language models for complex reasoning tasks. However, two critical limitations persist: 1) they often produce unfocused, verbose reasoning chains that obscure salient spatiotemporal cues and 2) binary rewarding fails to account for partially correct answers, resulting in high reward variance and inefficient learning. In this paper, we propos…

"Ocean currents can generate electricity, and our study shows Africa's seas have some of the strongest"
#Africa #Oceans #Energy

Ocean currents can generate electricity, and our study shows Africa's seas have some of the strongest
The world's oceans cover more than 70% of Earth's surface. They're filled with currents, some much stronger than the fastest flowing large rivers. These currents can be harnessed as clean, marine renewable energy. Marine energy is much more predictable and reliable than many other forms of renewable energy because unlike the sun and wind, which regularly do not produce electricity, ocean currents never stop moving around the planet. New research has found that the eastern and south-eastern coas…

Frågar du mig så finns det redan definitioner som är värda att testa i lite fler studier. Till exempel "Hyper-Palatable Foods". Jag har inte direkt något förtroende för att nuvarande administrationen i USA ska lyckas få till något som fungerar bra om det testas. Men ändå spännande att följa.
Defining Ultra-Processed Foods: Will FDA Do It Better?

Defining Ultra-Processed Foods: Will FDA Do It Better? - ConscienHealth
FDA Commissioner Marty Makary believes defining ultra-processed foods is key to overcoming the health problems they produce.

In this moment when equity & inclusion efforts are being hindered & erased, the DLF Committee on Equity & Inclusion is creating a Zotero library of related resources for the GLAM community. Let us know what topics would be most helpful to you: https://docs.google.com/f…

OPM Wants Supervisors to Produce a ‘Positive and Compliant’ Federal Workplace
https://www.fedweek.com/federal-managers-daily-report/opm-wants-supervisors-to-produce-a-positive-and-compliant-federal-workplace/

Does anyone have experience with running docker images of other Linux distros in GitHub Actions CI?
Use case is validating that things build and run under e.g. Fedora or Arch given an outer Ubuntu system. It won't be perfect because you're still running the Ubuntu kernel but should be good enough to find obvious build problems, produce usable nightly builds, etc.

This https://arxiv.org/abs/2505.20302 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csPL_…

VeriThoughts: Enabling Automated Verilog Code Generation using Reasoning and Formal Verification
This paper introduces VeriThoughts, a novel dataset designed for reasoning-based Verilog code generation. We establish a new benchmark framework grounded in formal verification methods to evaluate the quality and correctness of generated hardware descriptions. Additionally, we present a suite of specialized small-scale models optimized specifically for Verilog generation. Our work addresses the growing need for automated hardware design tools that can produce verifiably correct implementations …

Bears WR Rome Odunze worked on upper-body strength, ready to make Year 2 leap with Caleb Williams

https://www.cbssports.com/nfl/news/bears-w

Thanks for playing our games. See you at 9PM EST for the new game.
This may or may not be a clue for tonight's game.
See you then.

Exploiting Control-flow Enforcement Technology for Sound and Precise Static Binary Disassembly
Brian Zhao, Yiwei Yang, Yusheng Zheng, Andi Quinn
https://arxiv.org/abs/2506.09426

Exploiting Control-flow Enforcement Technology for Sound and Precise Static Binary Disassembly
Rewriting x86_64 binaries-whether for security hardening, dynamic instrumentation, or performance profiling is notoriously difficult due to variable-length instructions, interleaved code and data, and indirect jumps to arbitrary byte offsets. Existing solutions (e.g., "superset disassembly") ensure soundness but incur significant overhead and produce large rewritten binaries, especially for on-the-fly instrumentation. This paper addresses these challenges by introducing the Time Variance Author…

This https://arxiv.org/abs/2503.15487 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_ees…

Fast Two-photon Microscopy by Neuroimaging with Oblong Random Acquisition (NORA)
Advances in neural imaging have enabled neuroscientists to study how large neural populations conspire to produce perception, behavior and cognition. Despite many advances in optical methods, there exists a fundamental tradeoff between imaging speed, field of view, and resolution that limits the scope of neural imaging, especially for the raster-scanning multi-photon imaging needed to image deeper into the brain. One approach to overcoming this trade-off is computational imaging: the co-develop…

We’re witnessing unprecedented attempts to bully academic institutions with the administration’s ideological aims.
These attempts challenge long-standing norms of academic freedom
—that is, the ability of a teacher or researcher in higher education to investigate and discuss subjects without fear of political interference.
Our elected officials should stand up for scientific research and those who produce it in the face of politically motivated attacks.
But public a…

Apathy Over Attacks on Higher Education Needs to Change
If Americans don’t fight back against efforts to dismantle higher education, the U.S. will lose lifesaving medical research, innovation that spurs our economy and the ability to freely study science and society

This https://arxiv.org/abs/2504.21726 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_mat…

Notes sur l'application d'Albanese pour les zéro-cycles
For a smooth projective variety X over an arbitrary field k, we discuss the surjectivity of the Albanese map from the Chow group of zero-cycles of degree zero on X to the group of rational points of the Albanese variety of X. Over arithmetic fields, we use Severi-Brauer fibrations to produce examples where the map is not surjective. For varieties X over the complex field, we discuss the question after extension of the complex field to function fields of varieties. This is related to C. Voisin's…

Urban Discoveries II 🌆
都市发现 II 🌆
📷 Pentax MX
🎞️Lucky SHD 400
buy me ☕️ ?/请我喝杯☕️？
#filmphotography

Yiding Cao Geld mit PayPal.Me senden
Folgen Sie dem Link PayPal.Me/dingsterdingster und geben Sie den Betrag ein. Es ist bequem und sicher. Sie haben noch kein PayPal-Konto? Kein Problem.

GenAI is the new Offshoring #ai #llm
https://ardalis.com/genai-is-the-new-offshoring/

This https://arxiv.org/abs/2401.16508 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_mat…

A synthetic approach to detecting $v_1$-periodic families
We provide a simple proof that the unit map from the sphere spectrum to the connective image-of-$J$ spectrum $\mathrm{j}$ is surjective on homotopy groups. This is achieved using a novel $t$-structure on the category of $E$-synthetic spectra and a specific construction of $\mathbf{F}_p$- and BP-synthetic lifts of $\mathrm{j}$. These synthetic lifts then easily produce modified Adams and Adams--Novikov spectral sequences for $\mathrm{j}$ which we use the prove the above detection statement, all …

Kuo expects Apple to mass produce a Vision Pro with M5 in Q3 2025, a Vision Air in Q3 2027, a Vision Pro with a new design and XR glasses in H2 2028, and more (Marcus Mendes/9to5Mac)
https://9to5mac.com/2025/06/29/apple-working-on-7-head-mounted-products/

Kuo: Apple working on 7 head-mounted products - 9to5Mac
According to him, the company has at least seven head-worn devices in development across two product families, one coming in 2025.

Lambert's problem in orbital dynamics: a self--contained introduction
Lenox Helene Baloglou, Parneet Gill, Tonatiuh S\'anchez-Vizuet
https://arxiv.org/abs/2506.10556

Lambert's problem in orbital dynamics: a self--contained introduction
Lambert's problem is a classical boundary value problem in analytical mechanics. It arises when trying to determine the energy required to place a particle, subject to a central gravitational potential, in a free fall trajectory connecting two given points on a desired travel time. Due to its mathematical beauty and its relevance in aerospace engineering, it has been and remains the object of attention of countless engineers, mathematicians (pure and applied), and physicists seeking to produce …

This https://arxiv.org/abs/2412.02544 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_…

Testing inflation on all scales: a case study with $α$-attractors
A plethora of inflationary models can produce interesting small-scale phenomenology, such as enhanced scalar fluctuations leading to primordial black hole (PBH) production and large scalar-induced GW. Nevertheless, good models must simultaneously explain current observations on all scales. In this work, we showcase our methodology to establish the small-scale phenomenology of inflationary models on firm grounds. We consider the case of hybrid $α$-attractors, and focus on a reduced parameter sp…

🇺🇦 #NowPlaying on #BBC6Music's #TheCraigCharlesFunkAndSoulShow
Fat Produce:
🎵 No Way
#FatProduce
https://fatproduce.bandcamp.com/album/no-way-54-46-was-my-number
https://open.spotify.com/track/655GfEtxXoADMMYxM0ZScy

This https://arxiv.org/abs/2409.09685 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_mat…

On the critical finite-size gap scaling for frustration-free Hamiltonians
We prove that the critical finite-size gap scaling for frustration-free Hamiltonians is of inverse-square type. The result covers general graphs embedded in $\mathbb R^D$ and general finite-range interactions without requiring assumptions about the ground state correlations. Therefore, the inverse-square critical gap scaling is a robust, universal property of finite-range frustration-free Hamiltonians. This places further limits on their ability to produce conformal field theories in the contin…

rsyslog project ponders changing from tabs to spaces for C code.
"AI-powered tools like GitHub Copilot and other code generators are heavily optimized for space-indented code. They often struggle to produce correct, tab-indented code, which limits our ability to use these tools effectively."
https://www.

Hopeful Marxists: "If the #OneBigBeautifulBill doesn't produce some class consciousness of those getting impoverished, then nothing will!"
<time passes>
Marxists: "Alright, nothing will!"

AC/DC: LLM-based Audio Comprehension via Dialogue Continuation
Yusuke Fujita, Tomoya Mizumoto, Atsushi Kojima, Lianbo Liu, Yui Sudo
https://arxiv.org/abs/2506.10312

AC/DC: LLM-based Audio Comprehension via Dialogue Continuation
We propose an instruction-following audio comprehension model that leverages the dialogue continuation ability of large language models (LLMs). Instead of directly generating target captions in training data, the proposed method trains a model to produce responses as if the input caption triggered a dialogue. This dialogue continuation training mitigates the caption variation problem. Learning to continue a dialogue effectively captures the caption's meaning beyond its surface-level words. As a…

Scattering theory on Riemann surfaces I: Schiffer operators, cohomology, and index theorems
Eric Schippers, Wolfgang Staubach
https://arxiv.org/abs/2506.08160

Scattering theory on Riemann surfaces I: Schiffer operators, cohomology, and index theorems
We consider a compact Riemann surface $\mathscr{R}$ with a complex of non-intersecting Jordan curves, whose complement is a pair of Riemann surfaces with boundary, each of which may be possibly disconnected. We investigate conformally invariant integral operators of Schiffer, which act on $L^{2}$ anti-holomorphic one-forms on one of these surfaces with boundary and produce holomorphic one-forms on the disjoint union. These operators arise in potential theory, boundary value problems, approximat…

Iran will likely be able to produce enriched uranium 'in a matter of months', IAEA chief says (Agence France-Presse)
https://www.theguardian.com/world/2025/jun/29/iran-will-likely-be-able-to-produce-enriched-uranium-in-a-matter-of-months-iaea-chief-says
http://www.memeorandum.com/250629/p6#a250629p6

Honey bees remove 80% of pollen—leaving native bees with nothing #evironment

Honey bees remove 80% of pollen—leaving native bees with nothing
Feral honey bees, once celebrated for their agricultural value, are now threatening native ecosystems in Southern California by monopolizing pollen sources and overwhelming native pollinators. A new study reveals they remove up to 80% of pollen in a single day, severely disrupting food sources for over 700 species of native bees. Despite their benefits to agriculture, these invasive bees dominate nearly all bee biomass in the region and even produce lower-quality offspring when pollinating nati…

Athena: Enhancing Multimodal Reasoning with Data-efficient Process Reward Models
Shuai Wang, Zhenhua Liu, Jiaheng Wei, Xuanwu Yin, Dong Li, Emad Barsoum
https://arxiv.org/abs/2506.09532

Athena: Enhancing Multimodal Reasoning with Data-efficient Process Reward Models
We present Athena-PRM, a multimodal process reward model (PRM) designed to evaluate the reward score for each step in solving complex reasoning problems. Developing high-performance PRMs typically demands significant time and financial investment, primarily due to the necessity for step-level annotations of reasoning steps. Conventional automated labeling methods, such as Monte Carlo estimation, often produce noisy labels and incur substantial computational costs. To efficiently generate high-q…

From Intention to Execution: Probing the Generalization Boundaries of Vision-Language-Action Models
Irving Fang, Juexiao Zhang, Shengbang Tong, Chen Feng
https://arxiv.org/abs/2506.09930

From Intention to Execution: Probing the Generalization Boundaries of Vision-Language-Action Models
One promise that Vision-Language-Action (VLA) models hold over traditional imitation learning for robotics is to leverage the broad generalization capabilities of large Vision-Language Models (VLMs) to produce versatile, "generalist" robot policies. However, current evaluations of VLAs remain insufficient. Traditional imitation learning benchmarks are unsuitable due to the lack of language instructions. Emerging benchmarks for VLAs that incorporate language often come with limited evaluation ta…

What is the Cost of Differential Privacy for Deep Learning-Based Trajectory Generation?
Erik Buchholz, Natasha Fernandes, David D. Nguyen, Alsharif Abuadbba, Surya Nepal, Salil S. Kanhere
https://arxiv.org/abs/2506.09312

What is the Cost of Differential Privacy for Deep Learning-Based Trajectory Generation?
While location trajectories offer valuable insights, they also reveal sensitive personal information. Differential Privacy (DP) offers formal protection, but achieving a favourable utility-privacy trade-off remains challenging. Recent works explore deep learning-based generative models to produce synthetic trajectories. However, current models lack formal privacy guarantees and rely on conditional information derived from real data during generation. This work investigates the utility cost of e…

TTrace: Lightweight Error Checking and Diagnosis for Distributed Training
Haitian Jiang, Shaowei Zhu, Zhen Zhang, Zhenyu Song, Xinwei Fu, Zhen Jia, Yida Wang, Jinyang Li
https://arxiv.org/abs/2506.09280

TTrace: Lightweight Error Checking and Diagnosis for Distributed Training
Distributed training is essential for scaling the training of large neural network models, such as large language models (LLMs), across thousands of GPUs. However, the complexity of distributed training programs makes them particularly prone to silent bugs, which do not produce explicit error signal but lead to incorrect training outcome. Effectively detecting and localizing such silent bugs in distributed training is challenging. Common debugging practice using metrics like training loss or gr…

BugGen: A Self-Correcting Multi-Agent LLM Pipeline for Realistic RTL Bug Synthesis
Surya Jasper, Minh Luu, Evan Pan, Aakash Tyagi, Michael Quinn, Jiang Hu, David Kebo Houngninou
https://arxiv.org/abs/2506.10501

BugGen: A Self-Correcting Multi-Agent LLM Pipeline for Realistic RTL Bug Synthesis
Hardware complexity continues to strain verification resources, motivating the adoption of machine learning (ML) methods to improve debug efficiency. However, ML-assisted debugging critically depends on diverse and scalable bug datasets, which existing manual or automated bug insertion methods fail to reliably produce. We introduce BugGen, a first of its kind, fully autonomous, multi-agent pipeline leveraging Large Language Models (LLMs) to systematically generate, insert, and validate realisti…

SPARC: Concept-Aligned Sparse Autoencoders for Cross-Model and Cross-Modal Interpretability
Ali Nasiri-Sarvi, Hassan Rivaz, Mahdi S. Hosseini
https://arxiv.org/abs/2507.06265

SPARC: Concept-Aligned Sparse Autoencoders for Cross-Model and Cross-Modal Interpretability
Understanding how different AI models encode the same high-level concepts, such as objects or attributes, remains challenging because each model typically produces its own isolated representation. Existing interpretability methods like Sparse Autoencoders (SAEs) produce latent concepts individually for each model, resulting in incompatible concept spaces and limiting cross-model interpretability. To address this, we introduce SPARC (Sparse Autoencoders for Aligned Representation of Concepts), a…

Distinguishing Monochromatic Signals in LISA and Taiji: Ultralight Dark Matter versus Gravitational Waves
Heng-Tao Xu, Yue-Hui Yao, Yong Tang, Yue-Liang Wu
https://arxiv.org/abs/2506.09744

Distinguishing Monochromatic Signals in LISA and Taiji: Ultralight Dark Matter versus Gravitational Waves
Ultralight dark matter (ULDM) is an attractive candidate for cold dark matter, one of the main mysterious components of the Universe. Recent studies suggest that gravitational-wave (GW) laser interferometers can also detect bosonic ULDM fields, which would produce monochromatic signals resembling those from gravitational waves (GWs). Distinguishing between these potential origins therefore would be essential. In this work, we develop a method to address this challenge for space-based GW interfe…

Execution-Aware Program Reduction for WebAssembly via Record and Replay
Doehyun Baek, Daniel Lehmann, Ben L. Titzer, Sukyoung Ryu, Michael Pradel
https://arxiv.org/abs/2506.07834 …

Execution-Aware Program Reduction for WebAssembly via Record and Replay
WebAssembly (Wasm) programs may trigger bugs in their engine implementations. To aid debugging, program reduction techniques try to produce a smaller variant of the input program that still triggers the bug. However, existing execution-unaware program reduction techniques struggle with large and complex Wasm programs, because they rely on static information and apply syntactic transformations, while ignoring the valuable information offered by the input program's execution behavior. We presen…

This https://arxiv.org/abs/2505.22772 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csLG_…

Calibrated Value-Aware Model Learning with Stochastic Environment Models
The idea of value-aware model learning, that models should produce accurate value estimates, has gained prominence in model-based reinforcement learning. The MuZero loss, which penalizes a model's value function prediction compared to the ground-truth value function, has been utilized in several prominent empirical works in the literature. However, theoretical investigation into its strengths and weaknesses is limited. In this paper, we analyze the family of value-aware model learning losses, w…

Biotech uses fermentation to produce milk proteins without cows
https://phys.org/news/2025-06-biotech-fermentation-proteins-cows.html

Synthesizing Sun-as-a-star flare spectra from high-resolution solar observations
M. De Wilde, A. G. M. Pietrow, M. K. Druett, A. Pastor Yabar, J. Koza, I. Kontogiannis, O. Andriienko, A. Berlicki, A. R. Brunvoll, J. de la Cruz Rodr\'iguez, J. T. Faber, R. Joshi, D. Kuridze, D. N\'obrega-Siverio, L. H. M. Rouppe van der Voort, J. Ryb\'ak, E. Scullion, A. M. Silva, Z. Vashalomidze, A. Vicente Ar\'evalo, A. Wi\'sniewska, R. Yadav, T. V. Zaqarashvili, J. Zbinden, E. S. …

Synthesizing Sun-as-a-star flare spectra from high-resolution solar observations
Spatially resolved observations of the Sun and the astronomical sample size of stellar bodies are the respective key strengths of solar and stellar observations. However, the large difference in object brightness between the Sun and other stars has led to distinctly different instrumentation and methodologies between the two fields. We produce and analyze synthetic full-disk spectra derived from 19 small area field-of-view optical observations of solar flares acquired by the Swedish 1-m Solar T…

Ammonia, Methane, Hydrogen and Methanol Produced in Remote Renewable Energy Hubs: a Comparative Quantitative Analysis
Antoine Larbanois, Victor Dachet, Antoine Dubois, Rapha\"el Fonteneau, Damien Ernst
https://arxiv.org/abs/2507.07681

Ammonia, Methane, Hydrogen and Methanol Produced in Remote Renewable Energy Hubs: a Comparative Quantitative Analysis
Remote renewable energy hubs (RREHs) for synthetic fuel production are engineering systems harvesting renewable energy where it is particularly abundant. They produce transportable synthetic fuels for export to distant load centers. This article aims to evaluate the production costs of different energy carriers, and includes a discussion on advantages and disadvantages in terms of technical performance. To do so, we extend the study of Berger et al., (2021) which focuses on methane (CH4) as ene…

SPBA: Utilizing Speech Large Language Model for Backdoor Attacks on Speech Classification Models
Wenhan Yao, Fen Xiao, Xiarun Chen, Jia Liu, YongQiang He, Weiping Wen
https://arxiv.org/abs/2506.08346

SPBA: Utilizing Speech Large Language Model for Backdoor Attacks on Speech Classification Models
Deep speech classification tasks, including keyword spotting and speaker verification, are vital in speech-based human-computer interaction. Recently, the security of these technologies has been revealed to be susceptible to backdoor attacks. Specifically, attackers use noisy disruption triggers and speech element triggers to produce poisoned speech samples that train models to become vulnerable. However, these methods typically create only a limited number of backdoors due to the inherent cons…

A Survey of Pun Generation: Datasets, Evaluations and Methodologies
Yuchen Su, Yonghua Zhu, Ruofan Wang, Zijian Huang, Diana Benavides-Prado, Michael Witbrock
https://arxiv.org/abs/2507.04793

A Survey of Pun Generation: Datasets, Evaluations and Methodologies
Pun generation seeks to creatively modify linguistic elements in text to produce humour or evoke double meanings. It also aims to preserve coherence and contextual appropriateness, making it useful in creative writing and entertainment across various media and contexts. Although pun generation has received considerable attention in computational linguistics, there is currently no dedicated survey that systematically reviews this specific area. To bridge this gap, this paper provides a comprehen…

New resonances at LHC
Alain Le Yaouanc, Fran\c{c}ois Richard
https://arxiv.org/abs/2506.09490 https://arxiv.org/pdf/2506.09490…

New resonances at LHC
Due to the large QCD background, the tt spectroscopy is poorly doing at LHC with the exception of a toponium type candidate observed with high statistical significance by CMS. Our previous work offers an alternate interpretation of this narrow resonance as a Kaluza Klein graviton. For heavy scalars, following ATLAS, we describe how the top loop contribution to the gluon-gluon fusion mechanisms could produce a dip rather than a bump in the mass distribution, which further complicates searches fo…

Data collected by Mauna Loa from 1958 onward let us clearly see the evidence of climate change for the first time.
The station samples the air and measures global CO₂ levels.
Charles Keeling and his successors used this data to produce the famous Keeling curve – a graph showing carbon dioxide levels increasing year after year.
But this precious record is in peril.
Donald Trump has decided to defund the observatory recording the data -- as well as the widespread US gre…

Mauna Loa Observatory captured the reality of climate change. The US plans to shut it down
Cutting data climate collection – as the Trump administration is now trying to do – is like breaking a thermometer to avoid knowing you’ve got a fever.

An Execution Model for RICE
Steven Libby
https://arxiv.org/abs/2506.05839 https://arxiv.org/pdf/2506.05839

An Execution Model for RICE
In this paper, we build on the previous work of the RICE compiler by giving its execution model. We show the restrictions to the FlatCurry language that were made to produce executable code, and present the execution model using operational semantics similar to Launchbury. Finally, we show that the execution model conforms with the standard operational semantics for Curry.

Dynamics of Subsurface Flows in Solar Active Regions During the May 2024 Storm
B Lekshmi, Sushanta Tripathy, Kiran Jain, Alexei Pevtsov
https://arxiv.org/abs/2506.08141

Dynamics of Subsurface Flows in Solar Active Regions During the May 2024 Storm
In May 2024, the Sun exhibited intense magnetic activity, marked by numerous high-intensity flares resulting from the interaction and merging of NOAA ARs 13664 and 13668 in the southern hemisphere and AR 13663 in the northern hemisphere. Notably, AR 13664 displayed an extended lifetime, remaining visible after a full solar rotation and continuing to produce significant flaring activity. In this study, we investigate the evolution of sub-photospheric plasma flows associated with these ARs during…

Assessing the Alignment of Audio Representations with Timbre Similarity Ratings
Haokun Tian, Stefan Lattner, Charalampos Saitis
https://arxiv.org/abs/2507.07764

Assessing the Alignment of Audio Representations with Timbre Similarity Ratings
Psychoacoustical so-called "timbre spaces" map perceptual similarity ratings of instrument sounds onto low-dimensional embeddings via multidimensional scaling, but suffer from scalability issues and are incapable of generalization. Recent results from audio (music and speech) quality assessment as well as image similarity have shown that deep learning is able to produce embeddings that align well with human perception while being largely free from these constraints. Although the existing human-…

This https://arxiv.org/abs/2506.02943 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csSE_…

A Multi-agent LLM-based JUit Test Generation with Strong Oracles
Unit testing plays a critical role in ensuring software correctness. However, writing unit tests manually is laborious, especially for strong typed languages like Java, motivating the need for automated approaches. Traditional methods primarily rely on search-based or randomized algorithms to generate tests that achieve high code coverage and produce regression oracles, which are derived from the program's current behavior rather than its intended functionality. Recent advances in large languag…

TRUST: Test-time Resource Utilization for Superior Trustworthiness
Haripriya Harikumar, Santu Rana
https://arxiv.org/abs/2506.06048 https://

TRUST: Test-time Resource Utilization for Superior Trustworthiness
Standard uncertainty estimation techniques, such as dropout, often struggle to clearly distinguish reliable predictions from unreliable ones. We attribute this limitation to noisy classifier weights, which, while not impairing overall class-level predictions, render finer-level statistics less informative. To address this, we propose a novel test-time optimization method that accounts for the impact of such noise to produce more reliable confidence estimates. This score defines a monotonic subs…

As Russia wages its full-scale war against Ukraine, one of its most important missile plants is quietly expanding -- bypassing international sanctions.
The Votkinsk Plant, also known as the Votkinskiy Plant,
— a strategic, state-owned facility serving Russia’s nuclear forces
— has hired thousands of new workers,
added new buildings,
and brought in advanced machinery to significantly increase its missile production.
Ukrainians have felt it firsthand.
Is…

Investigation: Russia expands strategic plant producing ICBMs — with China’s help
Key findings: * Despite international sanctions, Russia's strategic missile plant was able to import complex machinery to dramatically increase missile production. * The Kyiv Independent has identified the equipment supplied to the plant, as well as the supply chains, mostly from China. * We located the plant's new premises, built to house the new machinery. * We obtained a document confirming that the plant received an order to produce intercontinental missiles capable of reaching the U.S.

This https://arxiv.org/abs/2408.01583 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_hepp…

Superconducting Levitated Detector of Gravitational Waves
A magnetically levitated mass couples to gravity and can act as an effective gravitational wave detector. We show that a superconducting sphere levitated in a quadrupolar magnetic field, when excited by a gravitational wave, will produce magnetic field fluctuations that can be read out using a flux tunable microwave resonator. With a readout operating at the standard quantum limit, such a system could achieve broadband strain noise sensitivity of $h \lesssim 10^{-20}/\sqrt{\rm Hz}$ for frequenc…

Private GPTs for LLM-driven testing in software development and machine learning
Jakub Jagielski, Markus Abel
https://arxiv.org/abs/2506.06509 https://

Private GPTs for LLM-driven testing in software development and machine learning
In this contribution, we examine the capability of private GPTs to automatically generate executable test code based on requirements. More specifically, we use acceptance criteria as input, formulated as part of epics, or stories, which are typically used in modern development processes. This gives product owners, or business intelligence, respectively, a way to directly produce testable criteria through the use of LLMs. We explore the quality of the so-produced tests in two ways: i) directly b…

Ukraine announced on June 13 that its short-range Sapsan ballistic missile would go into mass production
-- a major development in Kyiv's ongoing efforts to domestically produce the weapons it needs to fight Russia's full-scale invasion.
As Ukraine faces growing challenges in securing weapons from Western partners, and Russia continues launching drone and missile strikes on civilians, the development of Ukraine’s missile production is seen as a critical step toward strength…

Tired of military aid delays, Ukraine has designed its own ballistic missile — and it’s already in mass-production
Ukraine announced on June 13 that it's short-range Sapsan ballistic missile would go into mass production, a major development in Kyiv's ongoing efforts to domestically produce the weapons it needs to fight Russia's full-scale invasion. As Ukraine faces growing challenges in securing weapons from Western partners, and Russia continues launching drone and missile strikes on civilians, the development of Ukraine’s missile production is seen as a critical step toward strengthening its defense c…