Tootfinder

Tesla nimmt Model X und Model S in Deutschland aus dem Programm
Tesla hat seine Oberklasse-Modelle Model S und X in Deutschland aus dem Konfigurator genommen. Damit bietet der US-Autobauer hier nur noch Model 3 und Y an.

Etwas mehr der heute besonders häufig geteilten #News:
Tesla nimmt Model X und Model S in Deutschland aus dem Programm

CoEx -- Co-evolving World-model and Exploration
Minsoo Kim, Seung-won Hwang
https://arxiv.org/abs/2507.22281 https://arxiv.org/pdf/2507.22281

CoEx -- Co-evolving World-model and Exploration
Planning in modern LLM agents relies on the utilization of LLM as an internal world model, acquired during pretraining. However, existing agent designs fail to effectively assimilate new observations into dynamic updates of the world model. This reliance on the LLM's static internal world model is progressively prone to misalignment with the underlying true state of the world, leading to the generation of divergent and erroneous plans. We introduce a hierarchical agent architecture, CoEx, in wh…

KARMA Approach supporting Development Process Reconstruction in Model-based Systems Engineering
Jiawei Li, Zan Liang, Guoxin Wang, Jinzhi Lu, Yan Yan, Shouxuan Wu, Hao Wang
https://arxiv.org/abs/2506.22037

KARMA Approach supporting Development Process Reconstruction in Model-based Systems Engineering
Model reconstruction is a method used to drive the development of complex system development processes in model-based systems engineering. Currently, during the iterative design process of a system, there is a lack of an effective method to manage changes in development requirements, such as development cycle requirements and cost requirements, and to realize the reconstruction of the system development process model. To address these issues, this paper proposes a model reconstruction method to…

2nd Conference on Multiscale Modeling of Complex Systems SCALES, at Johannes Gutenberg University Mainz, Sept 2-5, 2025.
#geology
https://model.uni-mainz.de/scales-conference-202…

A model structure on the category of A$_\infty$-categories with strict morphisms
Mattia Ornaghi
https://arxiv.org/abs/2506.22847 https://

A model structure on the category of A$_\infty$-categories with strict morphisms
We prove that the category of (strictly unital) A$_\infty$-categories, linear over a commutative ring $R$, with strict A$_\infty$-morphisms has a cofibrantly generated model structure. In this model structure every object is fibrant and the cofibrant objects have cofibrant morphisms. As a consequence we prove that the semi-free A$_\infty$-categories (resp. resolutions) are cofibrant objects (resp. resolution) in this model structure.

Middo: Model-Informed Dynamic Data Optimization for Enhanced LLM Fine-Tuning via Closed-Loop Learning
Zinan Tang, Xin Gao, Qizhi Pei, Zhuoshi Pan, Mengzhang Cai, Jiang Wu, Conghui He, Lijun Wu
https://arxiv.org/abs/2508.21589

Middo: Model-Informed Dynamic Data Optimization for Enhanced LLM Fine-Tuning via Closed-Loop Learning
Supervised Fine-Tuning (SFT) Large Language Models (LLM) fundamentally rely on high-quality training data. While data selection and data synthesis are two common strategies to improve data quality, existing approaches often face limitations in static dataset curation that fail to adapt to evolving model capabilities. In this paper, we introduce Middo, a self-evolving Model-informed dynamic data optimization framework that uses model-aware data selection and context-preserving data refinement. U…

Assessing security and privacy
I sometimes get asked how I think about and evaluate the security and privacy of Internet services and applications. For me, a number of factors come into focus when assessing the privacy of an application or service. Some of the factors are technological and some of them are human, social and organizational. Additionally, some of the factors are critical to ensure privacy and some are important but less critical.
 
I think we need a new model …

I Stolenly Swear That I Am Up to (No) Good: Design and Evaluation of Model Stealing Attacks
Daryna Oliynyk, Rudolf Mayer, Kathrin Grosse, Andreas Rauber
https://arxiv.org/abs/2508.21654

I Stolenly Swear That I Am Up to (No) Good: Design and Evaluation of Model Stealing Attacks
Model stealing attacks endanger the confidentiality of machine learning models offered as a service. Although these models are kept secret, a malicious party can query a model to label data samples and train their own substitute model, violating intellectual property. While novel attacks in the field are continually being published, their design and evaluations are not standardised, making it challenging to compare prior works and assess progress in the field. This paper is the first to address…

Motion Tracking with Muscles: Predictive Control of a Parametric Musculoskeletal Canine Model
Vittorio La Barbera, Steven Bohez, Leonard Hasenclever, Yuval Tassa, John R. Hutchinson
https://arxiv.org/abs/2506.23768

Motion Tracking with Muscles: Predictive Control of a Parametric Musculoskeletal Canine Model
We introduce a novel musculoskeletal model of a dog, procedurally generated from accurate 3D muscle meshes. Accompanying this model is a motion capture-based locomotion task compatible with a variety of control algorithms, as well as an improved muscle dynamics model designed to enhance convergence in differentiable control frameworks. We validate our approach by comparing simulated muscle activation patterns with experimentally obtained electromyography (EMG) data from previous canine locomoti…

Hypernetworks for Model-Heterogeneous Personalized Federated Learning
Chen Zhang, Husheng Li, Xiang Liu, Linshan Jiang, Danxin Wang
https://arxiv.org/abs/2507.22330 https://

Hypernetworks for Model-Heterogeneous Personalized Federated Learning
Recent advances in personalized federated learning have focused on addressing client model heterogeneity. However, most existing methods still require external data, rely on model decoupling, or adopt partial learning strategies, which can limit their practicality and scalability. In this paper, we revisit hypernetwork-based methods and leverage their strong generalization capabilities to design a simple yet effective framework for heterogeneous personalized federated learning. Specifically, we…

"Certains signes d'effondrement de modèles d'IA commencent Š apparaître"
"Nous allons investir de plus en plus dans l'IA, jusqu'au moment où les modèles vont s'effondrer et que les réponses de l'IA seront si mauvaises que même un PDG au cerveau malade ne peut l'ignorer."

Google DeepMind unveils AlphaEarth Foundations, an AI model that unifies petabytes of Earth observation data and helps scientists create detailed maps on demand (Google DeepMind)
https://deepmind.google/discover/blog/alphaearth-found…

AlphaEarth Foundations helps map our planet in unprecedented detail
New AI model integrates petabytes of Earth observation data to generate a unified data representation that revolutionizes global mapping and monitoring

Non-classicality of two-qubit quantum collision model: non-Markovian effects
Jai Lalita, Subhashish Banerjee
https://arxiv.org/abs/2506.23818 https://

Non-classicality of two-qubit quantum collision model: non-Markovian effects
We investigate a two-qubit quantum system in contact with an environment modeled by a microscopic collision model with added ancilla-ancilla collisions in the non-Markovian regime. Two schemes of the two-qubit collision model with carried-forward correlations are introduced. In one scheme, a single stream of ancillae interacts with only one of the qubits of the two-qubit system; in the other, both the qubits interact with two independent sequences of ancillae, which could be at the same or diff…

Algorithm for computing the partition function of the Potts model for SP-graphs
Sofya Mukhamedzhanova, Bulat Sabirov, Amir Mukhamedzhanov
https://arxiv.org/abs/2507.22579 https:…

Algorithm for computing the partition function of the Potts model for SP-graphs
The q-state Potts model is a fundamental framework in statistical physics and graph theory, with its partition function encoding rich information about spin configurations. The multivariate Tutte polynomial (known as the partition function of the Potts model) can be defined on an arbitrary finite graph $G$ and encodes a lot of important combinatorial information about the graph. As a special case, it contains the familiar Tutte polynomial with two variables and, consequently, its specialization…

Generic derivations on algebraically bounded structures II Model theoretical properties
Fornasiero Antongiulio, Terzo Giuseppina
https://arxiv.org/abs/2507.22181 https://…

Generic derivations on algebraically bounded structures II Model theoretical properties
Let T be an algebraically bounded theory. We consider the $L(\barδ)$-expansions of T by a tuple $\bar δ$ of derivations (which may be commuting or not). We investigate the model completion of either of the above theories, whose existence has been established in [FT:24], with particular attention to its model-theoretic properties, including $ω$-stability, simplicity, open core, and elimination of imaginaries.

QPART: Adaptive Model Quantization and Dynamic Workload Balancing for Accuracy-aware Edge Inference
Xiangchen Li, Saeid Ghafouri, Bo Ji, Hans Vandierendonck, Deepu John, Dimitrios S. Nikolopoulos
https://arxiv.org/abs/2506.23934

QPART: Adaptive Model Quantization and Dynamic Workload Balancing for Accuracy-aware Edge Inference
As machine learning inferences increasingly move to edge devices, adapting to diverse computational capabilities, hardware, and memory constraints becomes more critical. Instead of relying on a pre-trained model fixed for all future inference queries across diverse edge devices, we argue that planning an inference pattern with a request-specific model tailored to the device's computational capacity, accuracy requirements, and time constraints is more cost-efficient and robust to diverse scenari…

Model-free Forecasting of Rogue Waves using Reservoir Computing
Abrari Noor Hasmi, Hadi Susanto
https://arxiv.org/abs/2506.21918 https://

Model-free Forecasting of Rogue Waves using Reservoir Computing
Recent research has demonstrated Reservoir Computing's capability to model various chaotic dynamical systems, yet its application to Hamiltonian systems remains relatively unexplored. This paper investigates the effectiveness of Reservoir Computing in capturing rogue wave dynamics from the nonlinear Schrödinger equation, a challenging Hamiltonian system with modulation instability. The model-free approach learns from breather simulations with five unstable modes. A properly tuned parallel Echo…

Heavy Quarkonia in the Light-Front Quark Model
Muhammad Ridwan, Ahmad Jafar Arifi, Terry Mart
https://arxiv.org/abs/2508.21145 https://arxiv.org/pdf/2508.2…

Heavy Quarkonia in the Light-Front Quark Model
We present recent progress in constructing heavy quarkonia light-front wave functions from the light-front quark model, a relativistic framework well-suited for describing partonic structure and hadronic form factors. In this proceeding, we discuss the recent application of the model to the $M1$ radiative transition such as $J/ψ\to η_c γ$ among other transitions. Additionally, we compare our results on the transition form factor with available lattice QCD data. These show our efforts to deep…

Pendulum Model of Spiking Neurons
Joy Bose
https://arxiv.org/abs/2507.22146 https://arxiv.org/pdf/2507.22146

Pendulum Model of Spiking Neurons
We propose a biologically inspired model of spiking neurons based on the dynamics of a damped, driven pendulum. Unlike traditional models such as the Leaky Integrate-and-Fire (LIF) neurons, the pendulum neuron incorporates second-order, nonlinear dynamics that naturally give rise to oscillatory behavior and phase-based spike encoding. This model captures richer temporal features and supports timing-sensitive computations critical for sequence processing and symbolic learning. We present an anal…

Testing parametric additive time-varying GARCH models
Niklas Ahlgren, Alexander Back, Timo Ter\"asvirta
https://arxiv.org/abs/2506.23821 https://

Testing parametric additive time-varying GARCH models
We develop misspecification tests for building additive time-varying (ATV-)GARCH models. In the model, the volatility equation of the GARCH model is augmented by a deterministic time-varying intercept modeled as a linear combination of logistic transition functions. The intercept is specified by a sequence of tests, moving from specific to general. The first test is the test of the standard stationary GARCH model against an ATV-GARCH model with one transition. The alternative model is unidentif…

The completed Kirillov model and local-global compatibility for functions on Igusa varieties
Sean Howe
https://arxiv.org/abs/2506.24089 https://

The completed Kirillov model and local-global compatibility for functions on Igusa varieties
We describe the cuspidal functions $\mathbb{V}_b^{\mathrm{cusp}}$ on the ordinary Caraiani-Scholze Igusa variety for $\mathrm{GL}_2$ as a completion of the smooth Kirillov model for classical cuspidal modular forms, and identify a variant of Hida's ordinary $p$-adic modular forms with the coinvariants of an action of $\tildeμ_{p^\infty}$ on $\mathbb{V}_b^{\mathrm{cusp}}$. As a consequence of these results, we establish a weak local-global compatibility theorem for eigenspaces in $\mathbb{V}_b^…

On the reggeon model with the pomeron and odderon: singularities with non-zero masses
M. A. Braun (Saint-Petersburg State University, Russia), E. M. Kuzminskii (Petersburg Nuclear Physics Institute, Russia), M. I. Vyazovsky (Saint-Petersburg State University, Russia)
https://arxiv.org/abs/2507.23086

On the reggeon model with the pomeron and odderon: singularities with non-zero masses
The Regge-Gribov model of the pomeron and odderon in the non-trivial transverse space is studied by the renormalization group technique in the single loop approximation. The pomeron and odderon are taken to have different bare intercepts and slopes. The behaviour when the intercepts move from below to their critical values compatible with the Froissart limitation is studied. The singuarities in the form of non-trivial branch points indicating a phase transition are found in the vicinity of five…

Steinberg-Guinan strength model for rhenium
Damian C. Swift
https://arxiv.org/abs/2507.22228 https://arxiv.org/pdf/2507.22228

Steinberg-Guinan strength model for rhenium
Rhenium, Re, is used as an x-ray shield in laser-driven material property experiments, where its strength at high pressures can be a consideration in the design, modeling, and interpretation. We present a Steinberg-Guinan (SG) strength model for Re, tailored for use in high-pressure dynamic loading simulations. Parameters for the SG model were derived from recent atom-in-jellium predictions of the shear modulus under compression and experimental data on work-hardening from rolled-bar studies. T…

Consistency of Feature Attribution in Deep Learning Architectures for Multi-Omics
Daniel Claborne, Javier Flores, Samantha Erwin, Luke Durell, Rachel Richardson, Ruby Fore, Lisa Bramer
https://arxiv.org/abs/2507.22877

Consistency of Feature Attribution in Deep Learning Architectures for Multi-Omics
Machine and deep learning have grown in popularity and use in biological research over the last decade but still present challenges in interpretability of the fitted model. The development and use of metrics to determine features driving predictions and increase model interpretability continues to be an open area of research. We investigate the use of Shapley Additive Explanations (SHAP) on a multi-view deep learning model applied to multi-omics data for the purposes of identifying biomolecules…

Active spin model for cell assemblies on 1D substrates
Harshal Potdar, Ignacio Pagonabarraga, Sudipto Muhuri
https://arxiv.org/abs/2507.22639 https://arxiv…

Active spin model for cell assemblies on 1D substrates
The experimental use of micropatterned quasi-1D substrates has emerged as an useful experimental tool to study the nature of cell-cell interactions and gain insight on collective behaviour of cell colonies. Inspired by these experiments, we propose an active spin model to investigate the emergent properties of the cell assemblies. The lattice gas model incorporates the interplay of self-propulsion, polarity directional switching, intra-cellular attraction, and contact Inhibition Locomotion (CIL…

Teaching a Language Model to Speak the Language of Tools
Simeon Emanuilov
https://arxiv.org/abs/2506.23394 https://arxiv.org/pdf/2506…

Teaching a Language Model to Speak the Language of Tools
External tool integration through function-calling is essential for practical language model applications, yet most multilingual models lack reliable tool-use capabilities in non-English languages. Even state-of-the-art multilingual models struggle with determining when to use tools and generating the structured outputs required for function calls, often exhibiting language confusion when prompted in lower-resource languages. This work presents a methodology for adapting existing language model…

Improving Spatio-Temporal Accuracy of the Stochastic Particle Fokker-Planck Model
Joonbeom Kim, Eunji Jun
https://arxiv.org/abs/2506.23142 https://

Improving Spatio-Temporal Accuracy of the Stochastic Particle Fokker-Planck Model
Accurate prediction of rarefied gas flows is important for space vehicle design, particularly in rarefied regimes where the Navier-Stokes equations are no more valid. While the direct simulation Monte Carlo (DSMC) method acts as a numerical solver for rarefied gas flows, it becomes inefficient when dealing with near-continuum regimes. The Fokker-Planck (FP) model improves computational efficiency by approximating particle collisions as a drift-diffusion process. The FP model has been extended t…

Real-time Generation of Various Types of Nodding for Avatar Attentive Listening System
Kazushi Kato, Koji Inoue, Divesh Lala, Keiko Ochi, Tatsuya Kawahara
https://arxiv.org/abs/2507.23298

Real-time Generation of Various Types of Nodding for Avatar Attentive Listening System
In human dialogue, nonverbal information such as nodding and facial expressions is as crucial as verbal information, and spoken dialogue systems are also expected to express such nonverbal behaviors. We focus on nodding, which is critical in an attentive listening system, and propose a model that predicts both its timing and type in real time. The proposed model builds on the voice activity projection (VAP) model, which predicts voice activity from both listener and speaker audio. We extend it …

Eerste AI-model in modeblad Vogue veroorzaakt veel discussie: 'Kunstmatige intelligentie haalt creativiteit en ziel eruit' | #EenVandaag

Analysis of quantities determining the critical inverse temperature in the annealed Potts model with Pareto vertex weights
A. J. E. M. Janssen
https://arxiv.org/abs/2508.21409 h…

Analysis of quantities determining the critical inverse temperature in the annealed Potts model with Pareto vertex weights
We consider in this work the crucial quantity $t_c$ that determines the critical inverse temperature $β_c$ in the $q$-state Potts model on sparse rank-1 random graphs where the vertices are equipped with a Pareto weight density $(τ-1)\,w^{-τ}\,{\cal X}_{[1,\infty)}(w)$. It is shown in \cite{ref1} that this $t_c$ is the unique positive zero of a function ${\cal K}$ that is obtained by an appropriate combination of the stationarity condition and the criticality condition for the case the exter…

Half-Physics: Enabling Kinematic 3D Human Model with Physical Interactions
Li Siyao, Yao Feng, Omid Tehari, Chen Change Loy, Michael J. Black
https://arxiv.org/abs/2507.23778 ht…

Half-Physics: Enabling Kinematic 3D Human Model with Physical Interactions
While current general-purpose 3D human models (e.g., SMPL-X) efficiently represent accurate human shape and pose, they lacks the ability to physically interact with the environment due to the kinematic nature. As a result, kinematic-based interaction models often suffer from issues such as interpenetration and unrealistic object dynamics. To address this limitation, we introduce a novel approach that embeds SMPL-X into a tangible entity capable of dynamic physical interactions with its surround…

Model-theoretic Forcing in Transition Algebra
Go Hashimoto, Daniel G\u{a}in\u{a}
https://arxiv.org/abs/2506.22828 https://arxiv.org/p…

Model-theoretic Forcing in Transition Algebra
We study Löwenheim-Skolem and Omitting Types theorems in Transition Algebra, a logical system obtained by enhancing many sorted first-order logic with features from dynamic logic. The sentences we consider include compositions, unions, and transitive closures of transition relations, which are treated similarly to actions in dynamic logics to define necessity and possibility operators. We show that Upward Löwenheim-Skolem theorem, any form of compactness, and joint Robinson consistency proper…

Minimal regular normal crossings models of superelliptic curves
Andrew Obus, Padmavathi Srinivasan
https://arxiv.org/abs/2506.24091 https://

Minimal regular normal crossings models of superelliptic curves
Let $K$ be a complete discretely valued field with perfect residue field $k$. If $X \to \mathbb{P}^1_K$ is a $\mathbb{Z}/d$-cover with $\text{char } k \nmid d$, we compute the minimal regular normal crossings model $\mathcal{X}$ of $X$ as the normalization of an explicit normal model $\mathcal{Y}$ of $\mathbb{P}^1_K$ in $K(X)$. The model $\mathcal{Y}$ is given using Mac Lane's description of discrete valuations on the rational function field $K(\mathbb{P}^1)$.

Inference in a generalized Bradley-Terry model for paired comparisons with covariates and a growing number of subjects
Ting Yan
https://arxiv.org/abs/2507.22472 https://

Inference in a generalized Bradley-Terry model for paired comparisons with covariates and a growing number of subjects
Motivated by the home-field advantage in sports, we propose a generalized Bradley--Terry model that incorporates covariate information for paired comparisons. It has an $n$-dimensional merit parameter $\bsβ$ and a fixed-dimensional regression coefficient $\bsγ$ for covariates. When the number of subjects $n$ approaches infinity and the number of comparisons between any two subjects is fixed, we show the uniform consistency of the maximum likelihood estimator (MLE) $(\widehat{\bsβ}, \widehat{…

General Signal Model and Capacity Limit for Rydberg Quantum Information System
Jieao Zhu, Linglong Dai
https://arxiv.org/abs/2506.23455 https://

General Signal Model and Capacity Limit for Rydberg Quantum Information System
Rydberg atomic receivers represent a transformative approach to achieving high-sensitivity, broadband, and miniaturized radio frequency (RF) reception. However, existing static signal models for Rydberg atomic receivers rely on the steady-state assumption of atomic quantum states, which cannot fully describe the signal reception process of dynamic signals. To fill in this gap, in this paper, we present a general model to compute the dynamic signal response of Rydberg atomic receivers in closed …

Rethinking Layer-wise Model Merging through Chain of Merges
Pietro Buzzega, Riccardo Salami, Angelo Porrello, Simone Calderara
https://arxiv.org/abs/2508.21421 https://

Rethinking Layer-wise Model Merging through Chain of Merges
Fine-tuning pretrained models has become a standard pathway to achieve state-of-the-art performance across a wide range of domains, leading to a proliferation of task-specific model variants. As the number of such specialized modules in-creases, merging them into a unified model without retraining has become a critical challenge. Existing merging techniques often rely on interference heuristics,importance weighting, or activation matching while treating each layer independently, thereby failing…

Economic Model Predictive Control with a Non-Fixed Reference Trajectory for Optimal Microgrid Dispatch
Avik Ghosh, Adil Khurram, Jan Kleissl, Sonia Martinez
https://arxiv.org/abs/2506.22406

Economic Model Predictive Control with a Non-Fixed Reference Trajectory for Optimal Microgrid Dispatch
Economic Model Predictive Control (EMPC), instead of stabilizing a reference trajectory/state in the objective function like a Tracking MPC, optimizes the economic performance over the prediction horizon, making it attractive for economical microgrid (MG) dispatch. However, the demand charge component in the monthly electricity cost, make it difficult to be encapsulated in additive stage costs, and can make solutions violate the principle of optimality if naively introduced in the objective fun…

WaveLLDM: Design and Development of a Lightweight Latent Diffusion Model for Speech Enhancement and Restoration
Kevin Putra Santoso, Rizka Wakhidatus Sholikah, Raden Venantius Hari Ginardi
https://arxiv.org/abs/2508.21153

WaveLLDM: Design and Development of a Lightweight Latent Diffusion Model for Speech Enhancement and Restoration
High-quality audio is essential in a wide range of applications, including online communication, virtual assistants, and the multimedia industry. However, degradation caused by noise, compression, and transmission artifacts remains a major challenge. While diffusion models have proven effective for audio restoration, they typically require significant computational resources and struggle to handle longer missing segments. This study introduces WaveLLDM (Wave Lightweight Latent Diffusion Model),…

Advanced Deep Learning Techniques for Automated Segmentation of Type B Aortic Dissections
Hao Xu, Ruth Lim, Brian E. Chapman
https://arxiv.org/abs/2506.22222

Advanced Deep Learning Techniques for Automated Segmentation of Type B Aortic Dissections
Purpose: Aortic dissections are life-threatening cardiovascular conditions requiring accurate segmentation of true lumen (TL), false lumen (FL), and false lumen thrombosis (FLT) from CTA images for effective management. Manual segmentation is time-consuming and variable, necessitating automated solutions. Materials and Methods: We developed four deep learning-based pipelines for Type B aortic dissection segmentation: a single-step model, a sequential model, a sequential multi-task model, and an…

Quantifying Label-Induced Bias in Large Language Model Self- and Cross-Evaluations
Muskan Saraf, Sajjad Rezvani Boroujeni, Justin Beaudry, Hossein Abedi, Tom Bush
https://arxiv.org/abs/2508.21164

Quantifying Label-Induced Bias in Large Language Model Self- and Cross-Evaluations
Large language models (LLMs) are increasingly used to evaluate outputs, yet their judgments may be influenced. This study examines bias in self- and cross-model evaluations by ChatGPT, Gemini, and Claude under four conditions: no labels, true labels, and two false-label scenarios. Blog posts authored by each model were evaluated by all three using both overall preference voting and quality ratings for Coherence, Informativeness, and Conciseness, with all scores expressed as percentages for dire…

X-ray polarization in radio-quiet AGN: Insights from the wedge corona model using Monte Carlo simulations
D. Tagliacozzo, S. Bianchi. V. Gianolli, A. Gnarini, A. Marinucci, G. Matt, F. Ursini, W. Zhang
https://arxiv.org/abs/2508.21203

X-ray polarization in radio-quiet AGN: Insights from the wedge corona model using Monte Carlo simulations
In this study, we present novel calculations of X-ray polarization from radio-quiet and unobscured active galactic nuclei (AGNs) using the Monte Carlo code MONK, which includes all general and special relativity effects. Our geometric model, referred to as the ``wedge corona'', features a homogeneous cloud of electrons characterized by an aspect ratio of h/r and a radius that extends down to the innermost stable circular orbit around the central black hole (BH). Adopting the physical parameters…

Model-Based Diagnosis: Automating End-to-End Diagnosis of Network Failures
Changrong Wu, Yiyao Yu, Myungjin Lee, Jayanth Srinivasa, Ennan Zhai, George Varghese, Yuval Tamir
https://arxiv.org/abs/2506.23083

Model-Based Diagnosis: Automating End-to-End Diagnosis of Network Failures
Fast diagnosis and repair of enterprise network failures is critically important since disruptions cause major business impacts. Prior works focused on diagnosis primitives or procedures limited to a subset of the problem, such as only data plane or only control plane faults. This paper proposes a new paradigm, model-based network diagnosis, that provides a systematic way to derive automated procedures for identifying the root cause of network failures, based on reports of end-to-end user-level…

Dimension and model reduction approaches for linear Bayesian inverse problems with rank-deficient prior covariances
Josie K\"onig, Elizabeth Qian, Melina A. Freitag
https://arxiv.org/abs/2506.23892

Dimension and model reduction approaches for linear Bayesian inverse problems with rank-deficient prior covariances
Bayesian inverse problems use observed data to update a prior probability distribution for an unknown state or parameter of a scientific system to a posterior distribution conditioned on the data. In many applications, the unknown parameter is high-dimensional, making computation of the posterior expensive due to the need to sample in a high-dimensional space and the need to evaluate an expensive high-dimensional forward model relating the unknown parameter to the data. However, inverse problem…

Discover how the energy as a service model is revolutionizing industrial energy procurement. #climatechange #climatesolutions #climate

How the energy as a service model is disrupting industrial energy procurement
Fast Company Executive Board member Jay Bhatty highlights how EaaS is changing traditional utility models.

Quantum Variational Transformer Model for Enhanced Cancer Classification
Don Roosan, Rubayat Khan, Md Rahatul Ashakin, Tiffany Khou, Saif Nirzhor, Mohammad Rifat Haider
https://arxiv.org/abs/2506.21641

Quantum Variational Transformer Model for Enhanced Cancer Classification
Accurate prediction of cancer type and primary tumor site is critical for effective diagnosis, personalized treatment, and improved outcomes. Traditional models struggle with the complexity of genomic and clinical data, but quantum computing offers enhanced computational capabilities. This study develops a hybrid quantum-classical transformer model, incorporating quantum attention mechanisms via variational quantum circuits (VQCs) to improve prediction accuracy. Using 30,000 anonymized cancer s…

Genesis AI, which generates its own synthetic data for training an AI model for robots, emerges from stealth with a $105M seed co-led by Eclipse and Khosla (Marina Temkin/TechCrunch)
https://techcrunch.com/2025/07/01/gene

Genesis AI launches with $105M seed funding from Eclipse, Khosla to build AI models for robots | TechCrunch
Genesis AI, which aims to build a foundational model for powering all kinds of robots, has emerged from stealth with $105M in seed funding.

Robust Convex Model Predictive Control with collision avoidance guarantees for robot manipulators
Bernhard Wullt, Johannes K\"ohler, Per Mattsson, Mikeal Norrl\"of, Thomas B. Sch\"on
https://arxiv.org/abs/2508.21677

Robust Convex Model Predictive Control with collision avoidance guarantees for robot manipulators
Industrial manipulators are normally operated in cluttered environments, making safe motion planning important. Furthermore, the presence of model-uncertainties make safe motion planning more difficult. Therefore, in practice the speed is limited in order to reduce the effect of disturbances. There is a need for control methods that can guarantee safe motions that can be executed fast. We address this need by suggesting a novel model predictive control (MPC) solution for manipulators, where our…

Privacy-Preserving Federated Learning Scheme with Mitigating Model Poisoning Attacks: Vulnerabilities and Countermeasures
Jiahui Wu, Fucai Luo, Tiecheng Sun, Haiyan Wang, Weizhe Zhang
https://arxiv.org/abs/2506.23622

Privacy-Preserving Federated Learning Scheme with Mitigating Model Poisoning Attacks: Vulnerabilities and Countermeasures
The privacy-preserving federated learning schemes based on the setting of two honest-but-curious and non-colluding servers offer promising solutions in terms of security and efficiency. However, our investigation reveals that these schemes still suffer from privacy leakage when considering model poisoning attacks from malicious users. Specifically, we demonstrate that the privacy-preserving computation process for defending against model poisoning attacks inadvertently leaks privacy to one of t…

A Survey on Autonomy-Induced Security Risks in Large Model-Based Agents
Hang Su, Jun Luo, Chang Liu, Xiao Yang, Yichi Zhang, Yinpeng Dong, Jun Zhu
https://arxiv.org/abs/2506.23844

A Survey on Autonomy-Induced Security Risks in Large Model-Based Agents
Recent advances in large language models (LLMs) have catalyzed the rise of autonomous AI agents capable of perceiving, reasoning, and acting in dynamic, open-ended environments. These large-model agents mark a paradigm shift from static inference systems to interactive, memory-augmented entities. While these capabilities significantly expand the functional scope of AI, they also introduce qualitatively novel security risks - such as memory poisoning, tool misuse, reward hacking, and emergent mi…

Axioms for Model Fidelity Evaluation
Evan Taylor, Edward Louis, Gregory Mocko
https://arxiv.org/abs/2507.23020 https://arxiv.org/pdf/2507.23020

Axioms for Model Fidelity Evaluation
Digital engineering has transformed the design and development process. However, the utility of digital engineering is fundamentally dependent on the assumption that a simulation provides information consistent with reality. This relationship is described as model fidelity. Despite the widespread use of the term, existing definitions of model fidelity often lack formal rigor in practical application, which leaves ambiguity in how this similarity should be evaluated. This paper presents seven fu…

Plasmon-assisted photoelectron emission in a model cluster using time-dependent density functional theory and the time-dependent surface-flux method
Mikhail Bednov, Waqas Pervez, Ingo Barke, Dieter Bauer
https://arxiv.org/abs/2507.22709

Plasmon-assisted photoelectron emission in a model cluster using time-dependent density functional theory and the time-dependent surface-flux method
We investigate plasmon-assisted photoelectron emission using a one-dimensional time-dependent density-functional theory (TDDFT) model. Photoelectron spectra are computed with the time-dependent surface-flux (t-SURFF) method. In addition to the expected above-threshold ionization (ATI) comb, we observe peaks that arise from long-lived plasmon oscillations and the associated electron emission occurring after the laser pulse. We further analyze the positions of these peaks and their scaling behavi…

Removing cutoff artifacts in the NJL model by a Renormalization Group consistent treatment
Hosein Gholami, Marco Hofmann, Michael Buballa
https://arxiv.org/abs/2508.21735 https:…

Removing cutoff artifacts in the NJL model by a Renormalization Group consistent treatment
We summarize how a renormalization-group (RG)-consistent treatment removes well-known artifacts in NJL-model descriptions of color-superconducting quark matter. We introduce two RG-consistent schemes, "minimal" and "massless", and present analytic solutions for the diquark gap at $T=0$ and for the phase boundary $T_c(μ)$ in symmetric massless matter, representing the high-density limit of the model. We compare the pairing gaps, phase diagram, and speed of sound with results obtained using conv…

Repair Ingredients Are All You Need: Improving Large Language Model-Based Program Repair via Repair Ingredients Search
Jiayi Zhang, Kai Huang, Jian Zhang, Yang Liu, Chunyang Chen
https://arxiv.org/abs/2506.23100

Repair Ingredients Are All You Need: Improving Large Language Model-Based Program Repair via Repair Ingredients Search
Automated Program Repair (APR) techniques aim to automatically fix buggy programs. Among these, Large Language Model-based (LLM-based) approaches have shown great promise. Recent advances demonstrate that directly leveraging LLMs can achieve leading results. However, these techniques remain suboptimal in generating contextually relevant and accurate patches, as they often overlook repair ingredients crucial for practical program repair. In this paper, we propose ReinFix, a novel framework that …

No covering with nowhere dense \textsf{P}-sets in the Cohen model
Alan Dow, Osvaldo Guzm\'an
https://arxiv.org/abs/2507.22476 https://arxiv.org/pdf/250…

No covering with nowhere dense \textsf{P}-sets in the Cohen model
We prove that if less than $\aleph_ω$-many Cohen reals are added to a model of \textsf{CH}, then $ω^{\ast}$ can not be covered by nowhere dense \textsf{P}-sets (equivalently, there is an ultrafilter on $ω$ that does not contain a tower).

Prime graphical parking functions and strongly recurrent configurations of the Abelian sandpile model
Thomas Selig, Haoyue Zhu
https://arxiv.org/abs/2506.23237

Prime graphical parking functions and strongly recurrent configurations of the Abelian sandpile model
This work investigates the duality between two discrete dynamical processes: parking functions, and the Abelian sandpile model (ASM). Specifically, we are interested in the extension of classical parking functions, called $G$-parking functions, introduced by Postnikov and Shapiro in 2004. $G$-parking functions are in bijection with recurrent configurations of the ASM on $G$. In this work, we define a notion of prime $G$-parking functions. These are parking functions that are in a sense "indecom…

Anomalous dimensions and critical exponents for the Gross-Neveu-Yukawa model at five loops
J. A. Gracey, A. Maier, P. Marquard, Y. Schr\"oder
https://arxiv.org/abs/2507.22594

Anomalous dimensions and critical exponents for the Gross-Neveu-Yukawa model at five loops
We renormalize the Gross-Neveu-Yukawa model with an $O(N)$ symmetry to $\mathcal{O}(ε^5)$ in $d=4-ε$ dimensions and determine the anomalous dimensions of the fermion and scalar fields, $β$-functions as well as the scalar field's mass operator. These are used to construct several $N$ dependent critical exponents relevant for quantum transitions in semi-metals and in particular those connected with graphene in three dimensions when $N=2$. Improved exponent estimates for scalar fermion transiti…

Energetic variational modeling of active nematics: coupling the Toner-Tu model with ATP hydrolysis
Yiwei Wang
https://arxiv.org/abs/2506.23043 https://

Energetic variational modeling of active nematics: coupling the Toner-Tu model with ATP hydrolysis
We present a thermodynamically consistent energetic variational model for active nematics driven by ATP hydrolysis, with a focus on the coupling between chemical reactions and mechanical dynamics. Extending the classical Toner-Tu framework, we introduce a chemo-mechanical coupling mechanism in which the self-advection and polarization dynamics are modulated by the ATP hydrolysis rate. The model is derived using an energetic variational approach that integrates both chemical free energy and mech…

IRanker: Towards Ranking Foundation Model
Tao Feng, Zhigang Hua, Zijie Lei, Yan Xie, Shuang Yang, Bo Long, Jiaxuan You
https://arxiv.org/abs/2506.21638 htt…

IRanker: Towards Ranking Foundation Model
Ranking tasks are ubiquitous, encompassing applications such as recommendation systems, LLM routing, and item re-ranking. We propose to unify these tasks using a single ranking foundation model (FM), as it eliminates the need for designing different models for each specific ranking task. However, unlike general supervision tasks in LLMs, ranking tasks do not have clear labels for supervision, posing great challenges to developing a ranking FM. To overcome these challenges, we propose IRanker, a…

Towards Building Private LLMs: Exploring Multi-Node Expert Parallelism on Apple Silicon for Mixture-of-Experts Large Language Model
Mu-Chi Chen, Po-Hsuan Huang, Xiangrui Ke, Chia-Heng Tu, Chun Jason Xue, Shih-Hao Hung
https://arxiv.org/abs/2506.23635

Towards Building Private LLMs: Exploring Multi-Node Expert Parallelism on Apple Silicon for Mixture-of-Experts Large Language Model
Large Language Models (LLMs) have revolutionized Artificial Intelligence (AI) with significant advancements such as OpenAI's ChatGPT, Meta's Llama, and Databricks' DBRX. This paper addresses the cost and scalability challenges encountered when constructing private LLM systems for personal or small group services, as aimed by Apple Intelligence. A Mac Studio cluster with Apple's M2 Ultra chips is established as a cost-efficient solution to host and accelerate the pretrained DBRX model with the M…

CAD2DMD-SET: Synthetic Generation Tool of Digital Measurement Device CAD Model Datasets for fine-tuning Large Vision-Language Models
Jo\~ao Valente, Atabak Dehban, Rodrigo Ventura
https://arxiv.org/abs/2508.21732

CAD2DMD-SET: Synthetic Generation Tool of Digital Measurement Device CAD Model Datasets for fine-tuning Large Vision-Language Models
Recent advancements in Large Vision-Language Models (LVLMs) have demonstrated impressive capabilities across various multimodal tasks. They continue, however, to struggle with trivial scenarios such as reading values from Digital Measurement Devices (DMDs), particularly in real-world conditions involving clutter, occlusions, extreme viewpoints, and motion blur; common in head-mounted cameras and Augmented Reality (AR) applications. Motivated by these limitations, this work introduces CAD2DMD-SE…

Multimodal Late Fusion Model for Problem-Solving Strategy Classification in a Machine Learning Game
Clemens Witt, Thiemo Leonhardt, Nadine Bergner, Mareen Grillenberger
https://arxiv.org/abs/2507.22426

Multimodal Late Fusion Model for Problem-Solving Strategy Classification in a Machine Learning Game
Machine learning models are widely used to support stealth assessment in digital learning environments. Existing approaches typically rely on abstracted gameplay log data, which may overlook subtle behavioral cues linked to learners' cognitive strategies. This paper proposes a multimodal late fusion model that integrates screencast-based visual data and structured in-game action sequences to classify students' problem-solving strategies. In a pilot study with secondary school students (N=149) p…

Stochastic forest transition model dynamics and parameter estimation via deep learning
Satoshi Kumabe, Tianyu Song, Ton Viet Ta
https://arxiv.org/abs/2507.21486 https://

Stochastic forest transition model dynamics and parameter estimation via deep learning
Forest transitions, characterized by dynamic shifts between forest, agricultural, and abandoned lands, are complex phenomena. This study developed a stochastic differential equation model to capture the intricate dynamics of these transitions. We established the existence of global positive solutions for the model and conducted numerical analyses to assess the impact of model parameters on deforestation incentives. To address the challenge of parameter estimation, we proposed a novel deep learn…

The Flexible Accumulation Model for High Density Temporal Exposures
Xinkai Zhou, Lee Goeddel, Nauder Faraday, Ciprian M. Crainiceanu
https://arxiv.org/abs/2506.22805

The Flexible Accumulation Model for High Density Temporal Exposures
Emerging technologies enable continuous monitoring of temporal exposures to disease risk factors, leading to complex structures in the exposure process that consists of a subject-specific number and duration of exposure episodes. A key scientific question is how does the number and duration of episodic exposure affect disease risk. We address this question by introducing the FLexible Accumulation ModEl (FLAME) and the associated inferential tools, whose finite sample performance is evaluated th…

Training Language Model to Critique for Better Refinement
Tianshu Yu, Chao Xiang, Mingchuan Yang, Pei Ke, Bosi Wen, Cunxiang Wang, Jiale Cheng, Li Zhang, Xinyu Mu, Chuxiong Sun, Minlie Huang
https://arxiv.org/abs/2506.22157

Training Language Model to Critique for Better Refinement
Large language models (LLMs) have demonstrated remarkable evaluation and critique capabilities, providing insightful feedback and identifying flaws in various tasks. However, limited research has explored which types of critiques are most effective for improving model responses or how to generate such critiques. To address this gap, we introduce \textbf{R}efinement-oriented \textbf{C}ritique \textbf{O}ptimization (RCO), a novel framework designed to train critic models using refinement signals.…

Explicit modified Euler approximations of the A\"{i}t-Sahalia type model with Poisson jumps
Yingsong Jiang, Ruishu Liu, Minhong Xu
https://arxiv.org/abs/2506.23947

Explicit modified Euler approximations of the Aït-Sahalia type model with Poisson jumps
This paper focuses on mean-square approximations of a generalized Aït-Sahalia interest rate model with Poisson jumps. The main challenge in the construction and analysis of time-discrete numerical schemes is caused by a drift that blows up at the origin, highly nonlinear drift and diffusion coefficients and positivity-preserving requirement. Due to the presence of the Poisson jumps, additional difficulties arise in recovering the exact order $1/2$ of convergence for the time-stepping schemes. …

From MBQI to Enumerative Instantiation and Back
Marek Dan\v{c}o, Petra Hozzov\'a, Mikol\'a\v{s} Janota
https://arxiv.org/abs/2506.22584 https://

From MBQI to Enumerative Instantiation and Back
This work investigates the relation between model-based quantifier instantiation (MBQI) and enumerative instantiation (EI) in Satisfiability Modulo Theories (SMT). MBQI operates at the semantic level and guarantees to find a counterexample to a given a non-model. However, it may lead to weak instantiations. In contrast, EI strives for completeness by systematically enumerating terms at the syntactic level. However, such terms may not be counter-examples. Here we investigate the relation between…

Google fully releases Gemma 3n, an open-source, multimodal AI model that can run on as little as 2GB of memory; the model was previously available as a preview (David Uzondu/Neowin)
https://www.neowin.net/news/google-releases-gemma-3n-…

Google releases Gemma 3n, a new AI model built for mobile devices
Last month, Google previewed Gemma 3n, an AI model optimized for mobile. Now, it's officially available for developers to test and build with.

Multi-Modal Model Predictive Path Integral Control for Collision Avoidance
Alberto Bertipaglia, Dariu M. Gavrila, Barys Shyrokau
https://arxiv.org/abs/2508.21364 https://…

Multi-Modal Model Predictive Path Integral Control for Collision Avoidance
This paper proposes a novel approach to motion planning and decision-making for automated vehicles, using a multi-modal Model Predictive Path Integral control algorithm. The method samples with Sobol sequences around the prior input and incorporates analytical solutions for collision avoidance. By leveraging multiple modes, the multi-modal control algorithm explores diverse trajectories, such as manoeuvring around obstacles or stopping safely before them, mitigating the risk of sub-optimal solu…

A Large-Scale Evolvable Dataset for Model Context Protocol Ecosystem and Security Analysis
Zhiwei Lin, Bonan Ruan, Jiahao Liu, Weibo Zhao
https://arxiv.org/abs/2506.23474

A Large-Scale Evolvable Dataset for Model Context Protocol Ecosystem and Security Analysis
The Model Context Protocol (MCP) has recently emerged as a standardized interface for connecting language models with external tools and data. As the ecosystem rapidly expands, the lack of a structured, comprehensive view of existing MCP artifacts presents challenges for research. To bridge this gap, we introduce MCPCorpus, a large-scale dataset containing around 14K MCP servers and 300 MCP clients. Each artifact is annotated with 20+ normalized attributes capturing its identity, interface conf…

Attestable Audits: Verifiable AI Safety Benchmarks Using Trusted Execution Environments
Christoph Schnabl, Daniel Hugenroth, Bill Marino, Alastair R. Beresford
https://arxiv.org/abs/2506.23706

Attestable Audits: Verifiable AI Safety Benchmarks Using Trusted Execution Environments
Benchmarks are important measures to evaluate safety and compliance of AI models at scale. However, they typically do not offer verifiable results and lack confidentiality for model IP and benchmark datasets. We propose Attestable Audits, which run inside Trusted Execution Environments and enable users to verify interaction with a compliant AI model. Our work protects sensitive data even when model provider and auditor do not trust each other. This addresses verification challenges raised in re…

Reconstruction of full-space quantum Hamiltonian from its effective, energy-dependent model-space projection
Miloslav Znojil
https://arxiv.org/abs/2506.22854

Reconstruction of full-space quantum Hamiltonian from its effective, energy-dependent model-space projection
Reconstruction of a full-space quantum Hamiltonian from its effective Feshbach's model-space avatar is shown feasible. In a preparatory step the information carried by the effective Hamiltonian is compactified using a linear algebraic operation (matrix inversion). A ``universal'' coupled set of polynomial algebraic equations it then obtained. In a few simplest special cases their solution is given and discussed.

Study of excited $D$ and $D_s$ mesons in a relativized quark model
Saba Noor, Faisal Akram, Bilal Masud
https://arxiv.org/abs/2506.22071 https://

Study of excited $D$ and $D_s$ mesons in a relativized quark model
We use a modified relativistic quark model to study the properties of excited charmed and charmed strange mesons. We calculate the masses and wave functions of conventional charmed and charmed strange mesons incorporating both spin and $S$-$D$ mixing effects and fit parameters of the potential model with known experimental states using differential evolution technique. Using Leading Born-Oppenheimer expansion, we also compute the spectrum and wave functions of first gluonic excited state of cha…

Comparing Cluster-Based Cross-Validation Strategies for Machine Learning Model Evaluation
Afonso Martini Spezia, Mariana Recamonde-Mendoza
https://arxiv.org/abs/2507.22299 https…

Comparing Cluster-Based Cross-Validation Strategies for Machine Learning Model Evaluation
Cross-validation plays a fundamental role in Machine Learning, enabling robust evaluation of model performance and preventing overestimation on training and validation data. However, one of its drawbacks is the potential to create data subsets (folds) that do not adequately represent the diversity of the original dataset, which can lead to biased performance estimates. The objective of this work is to deepen the investigation of cluster-based cross-validation strategies by analyzing the perform…

Harmonic Superspace for Ali-Ilahi's ADHM Instanton Sigma Model
Abbas Ali, Mohsin Ilahi, P. P. Abdul Salih, Shafeeq Rahman Thottoli
https://arxiv.org/abs/2507.22948 https://

Harmonic Superspace for Ali-Ilahi's ADHM Instanton Sigma Model
ADHM Yang-Mills instantons are extended field theoretical objects. These are more general than the more familiar 't Hooft Yang-Mills instantons. Their counter parts exist in string theory in terms of sigma models. Nearly three decades ago Witten constructed a (0,4) supersymmetric linear sigma model incorporating ADHM instantons. Witten's construction was in component form. Galperin and Sokatchev constructed the corresponding off-shell supersymmetric version using the harmonic superspace. Recent…

Are Recommenders Self-Aware? Label-Free Recommendation Performance Estimation via Model Uncertainty
Jiayu Li, Ziyi Ye, Guohao Jian, Zhiqiang Guo, Weizhi Ma, Qingyao Ai, Min Zhang
https://arxiv.org/abs/2507.23208

Are Recommenders Self-Aware? Label-Free Recommendation Performance Estimation via Model Uncertainty
Can a recommendation model be self-aware? This paper investigates the recommender's self-awareness by quantifying its uncertainty, which provides a label-free estimation of its performance. Such self-assessment can enable more informed understanding and decision-making before the recommender engages with any users. To this end, we propose an intuitive and effective method, probability-based List Distribution uncertainty (LiDu). LiDu measures uncertainty by determining the probability that a rec…

Detection of Personal Data in Structured Datasets Using a Large Language Model
Albert Agisha Ntwali, Luca R\"uck, Martin Heckmann
https://arxiv.org/abs/2506.22305

Detection of Personal Data in Structured Datasets Using a Large Language Model
We propose a novel approach for detecting personal data in structured datasets, leveraging GPT-4o, a state-of-the-art Large Language Model. A key innovation of our method is the incorporation of contextual information: in addition to a feature's name and values, we utilize information from other feature names within the dataset as well as the dataset description. We compare our approach to alternative methods, including Microsoft Presidio and CASSED, evaluating them on multiple datasets: DeSSI,…

An Information Bottleneck Asset Pricing Model
Che Sun
https://arxiv.org/abs/2507.23218 https://arxiv.org/pdf/2507.23218

An Information Bottleneck Asset Pricing Model
Deep neural networks (DNNs) have garnered significant attention in financial asset pricing, due to their strong capacity for modeling complex nonlinear relationships within financial data. However, sophisticated models are prone to over-fitting to the noise information in financial data, resulting in inferior performance. To address this issue, we propose an information bottleneck asset pricing model that compresses data with low signal-to-noise ratios to eliminate redundant information and ret…

Multi-robot Path Planning and Scheduling via Model Predictive Optimal Transport (MPC-OT)
Usman A. Khan, Mouhacine Benosman, Wenliang Liu, Federico Pecora, Joseph W. Durham
https://arxiv.org/abs/2508.21205

Multi-robot Path Planning and Scheduling via Model Predictive Optimal Transport (MPC-OT)
In this paper, we propose a novel methodology for path planning and scheduling for multi-robot navigation that is based on optimal transport theory and model predictive control. We consider a setup where $N$ robots are tasked to navigate to $M$ targets in a common space with obstacles. Mapping robots to targets first and then planning paths can result in overlapping paths that lead to deadlocks. We derive a strategy based on optimal transport that not only provides minimum cost paths from robot…

A surrogate model for topology optimisation of elastic structures via parametric autoencoders
Matteo Giacomini, Antonio Huerta
https://arxiv.org/abs/2507.22539 https://

A surrogate model for topology optimisation of elastic structures via parametric autoencoders
A surrogate-based topology optimisation algorithm for linear elastic structures under parametric loads and boundary conditions is proposed. Instead of learning the parametric solution of the state (and adjoint) problems or the optimisation trajectory as a function of the iterations, the proposed approach devises a surrogate version of the entire optimisation pipeline. First, the method predicts a quasi-optimal topology for a given problem configuration as a surrogate model of high-fidelity topo…

Sources: Meta aims to launch its next-gen Llama AI model, Llama 4.X, by the end of 2025; the TBD team developing Llama 4.X is also trying to fix Llama 4 bugs (Jyoti Mann/Business Insider)
https://www.businessinsider.com/meta-superintelligen…

Meta races to launch latest Llama AI model by year-end, sources say
Meta plans to launch its latest AI model, called Llama 4.X, by year-end, two people familiar with the matter told Business Insider.

QZhou-Embedding Technical Report
Peng Yu, En Xu, Bin Chen, Haibiao Chen, Yinfei Xu
https://arxiv.org/abs/2508.21632 https://arxiv.org/pdf/2508.21632…

QZhou-Embedding Technical Report
We present QZhou-Embedding, a general-purpose contextual text embedding model with exceptional text representation capabilities. Built upon the Qwen2.5-7B-Instruct foundation model, we designed a unified multi-task framework comprising specialized data transformation and training strategies. The data transformation scheme enables the incorporation of more diverse textual training datasets, while the task-specific training strategies enhance model learning efficiency. We developed a data synthes…

Leveraging Imperfection with MEDLEY A Multi-Model Approach Harnessing Bias in Medical AI
Farhad Abtahi, Mehdi Astaraki, Fernando Seoane
https://arxiv.org/abs/2508.21648 https://…

Leveraging Imperfection with MEDLEY A Multi-Model Approach Harnessing Bias in Medical AI
Bias in medical artificial intelligence is conventionally viewed as a defect requiring elimination. However, human reasoning inherently incorporates biases shaped by education, culture, and experience, suggesting their presence may be inevitable and potentially valuable. We propose MEDLEY (Medical Ensemble Diagnostic system with Leveraged diversitY), a conceptual framework that orchestrates multiple AI models while preserving their diverse outputs rather than collapsing them into a consensus. U…

The goofy-symmetric Standard Model and the Hierarchy Problem
Thede de Boer, Florian Goertz, Andrea Incrocci
https://arxiv.org/abs/2507.22111 https://arxiv.…

The goofy-symmetric Standard Model and the Hierarchy Problem
A new class of so-called "goofy" symmetries has been shown to lead to renormalization-group stable relations between parameters in two-Higgs-Doublet Models, not known before. In this work we investigate goofy transformations in the Standard Model (SM), extending them to the fermion sector. We show that the SM action is goofy "symmetric", if the Higgs mass term vanishes, while other SM interactions are allowed. Goofy symmetry could thus add another option to the very few means to successfully fo…

JointRank: Rank Large Set with Single Pass
Evgeny Dedov
https://arxiv.org/abs/2506.22262 https://arxiv.org/pdf/2506.22262

JointRank: Rank Large Set with Single Pass
Efficiently ranking relevant items from large candidate pools is a cornerstone of modern information retrieval systems -- such as web search, recommendation, and retrieval-augmented generation. Listwise rerankers, which improve relevance by jointly considering multiple candidates, are often limited in practice: either by model input size constraints, or by degraded quality when processing large sets. We propose a model-agnostic method for fast reranking large sets that exceed a model input limi…

SoK: Large Language Model-Generated Textual Phishing Campaigns End-to-End Analysis of Generation, Characteristics, and Detection
Fengchao Chen, Tingmin Wu, Van Nguyen, Carsten Rudolph
https://arxiv.org/abs/2508.21457

SoK: Large Language Model-Generated Textual Phishing Campaigns End-to-End Analysis of Generation, Characteristics, and Detection
Phishing is a pervasive form of social engineering in which attackers impersonate trusted entities to steal information or induce harmful actions. Text-based phishing dominates for its low cost, scalability, and concealability, advantages recently amplified by large language models (LLMs) that enable ``Phishing-as-a-Service'' attacks at scale within minutes. Despite the growing research into LLM-facilitated phishing attacks, consolidated systematic research on the phishing attack life cycle rem…

Z.ai, formerly known as Zhipu and which has raised $1.5B from Tencent and others, releases GLM-4.5, an open source AI model that's cheaper to use than DeepSeek (Evelyn Cheng/CNBC)
https://www.cnbc.com/2025/07/28/chinas-latest-ai-…

China's latest AI model claims to be even cheaper to use than DeepSeek
Startup Z.ai, formerly known as Zhipu, announced Monday a new, low-cost AI model as Chinese companies race to stay at the frontlines of the tech.

Efficient Multilingual ASR Finetuning via LoRA Language Experts
Jiahong Li, Yiwen Shao, Jianheng Zhuo, Chenda Li, Liliang Tang, Dong Yu, Yanmin Qian
https://arxiv.org/abs/2506.21555

Efficient Multilingual ASR Finetuning via LoRA Language Experts
Recent advancements in deep learning have significantly enhanced multilingual automatic speech recognition (ASR) due to the development of advanced model architectures and available large-scale multilingual datasets. Despite that, multilingual ASR still suffers from the curse of multilinguality in that different languages tend to interfere with each other, making it difficult for the ASR model to identify multiple languages effectively while sharing model capacity across them. This paper propos…

FedCVD : Communication-Efficient Federated Learning for Cardiovascular Risk Prediction with Parametric and Non-Parametric Model Optimization
Abdelrhman Gaber, Hassan Abd-Eltawab, John Elgallab, Youssif Abuzied, Dineo Mpanya, Turgay Celik, Swarun Kumar, Tamer ElBatt
https://arxiv.org/abs/2507.22963…

FedCVD++: Communication-Efficient Federated Learning for Cardiovascular Risk Prediction with Parametric and Non-Parametric Model Optimization
Cardiovascular diseases (CVD) cause over 17 million deaths annually worldwide, highlighting the urgent need for privacy-preserving predictive systems. We introduce FedCVD++, an enhanced federated learning (FL) framework that integrates both parametric models (logistic regression, SVM, neural networks) and non-parametric models (Random Forest, XGBoost) for coronary heart disease risk prediction. To address key FL challenges, we propose: (1) tree-subset sampling that reduces Random Forest communi…

Features of Charged Lepton Flavor Violation in an $A_4$ Symmetric Neutrino Mass Model
Pravesh Chndra Awasthi, Jai More, Akhila Kumar Pradhan, Kumar Rao, Purushottam Sahu, S. Uma Sankar
https://arxiv.org/abs/2507.23747

Features of Charged Lepton Flavor Violation in an $A_4$ Symmetric Neutrino Mass Model
Neutrino flavour oscillations imply that there must be charged lepton flavour violation (CLFV) also. Different neutrino mass models predict different patterns of CLFV decays. Neutrino mass generation through standard see-saw mechanisms leads to the prediction that the branching ratios of meson CLFV decays will always be smaller than the corresponding radiative CLFV decays. In this work, we analyse an interesting neutrino mass model, based on $A_4$ symmetry, in which the symmetry and the symmetr…

Counterfactual Scenarios for Automated Planning
Nicola Gigante, Francesco Leofante, Andrea Micheli
https://arxiv.org/abs/2508.21521 https://arxiv.org/pdf/2…

Counterfactual Scenarios for Automated Planning
Counterfactual Explanations (CEs) are a powerful technique used to explain Machine Learning models by showing how the input to a model should be minimally changed for the model to produce a different output. Similar proposals have been made in the context of Automated Planning, where CEs have been characterised in terms of minimal modifications to an existing plan that would result in the satisfaction of a different goal. While such explanations may help diagnose faults and reason about the cha…

Large Language Model-Based Framework for Explainable Cyberattack Detection in Automatic Generation Control Systems
Muhammad Sharshar, Ahmad Mohammad Saber, Davor Svetinovic, Amr M. Youssef, Deepa Kundur, Ehab F. El-Saadany
https://arxiv.org/abs/2507.22239

Large Language Model-Based Framework for Explainable Cyberattack Detection in Automatic Generation Control Systems
The increasing digitization of smart grids has improved operational efficiency but also introduced new cybersecurity vulnerabilities, such as False Data Injection Attacks (FDIAs) targeting Automatic Generation Control (AGC) systems. While machine learning (ML) and deep learning (DL) models have shown promise in detecting such attacks, their opaque decision-making limits operator trust and real-world applicability. This paper proposes a hybrid framework that integrates lightweight ML-based attac…

European Commission says Meta has proposed limited changes to its pay-or-consent ad model, and Meta faces daily fines if it determines they are not sufficient (Reuters)
https://economictimes.indiatimes.com/t

Meta will only make limited changes to pay-or-consent model, EU says
Meta Platforms made only limited changes to its pay-or-consent model, and EU regulators cannot confirm compliance with an antitrust order. The European Commission may impose periodic penalties starting June 27, if Meta continues to fail meeting the required standards.

Can a mobile robot learn from a pedestrian model to prevent the sidewalk salsa?
Olger Siebinga, David Abbink
https://arxiv.org/abs/2508.21690 https://arxiv…

Can a mobile robot learn from a pedestrian model to prevent the sidewalk salsa?
Pedestrians approaching each other on a sidewalk sometimes end up in an awkward interaction known as the "sidewalk salsa": they both (repeatedly) deviate to the same side to avoid a collision. This provides an interesting use case to study interactions between pedestrians and mobile robots because, in the vast majority of cases, this phenomenon is avoided through a negotiation based on implicit communication. Understanding how it goes wrong and how pedestrians end up in the sidewalk salsa will …

Measuring Time-Series Dataset Similarity using Wasserstein Distance
Hongjie Chen, Akshay Mehra, Josh Kimball, Ryan A. Rossi
https://arxiv.org/abs/2507.22189 https://

Measuring Time-Series Dataset Similarity using Wasserstein Distance
The emergence of time-series foundation model research elevates the growing need to measure the (dis)similarity of time-series datasets. A time-series dataset similarity measure aids research in multiple ways, including model selection, finetuning, and visualization. In this paper, we propose a distribution-based method to measure time-series dataset similarity by leveraging the Wasserstein distance. We consider a time-series dataset an empirical instantiation of an underlying multivariate norm…

ControlMed: Adding Reasoning Control to Medical Language Model
Sung-Min Lee, Siyoon Lee, Juyeon Kim, Kyungmin Roh
https://arxiv.org/abs/2507.22545 https://…

ControlMed: Adding Reasoning Control to Medical Language Model
Reasoning Large Language Models (LLMs) with enhanced accuracy and explainability are increasingly being adopted in the medical domain, as the life-critical nature of clinical decision-making demands reliable support. Despite these advancements, existing reasoning LLMs often generate unnecessarily lengthy reasoning processes, leading to significant computational overhead and response latency. These limitations hinder their practical deployment in real-world clinical environments. To address thes…

Huawei open sources its 7B and 72B Pangu AI models, and some of its model reasoning tech, to incentivize the use of Huawei products like its Ascend AI chips (Dylan Butts/CNBC)
https://www.cnbc.com/2025/07/01/huawei-open-sour…

China's Huawei open-sources AI models as it seeks adoption across the global AI market
Huawei has open-sourced more of its AI models. Tech experts say it'll help the U.S.-blacklisted company continue to expand throughout the AI ecosystem.

Failure Prediction Is a Better Performance Proxy for Early-Exit Networks Than Calibration
Piotr Kubaty, Filip Szatkowski, Metod Jazbec, Bartosz W\'ojcik
https://arxiv.org/abs/2508.21495

Failure Prediction Is a Better Performance Proxy for Early-Exit Networks Than Calibration
Early-exit models speed up inference by attaching internal classifiers to intermediate layers of the model and allowing computation to stop once a prediction satisfies an exit criterion. Most early-exit methods rely on confidence-based exit strategies, which motivated some works to calibrate intermediate classifiers to improve the performance of the entire model. In this paper, we show that calibration measures can be misleading indicators of the performance of multi-exit models: a well-calibra…

Using Scaling Laws for Data Source Utility Estimation in Domain-Specific Pre-Training
Oleksiy Ostapenko, Charles Guille-Escuret, Luke Kumar, Max Tian, Denis Kocetkov, Gopeshh Subbaraj, Raymond Li, Joel Lamy-Poirier, Sebastien Paquet, Torsten Scholak
https://arxiv.org/abs/2507.22250

Using Scaling Laws for Data Source Utility Estimation in Domain-Specific Pre-Training
We introduce a framework for optimizing domain-specific dataset construction in foundation model training. Specifically, we seek a cost-efficient way to estimate the quality of data sources (e.g. synthetically generated or filtered web data, etc.) in order to make optimal decisions about resource allocation for data sourcing from these sources for the stage two pre-training phase, aka annealing, with the goal of specializing a generalist pre-trained model to specific domains. Our approach exten…

Benchmarking the State of Networks with a Low-Cost Method Based on Reservoir Computing
Felix Simon Reimers, Carl-Hendrik Peters, Stefano Nichele
https://arxiv.org/abs/2508.21420

Benchmarking the State of Networks with a Low-Cost Method Based on Reservoir Computing
Using data from mobile network utilization in Norway, we showcase the possibility of monitoring the state of communication and mobility networks with a non-invasive, low-cost method. This method transforms the network data into a model within the framework of reservoir computing and then measures the model's performance on proxy tasks. Experimentally, we show how the performance on these proxies relates to the state of the network. A key advantage of this approach is that it uses readily availa…