Tootfinder

I have a nice Anker power bank with a built-in USB-C cable strong enough to hold it by.
But recently they recalled it and offered to send a replacement. I just unboxed the replacement… and it’s a different model. One which is physically larger and has no built-in cable. Same capacity, higher wattage, but not what I want.
In fact the original model is no longer sold (at that capacity). I didn’t realize they’d recalled all of them…
Oh well. I imagine they’ll bring them b…

Scaling Learned Image Compression Models up to 1 Billion
Yuqi Li, Haotian Zhang, Li Li, Dong Liu, Feng Wu
https://arxiv.org/abs/2508.09075 https://arxiv.or…

Scaling Learned Image Compression Models up to 1 Billion
Recent advances in large language models (LLMs) highlight a strong connection between intelligence and compression. Learned image compression, a fundamental task in modern data compression, has made significant progress in recent years. However, current models remain limited in scale, restricting their representation capacity, and how scaling model size influences compression performance remains unexplored. In this work, we present a pioneering study on scaling up learned image compression mode…

This https://arxiv.org/abs/2505.15030 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csLG_…

Harnessing On-Device Large Language Model: Empirical Results and Implications for AI PC
The increasing deployment of Large Language Models (LLMs) on edge devices, driven by model advancements and hardware improvements, offers significant privacy benefits. However, these on-device LLMs inherently face performance limitations due to reduced model capacity and necessary compression techniques. To address this, we introduce a systematic methodology -- encompassing model capability, development efficiency, and system resources -- for evaluating on-device LLMs. Our comprehensive evaluat…

Introduction to Sachdev-Ye-Kitaev Model: A Strongly Correlated System Perspective
Rishabh Jha
https://arxiv.org/abs/2507.07195 https://

Introduction to Sachdev-Ye-Kitaev Model: A Strongly Correlated System Perspective
The Sachdev-Ye-Kitaev (SYK) model delivers an analytically tractable framework for exotic strongly correlated phases where conventional paradigms like Landau's Fermi liquid theory collapse. This review offers a pedagogical introduction to the SYK physics, highlighting its unique capacity to model \textit{strange metals} -- systems exhibiting linear-in-temperature resistivity, Planckian dissipation, and quasiparticle breakdown. We systematically construct both Majorana and complex fermion varian…

Learning an Implicit Physics Model for Image-based Fluid Simulation
Emily Yue-Ting Jia, Jiageng Mao, Zhiyuan Gao, Yajie Zhao, Yue Wang
https://arxiv.org/abs/2508.08254 https://

Learning an Implicit Physics Model for Image-based Fluid Simulation
Humans possess an exceptional ability to imagine 4D scenes, encompassing both motion and 3D geometry, from a single still image. This ability is rooted in our accumulated observations of similar scenes and an intuitive understanding of physics. In this paper, we aim to replicate this capacity in neural networks, specifically focusing on natural fluid imagery. Existing methods for this task typically employ simplistic 2D motion estimators to animate the image, leading to motion predictions that …

Geometry-Aware Spiking Graph Neural Network
Bowen Zhang, Genan Dai, Hu Huang, Long Lan
https://arxiv.org/abs/2508.06793 https://arxiv.org/pdf/2508.06793

Geometry-Aware Spiking Graph Neural Network
Graph Neural Networks (GNNs) have demonstrated impressive capabilities in modeling graph-structured data, while Spiking Neural Networks (SNNs) offer high energy efficiency through sparse, event-driven computation. However, existing spiking GNNs predominantly operate in Euclidean space and rely on fixed geometric assumptions, limiting their capacity to model complex graph structures such as hierarchies and cycles. To overcome these limitations, we propose \method{}, a novel Geometry-Aware Spikin…

README: Robust Error-Aware Digital Signature Framework via Deep Watermarking Model
Hyunwook Choi, Sangyun Won, Daeyeon Hwang, Junhyeok Choi
https://arxiv.org/abs/2507.04495

README: Robust Error-Aware Digital Signature Framework via Deep Watermarking Model
Deep learning-based watermarking has emerged as a promising solution for robust image authentication and protection. However, existing models are limited by low embedding capacity and vulnerability to bit-level errors, making them unsuitable for cryptographic applications such as digital signatures, which require over 2048 bits of error-free data. In this paper, we propose README (Robust Error-Aware Digital Signature via Deep WaterMarking ModEl), a novel framework that enables robust, verifiabl…

This https://arxiv.org/abs/2505.12538 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_eco…

On long-duration storage, weather uncertainty and limited foresight
Long-duration energy storage (LDES) is a key component for fully renewable, sector-coupled energy systems based on wind and solar. While capacity expansion planning has begun to take into account interannual weather variability, it often ignores weather uncertainty and limited foresight in capacity and operational decisions. We build a stochastic capacity expansion model for fully decarbonized energy systems with LDES in Europe accounting for weather uncertainty - isolating the effect of limite…

EC2MoE: Adaptive End-Cloud Pipeline Collaboration Enabling Scalable Mixture-of-Experts Inference
Zheming Yang, Yunqing Hu, Sheng Sun, Wen Ji
https://arxiv.org/abs/2508.06024 htt…

EC2MoE: Adaptive End-Cloud Pipeline Collaboration Enabling Scalable Mixture-of-Experts Inference
The Mixture-of-Experts (MoE) paradigm has emerged as a promising solution to scale up model capacity while maintaining inference efficiency. However, deploying MoE models across heterogeneous end-cloud environments poses new challenges in expert scheduling, communication overhead, and resource heterogeneity. In this paper, we propose EC2MoE, an adaptive framework for scalable MoE inference via end-cloud pipeline collaboration. First, we design a hardware-aware lightweight group gate network tha…

Efficient Edge LLMs Deployment via HessianAware Quantization and CPU GPU Collaborative
Tuo Zhang, Ning Li, Xin Yuan, Wenchao Xu, Quan Chen, Song Guo, Haijun Zhang
https://arxiv.org/abs/2508.07329

Efficient Edge LLMs Deployment via HessianAware Quantization and CPU GPU Collaborative
With the breakthrough progress of large language models (LLMs) in natural language processing and multimodal tasks, efficiently deploying them on resource-constrained edge devices has become a critical challenge. The Mixture of Experts (MoE) architecture enhances model capacity through sparse activation, but faces two major difficulties in practical deployment: (1) The presence of numerous outliers in activation distributions leads to severe degradation in quantization accuracy for both activat…

Optimal Sizing and Control of a Grid-Connected Battery in a Stacked Revenue Model Including an Energy Community
Tudor Octavian Pocola, Valentin Robu, Jip Rietveld, Sonam Norbu, Benoit Couraud, Merlinda Andoni, David Flynn, H. Vincent Poor
https://arxiv.org/abs/2507.04343

Optimal Sizing and Control of a Grid-Connected Battery in a Stacked Revenue Model Including an Energy Community
Recent years have seen rapid increases in intermittent renewable generation, requiring novel battery energy storage systems (BESS) solutions. One recent trend is the emergence of large grid-connected batteries, that can be controlled to provide multiple storage and flexibility services, using a stacked revenue model. Another emerging development is renewable energy communities (REC), in which prosumers invest in their own renewable generation capacity, but also requiring battery storage for fle…

A Model-Data Dual-Driven Resource Allocation Scheme for IREE Oriented 6G Networks
Tao Yu, Simin Wang, Shunqing Zhang, Xiaojing Chen, Zi Xu, Xin Wang, Jiandong Li, Junyu Liu, Sihai Zhang
https://arxiv.org/abs/2506.03508

A Model-Data Dual-Driven Resource Allocation Scheme for IREE Oriented 6G Networks
The rapid and substantial fluctuations in wireless network capacity and traffic demand, driven by the emergence of 6G technologies, have exacerbated the issue of traffic-capacity mismatch, raising concerns about wireless network energy consumption. To address this challenge, we propose a model-data dual-driven resource allocation (MDDRA) algorithm aimed at maximizing the integrated relative energy efficiency (IREE) metric under dynamic traffic conditions. Unlike conventional model-driven or dat…

Powers of Magnetic Graph Matrix: Fourier Spectrum, Walk Compression, and Applications
Yinan Huang, David F. Gleich, Pan Li
https://arxiv.org/abs/2506.07343

Powers of Magnetic Graph Matrix: Fourier Spectrum, Walk Compression, and Applications
Magnetic graphs, originally developed to model quantum systems under magnetic fields, have recently emerged as a powerful framework for analyzing complex directed networks. Existing research has primarily used the spectral properties of the magnetic graph matrix to study global and stationary network features. However, their capacity to model local, non-equilibrium behaviors, often described by matrix powers, remains largely unexplored. We present a novel combinatorial interpretation of the mag…

RoboLinker: A Diffusion-model-based Matching Clothing Generator Between Humans and Companion Robots
Jing Tang, Qing Xiao, Kunxu Du, Zaiqiao Ye
https://arxiv.org/abs/2508.01165 h…

RoboLinker: A Diffusion-model-based Matching Clothing Generator Between Humans and Companion Robots
We present RoboLinker, a generative design system that creates matching outfits for humans and their robots. Using a diffusion-based model, the system takes a robot image and a style prompt from users as input, and outputs a human outfit that visually complements the robot's attire. Through an interactive interface, users can refine the generated designs. We evaluate RoboLinker with both humanoid and pet-like robots, demonstrating its capacity to produce stylistically coherent and emotionally r…

A Spin Glass Characterization of Neural Networks
Jun Li
https://arxiv.org/abs/2508.07397 https://arxiv.org/pdf/2508.07397

A Spin Glass Characterization of Neural Networks
This work presents a statistical mechanics characterization of neural networks, motivated by the replica symmetry breaking (RSB) phenomenon in spin glasses. A Hopfield-type spin glass model is constructed from a given feedforward neural network (FNN). Overlaps between simulated replica samples serve as a characteristic descriptor of the FNN. The connection between the spin-glass description and commonly studied properties of the FNN -- such as data fitting, capacity, generalization, and robustn…

Modeling Optical Key Distribution over a Satellite-to-Ground Link Under Weak Atmospheric Turbulence
Artur Czerwinski, Miko{\l}aj Lasota, Marcin Jarzyna, Mateusz Kucharczyk, Micha{\l} Jachura, Konrad Banaszek
https://arxiv.org/abs/2508.05807

Modeling Optical Key Distribution over a Satellite-to-Ground Link Under Weak Atmospheric Turbulence
In this study, we analyze the secret key capacity of intensity modulation/direct detection optical key distribution (IM/DD OKD) for a free-space optical (FSO) link between a low-Earth orbit satellite and an optical ground station. Focusing on downlink communication, we account for atmospheric turbulence, which causes random variations in the transmittance of the FSO channel. We implement an atmospheric channel model that accounts for absorption and scattering, geometric losses, pointing errors,…

MedMambaLite: Hardware-Aware Mamba for Medical Image Classification
Romina Aalishah, Mozhgan Navardi, Tinoosh Mohsenin
https://arxiv.org/abs/2508.05049 https://

MedMambaLite: Hardware-Aware Mamba for Medical Image Classification
AI-powered medical devices have driven the need for real-time, on-device inference such as biomedical image classification. Deployment of deep learning models at the edge is now used for applications such as anomaly detection and classification in medical images. However, achieving this level of performance on edge devices remains challenging due to limitations in model size and computational capacity. To address this, we present MedMambaLite, a hardware-aware Mamba-based model optimized throug…

Enhancing Environment Generalizability for Deep Learning-Based CSI Feedback
Haoyu Wang, Shuangfeng Han, Xiaoyun Wang, Zhi Sun
https://arxiv.org/abs/2507.06833

Enhancing Environment Generalizability for Deep Learning-Based CSI Feedback
Accurate and low-overhead channel state information (CSI) feedback is essential to boost the capacity of frequency division duplex (FDD) massive multiple-input multiple-output (MIMO) systems. Deep learning-based CSI feedback significantly outperforms conventional approaches. Nevertheless, current deep learning-based CSI feedback algorithms exhibit limited generalizability to unseen environments, which obviously increases the deployment cost. In this paper, we first model the distribution shift …

A learning model predictive control for virtual coupling in railroads
Miguel A. Vaquero-Serrano, Francesco Borrelli, Jesus Felez
https://arxiv.org/abs/2507.02383

A learning model predictive control for virtual coupling in railroads
The objective of this paper is to present a novel intelligent train control system for virtual coupling in railroads based on a Learning Model Predictive Control (LMPC). Virtual coupling is an emerging railroad technology that reduces the distance between trains to increase the capacity of the line, whereas LMPC is an optimization-based controller that incorporates artificial intelligence methods to improve its control policies. By incorporating data from past experiences into the optimization …

IGSM: Improved Geometric and Sensitivity Matching for Finetuning Pruned Diffusion Models
Caleb Zheng, Eli Shlizerman
https://arxiv.org/abs/2506.05398 https…

IGSM: Improved Geometric and Sensitivity Matching for Finetuning Pruned Diffusion Models
Diffusion models achieve realistic outcomes across a wide range of generative tasks, but their high computational cost remains a major barrier to deployment. Model pruning has emerged as a promising strategy to reduce inference cost and enable lightweight diffusion models. While effective, pruned diffusion models are proned to quality reduction due to limited capacity. A key limitation of current pruning approaches is that pruned models are finetuned using the same objective as the dense model,…

AdamMeme: Adaptively Probe the Reasoning Capacity of Multimodal Large Language Models on Harmfulness
Zixin Chen, Hongzhan Lin, Kaixin Li, Ziyang Luo, Zhen Ye, Guang Chen, Zhiyong Huang, Jing Ma
https://arxiv.org/abs/2507.01702

AdamMeme: Adaptively Probe the Reasoning Capacity of Multimodal Large Language Models on Harmfulness
The proliferation of multimodal memes in the social media era demands that multimodal Large Language Models (mLLMs) effectively understand meme harmfulness. Existing benchmarks for assessing mLLMs on harmful meme understanding rely on accuracy-based, model-agnostic evaluations using static datasets. These benchmarks are limited in their ability to provide up-to-date and thorough assessments, as online memes evolve dynamically. To address this, we propose AdamMeme, a flexible, agent-based evalua…

G-Memory: Tracing Hierarchical Memory for Multi-Agent Systems
Guibin Zhang, Muxin Fu, Guancheng Wan, Miao Yu, Kun Wang, Shuicheng Yan
https://arxiv.org/abs/2506.07398

G-Memory: Tracing Hierarchical Memory for Multi-Agent Systems
Large language model (LLM)-powered multi-agent systems (MAS) have demonstrated cognitive and execution capabilities that far exceed those of single LLM agents, yet their capacity for self-evolution remains hampered by underdeveloped memory architectures. Upon close inspection, we are alarmed to discover that prevailing MAS memory mechanisms (1) are overly simplistic, completely disregarding the nuanced inter-agent collaboration trajectories, and (2) lack cross-trial and agent-specific customiza…

Continuous-time multifarious systems -- Part II: non-reciprocal multifarious self-organization
Jakob Metson, Saeed Osat, Ramin Golestanian
https://arxiv.org/abs/2506.07649

Continuous-time multifarious systems -- Part II: non-reciprocal multifarious self-organization
In the context of self-assembly, where complex structures can be assembled from smaller units, it is desirable to devise strategies towards disassembly and reassembly processes that reuse the constituent parts. A non-reciprocal multifarious self-organization strategy has been recently introduced, and shown to have the capacity to exhibit this complex property. In this work, we study the model using continuous-time Gillespie simulations, and compare the results against discrete-time Monte Carlo …

This is welcome from #Burnham.
(Warning 15 cookies)
'Greater Manchester mayor Andy Burnham hit out, saying the decision would help “overheat” the economy. He warned: “It’s a model for an ever-overheating UK economy rather than a more balanced, levelled-up economy, which is what we would argue for. Manchester Airport has two runways and it doesn’t operate at full capacity.”'

Labour split over backing for Heathrow expansion with Reeves set to give her support
Chancellor Rachel Reeves is expected to voice her support for a third runway at Heathrow despite opposition from cabinet collegue Ed Miliband and London mayor Sadiq Khan

This https://arxiv.org/abs/2505.16865 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csIR_…

LARES: Latent Reasoning for Sequential Recommendation
Sequential recommender systems have become increasingly important in real-world applications that model user behavior sequences to predict their preferences. However, existing sequential recommendation methods predominantly rely on non-reasoning paradigms, which may limit the model's computational capacity and result in suboptimal recommendation performance. To address these limitations, we present LARES, a novel and scalable LAtent REasoning framework for Sequential recommendation that enhance…

This https://arxiv.org/abs/2506.04168 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csLG_…

Horizon Reduction Makes RL Scalable
In this work, we study the scalability of offline reinforcement learning (RL) algorithms. In principle, a truly scalable offline RL algorithm should be able to solve any given problem, regardless of its complexity, given sufficient data, compute, and model capacity. We investigate if and how current offline RL algorithms match up to this promise on diverse, challenging, previously unsolved tasks, using datasets up to 1000x larger than typical offline RL datasets. We observe that despite scaling…

Meta's AI Model 'Memorized' Huge Chunks of Books, Including 'Harry Potter' and '1984' https://www.404media.co/meta-ai-model-memorized-harry-potter-books/

System-performance and cost modeling of Large Language Model training and inference
Wenzhe Guo, Joyjit Kundu, Uras Tos, Weijiang Kong, Giuliano Sisto, Timon Evenblij, Manu Perumkunnil
https://arxiv.org/abs/2507.02456

System-performance and cost modeling of Large Language Model training and inference
Large language models (LLMs), based on transformer architectures, have revolutionized numerous domains within artificial intelligence, science, and engineering due to their exceptional scalability and adaptability. However, the exponential growth in LLM size and complexity has outpaced advancements in compute capacity, memory bandwidth, network performance, and cost efficiency, posing significant challenges to their scalability on distributed systems. To address these limitations, alternative m…

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…

This https://arxiv.org/abs/2503.19217 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csSE_…

LLM Benchmarking with LLaMA2: Evaluating Code Development Performance Across Multiple Programming Languages
The rapid evolution of large language models (LLMs) has opened new possibilities for automating various tasks in software development. This paper evaluates the capabilities of the Llama 2-70B model in automating these tasks for scientific applications written in commonly used programming languages. Using representative test problems, we assess the model's capacity to generate code, documentation, and unit tests, as well as its ability to translate existing code between commonly used programming…

Fast Forward and Inverse Thermal Modeling for Parameter Estimation of Multi-Layer Composites -- Part II: Inverse Modeling and Applications
Gan Fu, Mitrofan Curti, Calina Ciuhu, Elena A. Lomonova
https://arxiv.org/abs/2507.06746

Fast Forward and Inverse Thermal Modeling for Parameter Estimation of Multi-Layer Composites -- Part II: Inverse Modeling and Applications
A fast inverse heat conduction model (IHCM) is developed for estimating unknown properties of multi-layer composites considering internal heat generation. This work builds on the validated analytical forward models presented in Part I. Transient temperature at a single point is used as input, with the objective function minimized through an interior-point optimization algorithm. The IHCM accurately estimates thermal properties such as thermal conductivity, specific heat capacity, density, and h…

Cognitive Load and Information Processing in Financial Markets: Theory and Evidence from Disclosure Complexity
Yimin Du, Guolin Tang
https://arxiv.org/abs/2507.07037 https://

Cognitive Load and Information Processing in Financial Markets: Theory and Evidence from Disclosure Complexity
We develop a theoretical framework for understanding how cognitive load affects information processing in financial markets and test it using exogenous variation in disclosure complexity. Our model distinguishes between attention allocation and cognitive processing capacity, showing that complex information creates differential effects across investor types. Using a comprehensive dataset of corporate disclosures and a novel identification strategy based on regulatory changes, we find that cogni…

AlphaGo Moment for Model Architecture Discovery
Yixiu Liu, Yang Nan, Weixian Xu, Xiangkun Hu, Lyumanshan Ye, Zhen Qin, Pengfei Liu
https://arxiv.org/abs/2507.18074 https://

AlphaGo Moment for Model Architecture Discovery
While AI systems demonstrate exponentially improving capabilities, the pace of AI research itself remains linearly bounded by human cognitive capacity, creating an increasingly severe development bottleneck. We present ASI-Arch, the first demonstration of Artificial Superintelligence for AI research (ASI4AI) in the critical domain of neural architecture discovery--a fully autonomous system that shatters this fundamental constraint by enabling AI to conduct its own architectural innovation. Movi…

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…

Accounting for Subsystem Aging Variability in Battery Energy Storage System Optimization
Melina Grane, Martin Cornejo, Holger Hesse, Andreas Jossen
https://arxiv.org/abs/2507.04813

Accounting for Subsystem Aging Variability in Battery Energy Storage System Optimization
This paper presents a degradation-cost-aware optimization framework for multi-string battery energy storage systems, emphasizing the impact of inhomogeneous subsystem-level aging in operational decision-making. We evaluate four scenarios for an energy arbitrage scenario, that vary in model precision and treatment of aging costs. Key performance metrics include operational revenue, power schedule mismatch, missed revenues, capacity losses, and revenue generated per unit of capacity loss. Our ana…

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 …

An Extended Model of Fractional-Dimensional Space for Anisotropic Solids with Deformed Derivatives
Jos\'e Weberszpil
https://arxiv.org/abs/2506.18127 h…

An Extended Model of Fractional-Dimensional Space for Anisotropic Solids with Deformed Derivatives
In this work, we extend a fractional-dimensional space model for anisotropic solids by incorporating a q-deformed derivative operator, inspired by Tsallis' nonadditive entropy framework. This generalization provides an analytical framework for exploring anisotropic thermal properties, within a unified and flexible mathematical formalism. We derive modified expressions for the phonon density of states and specific heat capacity, highlighting the impact of the deformation parameters on thermodyna…

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…

MaXsive: High-Capacity and Robust Training-Free Generative Image Watermarking in Diffusion Models
Po-Yuan Mao, Cheng-Chang Tsai, Chun-Shien Lu
https://arxiv.org/abs/2507.21195 h…

MaXsive: High-Capacity and Robust Training-Free Generative Image Watermarking in Diffusion Models
The great success of the diffusion model in image synthesis led to the release of gigantic commercial models, raising the issue of copyright protection and inappropriate content generation. Training-free diffusion watermarking provides a low-cost solution for these issues. However, the prior works remain vulnerable to rotation, scaling, and translation (RST) attacks. Although some methods employ meticulously designed patterns to mitigate this issue, they often reduce watermark capacity, which c…

This https://arxiv.org/abs/2505.23337 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csLG_…

Matryoshka Model Learning for Improved Elastic Student Models
Industry-grade ML models are carefully designed to meet rapidly evolving serving constraints, which requires significant resources for model development. In this paper, we propose MatTA, a framework for training multiple accurate Student models using a novel Teacher-TA-Student recipe. TA models are larger versions of the Student models with higher capacity, and thus allow Student models to better relate to the Teacher model and also bring in more domain-specific expertise. Furthermore, multiple…

The Urn of Hill, Lane and Sudderth
Simone Franchini
https://arxiv.org/abs/2506.20826 https://arxiv.org/pdf/2506.20826

The Urn of Hill, Lane and Sudderth
We review some facts, properties and applications of the urn of Hill, Lane and Sudderth, a paradigmatic model of stochastic process with memory, where the urn evolution is as follows: consider an urn of given capacity, at each step a new ball, black or white, is added to the urn with probability that is function (urn function) of the fraction of black balls. The process runs until capacity is reached.

Thermodynamic properties of CrMnFeCoNi high entropy alloy at elevated electronic temperatures
Nikita Medvedev
https://arxiv.org/abs/2506.23171 https://

Thermodynamic properties of CrMnFeCoNi high entropy alloy at elevated electronic temperatures
The Cantor alloy (equiatomic CrMnFeCoNi) is a high-entropy alloy with unique physical properties and radiation resistance. To model its response to intense laser pulses, the parameters of the electronic ensemble are required. In this work, the electronic heat capacity, thermal conductivity, and electron-phonon coupling strength at elevated electronic temperatures are evaluated using a combined approach that incorporates tight-binding molecular dynamics and the Boltzmann equation. The damage thr…

Overfitting in Histopathology Model Training: The Need for Customized Architectures
Saghir Alfasly, Ghazal Alabtah, H. R. Tizhoosh
https://arxiv.org/abs/2506.16631

Overfitting in Histopathology Model Training: The Need for Customized Architectures
This study investigates the critical problem of overfitting in deep learning models applied to histopathology image analysis. We show that simply adopting and fine-tuning large-scale models designed for natural image analysis often leads to suboptimal performance and significant overfitting when applied to histopathology tasks. Through extensive experiments with various model architectures, including ResNet variants and Vision Transformers (ViT), we show that increasing model capacity does not …

Assessing the Macroeconomic Impacts of Disasters: an Updated Multi-Regional Impact Assessment (MRIA) model
Surender Raj Vanniya Perumal, Mark Thissen, Marleen de Ruiter, Elco E. Koks
https://arxiv.org/abs/2508.00510

Assessing the Macroeconomic Impacts of Disasters: an Updated Multi-Regional Impact Assessment (MRIA) model
Disasters often impact supply chains, leading to cascading effects across regions. While unaffected regions may attempt to compensate, their ability is constrained by their available production capacity and logistical constraints between regions. This study introduces a Multi-Regional Impact Assessment (MRIA) model to evaluate the regional and macroeconomic consequences of disasters, capturing regional post-disaster trade dynamics and logistical constraints. Our findings emphasize that enhancin…

Decoding Cortical Microcircuits: A Generative Model for Latent Space Exploration and Controlled Synthesis
Xingyu Liu, Yubin Li, Guozhang Chen
https://arxiv.org/abs/2506.11062

Decoding Cortical Microcircuits: A Generative Model for Latent Space Exploration and Controlled Synthesis
A central idea in understanding brains and building artificial intelligence is that structure determines function. Yet, how the brain's complex structure arises from a limited set of genetic instructions remains a key question. The ultra high-dimensional detail of neural connections vastly exceeds the information storage capacity of genes, suggesting a compact, low-dimensional blueprint must guide brain development. Our motivation is to uncover this blueprint. We introduce a generative model, t…

SAM2-UNeXT: An Improved High-Resolution Baseline for Adapting Foundation Models to Downstream Segmentation Tasks
Xinyu Xiong, Zihuang Wu, Lei Zhang, Lei Lu, Ming Li, Guanbin Li
https://arxiv.org/abs/2508.03566

SAM2-UNeXT: An Improved High-Resolution Baseline for Adapting Foundation Models to Downstream Segmentation Tasks
Recent studies have highlighted the potential of adapting the Segment Anything Model (SAM) for various downstream tasks. However, constructing a more powerful and generalizable encoder to further enhance performance remains an open challenge. In this work, we propose SAM2-UNeXT, an advanced framework that builds upon the core principles of SAM2-UNet while extending the representational capacity of SAM2 through the integration of an auxiliary DINOv2 encoder. By incorporating a dual-resolution st…

Efficient LLM Inference: Bandwidth, Compute, Synchronization, and Capacity are all you need
Michael Davies, Neal Crago, Karthikeyan Sankaralingam, Christos Kozyrakis
https://arxiv.org/abs/2507.14397

Efficient LLM Inference: Bandwidth, Compute, Synchronization, and Capacity are all you need
This paper presents a limit study of transformer-based large language model (LLM) inference, focusing on the fundamental performance bottlenecks imposed by memory bandwidth, memory capacity, and synchronization overhead in distributed inference systems. We develop a hardware-agnostic performance model that abstracts away implementation details, enabling the analysis of a wide range of current and near-future hardware technologies. Our analysis spans from current HBM3 memory technology used in A…

MountainLion: A Multi-Modal LLM-Based Agent System for Interpretable and Adaptive Financial Trading
Siyi Wu, Zhaoyang Guan, Leyi Zhao, Xinyuan Song, Xinyu Ying, Hanlin Zhang, Michele Pak, Yangfan He, Yi Xin, Jianhui Wang, Tianyu Shi
https://arxiv.org/abs/2507.20474

MountainLion: A Multi-Modal LLM-Based Agent System for Interpretable and Adaptive Financial Trading
Cryptocurrency trading is a challenging task requiring the integration of heterogeneous data from multiple modalities. Traditional deep learning and reinforcement learning approaches typically demand large training datasets and encode diverse inputs into numerical representations, often at the cost of interpretability. Recent progress in large language model (LLM)-based agents has demonstrated the capacity to process multi-modal data and support complex investment decision-making. Building on t…

Rack-Aware MSR Codes with Linear Field Size and Smaller Sub-Packetization for Tolerating Multiple Erasures
Hengming Zhao, Dianhua Wu, Minquan Cheng
https://arxiv.org/abs/2506.18367

Rack-Aware MSR Codes with Linear Field Size and Smaller Sub-Packetization for Tolerating Multiple Erasures
In an $(n,k,d)$ rack-aware storage model, the system consists of $n$ nodes uniformly distributed across $\bar{n}$ successive racks, such that each rack contains $u$ nodes of equal capacity and the reconstructive degree satisfies $k=\bar{k}u+v$ where $0\leq v\leq u-1$. Suppose there are $h\geq1$ failed nodes in a rack (called the host rack). Then together with its surviving nodes, the host rack downloads recovery data from $\bar{d}$ helper racks and repairs its failed nodes. In this paper, we fo…

DiffPF: Differentiable Particle Filtering with Generative Sampling via Conditional Diffusion Models
Ziyu Wan, Lin Zhao
https://arxiv.org/abs/2507.15716 htt…

DiffPF: Differentiable Particle Filtering with Generative Sampling via Conditional Diffusion Models
This paper proposes DiffPF, a differentiable particle filter that leverages diffusion models for state estimation in dynamic systems. Unlike conventional differentiable particle filters, which require importance weighting and typically rely on predefined or low-capacity proposal distributions. DiffPF learns a flexible posterior sampler by conditioning a diffusion model on predicted particles and the current observation. This enables accurate, equally-weighted sampling from complex, high-dimensi…

Capacity Matters: a Proof-of-Concept for Transformer Memorization on Real-World Data
Anton Changalidis, Aki H\"arm\"a
https://arxiv.org/abs/2506.14704

Capacity Matters: a Proof-of-Concept for Transformer Memorization on Real-World Data
This paper studies how the model architecture and data configurations influence the empirical memorization capacity of generative transformers. The models are trained using synthetic text datasets derived from the Systematized Nomenclature of Medicine (SNOMED) knowledge graph: triplets, representing static connections, and sequences, simulating complex relation patterns. The results show that embedding size is the primary determinant of learning speed and capacity, while additional layers provi…

Analyzing the Impact of Strategic Bidding on the Reserve Capacity via a Bi-Level Model
Yun Xu, Yunxiao Bai, Yunyong Zhang, Peng Wang, Xuelin Wang, Jiqun Guo, Kaijun Xie, Rusheng Zhao
https://arxiv.org/abs/2506.20493

Analyzing the Impact of Strategic Bidding on the Reserve Capacity via a Bi-Level Model
The growing integration of renewable energy sources necessitates adequate reserve capacity to maintain power balance. However, in market clearing, power companies with flexible resources may submit strategic bids to maximize profits, potentially compromising system reserves. This paper examines the effects of such strategic behavior by modeling the market as a bi-level problem. The upper level represents a strategic company aiming to maximize profit, while the lower level simulates the system o…

Clustering scientific publications: lessons learned through experiments with a real citation network
Vu Thi Huong, Thorsten Koch
https://arxiv.org/abs/2505.18180

Clustering scientific publications: lessons learned through experiments with a real citation network
Clustering scientific publications can reveal underlying research structures within bibliographic databases. Graph-based clustering methods, such as spectral, Louvain, and Leiden algorithms, are frequently utilized due to their capacity to effectively model citation networks. However, their performance may degrade when applied to real-world data. This study evaluates the performance of these clustering algorithms on a citation graph comprising approx. 700,000 papers and 4.6 million citations ex…

This https://arxiv.org/abs/2412.03970 has been replaced.
initial toot: https://mastoxiv.page/@ar…

A Data-Driven Framework for Discovering Fractional Differential Equations in Complex Systems
In complex physical systems, conventional differential equations often fall short in capturing non-local and memory effects, as they are limited to local dynamics and integer-order interactions. This study introduces a stepwise data-driven framework for discovering fractional differential equations (FDEs) directly from data. FDEs, known for their capacity to model non-local dynamics with fewer parameters than integer-order derivatives, can represent complex systems with long-range interactions.…

A Foundation Model for Massive MIMO Precoding with an Adaptive per-User Rate-Power Tradeoff
J\'er\^ome Emery, Ali Hasanzadeh Karkan, Jean-Fran\c{c}ois Frigon, Fran\c{c}ois Leduc-Primeau
https://arxiv.org/abs/2507.18587

A Foundation Model for Massive MIMO Precoding with an Adaptive per-User Rate-Power Tradeoff
Deep learning (DL) has emerged as a solution for precoding in massive multiple-input multiple-output (mMIMO) systems due to its capacity to learn the characteristics of the propagation environment. However, training such a model requires high-quality, local datasets at the deployment site, which are often difficult to collect. We propose a transformer-based foundation model for mMIMO precoding that seeks to minimize the energy consumption of the transmitter while dynamically adapting to per-use…

Machine Learning-Driven Performance Analysis of Compressed Communication in Aerial-RIS Networks for Future 6G Networks
Muhammad Farhan Khan, Muhammad Ahmed Mohsin, Zeeshan Alam, Muhammad Saad, Muhammad Waqar
https://arxiv.org/abs/2508.01911

Machine Learning-Driven Performance Analysis of Compressed Communication in Aerial-RIS Networks for Future 6G Networks
In the future 6G and wireless networks, particularly in dense urban environments, bandwidth exhaustion and limited capacity pose significant challenges to enhancing data rates. We introduce a novel system model designed to improve the data rate of users in next-generation multi-cell networks by integrating Unmanned Aerial Vehicle (UAV)-Assisted Reconfigurable Intelligent Surfaces (RIS), Non-Orthogonal Multiple Access (NOMA), and Coordinated Multipoint Transmission (CoMP). Optimally deploying Ae…

Hidden Bose-Einstein Singularities in Correlated Electron Systems: III. Thermodynamic Signals
Takafumi Kita
https://arxiv.org/abs/2506.19211 https://

Hidden Bose-Einstein Singularities in Correlated Electron Systems: III. Thermodynamic Signals
We study thermodynamic consequences of the hidden Bose-Einstein singularities, which have been predicted to cause a pseudogap phase, based on quantum field theory of ordered phases. Starting from the Luttinger-Ward functional for the grand thermodynamic potential, we derive expressions of the Helmholtz free energy, internal energy, entropy, and heat capacity in a form suitable for numerical studies. They are applied to the weakly attractive Hubbard model in three dimensions to calculate the the…

Capacity Based Design of Slot Array Antennas
Volodymyr Shyianov, Bamelak Tadele, Vladimir I. Okhmatovski, Faouzi Bellili, Amine Mezghani
https://arxiv.org/abs/2506.17845

Capacity Based Design of Slot Array Antennas
Historically, the design of antenna arrays has evolved separately from Shannon theory. Shannon theory adopts a probabilistic approach in the design of communication systems, while antenna design approaches have relied on deterministic Maxwell theory alone. In this paper, we introduce a new approach to the design of antenna arrays based on information theoretic metrics. To this end, we develop a statistical model suitable for the numerical optimization of antenna systems. The model is utilized t…

AI, AGI, and learning efficiency
My 4-month-old kid is not DDoSing Wikipedia right now, nor will they ever do so before learning to speak, read, or write. Their entire "training corpus" will not top even 100 million "tokens" before they can speak & understand language, and do so with real intentionally.
Just to emphasize that point: 100 words-per-minute times 60 minutes-per-hour times 12 hours-per-day times 365 days-per-year times 4 years is a mere 105,120,000 words. That's a ludicrously *high* estimate of words-per-minute and hours-per-day, and 4 years old (the age of my other kid) is well after basic speech capabilities are developed in many children, etc. More likely the available "training data" is at least 1 or 2 orders of magnitude less than this.
The point here is that large language models, trained as they are on multiple *billions* of tokens, are not developing their behavioral capabilities in a way that's remotely similar to humans, even if you believe those capabilities are similar (they are by certain very biased ways of measurement; they very much aren't by others). This idea that humans must be naturally good at acquiring language is an old one (see e.g. #AI #LLM #AGI

MiCoTA: Bridging the Learnability Gap with Intermediate CoT and Teacher Assistants
Dongyi Ding, Tiannan Wang, Chenghao Zhu, Meiling Tao, Yuchen Eleanor Jiang, Wangchunshu Zhou
https://arxiv.org/abs/2507.01887

MiCoTA: Bridging the Learnability Gap with Intermediate CoT and Teacher Assistants
Large language models (LLMs) excel at reasoning tasks requiring long thought sequences for planning, reflection, and refinement. However, their substantial model size and high computational demands are impractical for widespread deployment. Yet, small language models (SLMs) often struggle to learn long-form CoT reasoning due to their limited capacity, a phenomenon we refer to as the "SLMs Learnability Gap". To address this, we introduce \textbf{Mi}d-\textbf{Co}T \textbf{T}eacher \textbf{A}ssist…

A Mean Field Game for Capacity Expansion Modeling
Emma Hubert, Dimitrios Lolas, Ronnie Sircar
https://arxiv.org/abs/2507.10604 https://

A Mean Field Game for Capacity Expansion Modeling
This paper studies the optimal investment behavior of renewable electricity producers in a competitive market, where both prices and installation costs are influenced by aggregate industry activity. We model the resulting crowding effects using a mean field game framework, capturing the strategic interactions among a continuum of heterogeneous producers. The equilibrium dynamics are characterized via a coupled system of Hamilton-Jacobi-Bellman and Fokker-Planck equations, which describe the val…

Theoretical Analysis of Near-Field MIMO Channel Capacity and Mid-Band Experimental Validation
Haiyang Miao, Jianhua Zhang, Pan Tang, Heng Wang, Lei Tian, Guangyi Liu
https://arxiv.org/abs/2506.15972

Theoretical Analysis of Near-Field MIMO Channel Capacity and Mid-Band Experimental Validation
With the increase of multiple-input-multiple-output (MIMO) array size and carrier frequency, near-field MIMO communications will become crucial in 6G wireless networks. Due to the increase of MIMO near-field range, the research of near-field MIMO capacity has aroused wide interest. In this paper, we focus on the theoretical analysis and empirical study of near-field MIMO capacity. First, the near-field channel model is characterized from the electromagnetic information perspective. Second, with…

Integrating Upstream Supply Chains into Generation Expansion Planning
Boyu Yao, Andrey Bernstein, Yury Dvorkin
https://arxiv.org/abs/2508.03001 https://arx…

Integrating Upstream Supply Chains into Generation Expansion Planning
Rising electricity demand underscores the need for secure and reliable generation expansion planning that accounts for upstream supply chain constraints. Traditional models often overlook limitations in materials, manufacturing capacity, lead times for deployment, and field availability, which can delay availability of planned resources and thus to threaten system reliability. This paper introduces a multi-stage supply chain-constrained generation expansion planning (SC-GEP) model that optimize…

Distributionally Robust Optimization for Aerial Multi-access Edge Computing via Cooperation of UAVs and HAPs
Ziye Jia, Can Cui, Chao Dong, Qihui Wu, Zhuang Ling, Dusit Niyato, Zhu Han
https://arxiv.org/abs/2506.01972

Distributionally Robust Optimization for Aerial Multi-access Edge Computing via Cooperation of UAVs and HAPs
With an extensive increment of computation demands, the aerial multi-access edge computing (MEC), mainly based on unmanned aerial vehicles (UAVs) and high altitude platforms (HAPs), plays significant roles in future network scenarios. In detail, UAVs can be flexibly deployed, while HAPs are characterized with large capacity and stability. Hence, in this paper, we provide a hierarchical model composed of an HAP and multi-UAVs, to provide aerial MEC services. Moreover, considering the errors of c…

NetRoller: Interfacing General and Specialized Models for End-to-End Autonomous Driving
Ren Xin, Hongji Liu, Xiaodong Mei, Wenru Liu, Maosheng Ye, Zhili Chen, Jun Ma
https://arxiv.org/abs/2506.14589

NetRoller: Interfacing General and Specialized Models for End-to-End Autonomous Driving
Integrating General Models (GMs) such as Large Language Models (LLMs), with Specialized Models (SMs) in autonomous driving tasks presents a promising approach to mitigating challenges in data diversity and model capacity of existing specialized driving models. However, this integration leads to problems of asynchronous systems, which arise from the distinct characteristics inherent in GMs and SMs. To tackle this challenge, we propose NetRoller, an adapter that incorporates a set of novel mechan…

TGIF: Talker Group-Informed Familiarization of Target Speaker Extraction
Tsun-An Hsieh, Minje Kim
https://arxiv.org/abs/2507.14044 https://

TGIF: Talker Group-Informed Familiarization of Target Speaker Extraction
State-of-the-art target speaker extraction (TSE) systems are typically designed to generalize to any given mixing environment, necessitating a model with a large enough capacity as a generalist. Personalized speech enhancement could be a specialized solution that adapts to single-user scenarios, but it overlooks the practical need for customization in cases where only a small number of talkers are involved, e.g., TSE for a specific family. We address this gap with the proposed concept, talker g…

Variable Min-Cut Max-Flow Bounds and Algorithms in Finite Regime
Rivka Gitik, Alejandro Cohen
https://arxiv.org/abs/2507.14852 https://

Variable Min-Cut Max-Flow Bounds and Algorithms in Finite Regime
The maximum achievable capacity from source to destination in a network is limited by the min-cut max-flow bound; this serves as a converse limit. In practice, link capacities often fluctuate due to dynamic network conditions. In this work, we introduce a novel analytical framework that leverages tools from computational geometry to analyze throughput in heterogeneous networks with variable link capacities in a finite regime. Within this model, we derive new performance bounds and demonstrate t…

Multi-type logistic branching processes with selection: frequency process and genealogy for large carrying capacities
Marta Dai Pra, Julian Kern
https://arxiv.org/abs/2507.12601

Multi-type logistic branching processes with selection: frequency process and genealogy for large carrying capacities
We present a model for growth in a multi-species population. We consider two types evolving as a logistic branching process with mutation, where one of the types has a selective advantage, and are interested in the regime in which the carrying capacity of the system goes to $\infty$. We first study the genealogy of the population up until it almost reaches carrying capacity through a coupling with an independent branching process. We then focus on the phase in which the population has reached c…

Group then Scale: Dynamic Mixture-of-Experts Multilingual Language Model
Chong Li, Yingzhuo Deng, Jiajun Zhang, Chengqing Zong
https://arxiv.org/abs/2506.12388

Group then Scale: Dynamic Mixture-of-Experts Multilingual Language Model
The curse of multilinguality phenomenon is a fundamental problem of multilingual Large Language Models (LLMs), where the competition between massive languages results in inferior performance. It mainly comes from limited capacity and negative transfer between dissimilar languages. To address this issue, we propose a method to dynamically group and scale up the parameters of multilingual LLM while boosting positive transfer among similar languages. Specifically, the model is first tuned on monol…

ModFus-DM: Explore the Representation in Modulated Signal Diffusion Generated Models
Haoyue Tan, Yu Li, Zhenxi Zhang, Xiaoran Shi, Feng Zhou
https://arxiv.org/abs/2508.01719 htt…

ModFus-DM: Explore the Representation in Modulated Signal Diffusion Generated Models
Automatic modulation classification (AMC) is essential for wireless communication systems in both military and civilian applications. However, existing deep learning-based AMC methods often require large labeled signals and struggle with non-fixed signal lengths, distribution shifts, and limited labeled signals. To address these challenges, we propose a modulation-driven feature fusion via diffusion model (ModFus-DM), a novel unsupervised AMC framework that leverages the generative capacity of …

A Joint Planning Model for Fixed and Mobile Electric Vehicle Charging Stations Considering Flexible Capacity Strategy
Zhe Yu, Xue Hu, Qin Wang
https://arxiv.org/abs/2507.17587 h…

A Joint Planning Model for Fixed and Mobile Electric Vehicle Charging Stations Considering Flexible Capacity Strategy
The widespread adoption of electric vehicles (EVs) has significantly increased demand on both transportation and power systems, posing challenges to their stable operation. To support the growing need for EV charging, both fixed charging stations (FCSs) and mobile charging stations (MCSs) have been introduced, serving as key interfaces between the power grid and traffic network. Recognizing the importance of collaborative planning across these sectors, this paper presents a two-stage joint plan…

Hierarchical Diffusion Framework for Pseudo-Healthy Brain MRI Inpainting with Enhanced 3D Consistency
Dou Hoon Kwark, Shirui Luo, Xiyue Zhu, Yudu Li, Zhi-Pei Liang, Volodymyr Kindratenko
https://arxiv.org/abs/2507.17911

Hierarchical Diffusion Framework for Pseudo-Healthy Brain MRI Inpainting with Enhanced 3D Consistency
Pseudo-healthy image inpainting is an essential preprocessing step for analyzing pathological brain MRI scans. Most current inpainting methods favor slice-wise 2D models for their high in-plane fidelity, but their independence across slices produces discontinuities in the volume. Fully 3D models alleviate this issue, but their high model capacity demands extensive training data for reliable, high-fidelity synthesis -- often impractical in medical settings. We address these limitations with a hi…

Preventing an Extractive Green Hydrogen Industry: Risks and Benefits of Grid Expansion and Green Hydrogen in and for Kenya
Xi Xi, Boniface Kinyanjui, Daniel M. Kammen
https://arxiv.org/abs/2507.14756

Preventing an Extractive Green Hydrogen Industry: Risks and Benefits of Grid Expansion and Green Hydrogen in and for Kenya
This study evaluates the role of grid-connected hydrogen electrolyzers in advancing a cost-effective and in particular an equitable green hydrogen industry in Kenya to serve both domestic and international needs and markets. Using a multi-nodal capacity expansion model with county-level spatial resolution, we assess how electrolyzer deployment affects electricity cost, grid flexibility, and carbon intensity under various renewable and demand scenarios. Results show that electrolyzers enable up …

Megrez2 Technical Report
Boxun Li, Yadong Li, Zhiyuan Li, Congyi Liu, Weilin Liu, Guowei Niu, Zheyue Tan, Haiyang Xu, Zhuyu Yao, Tao Yuan, Dong Zhou, Yueqing Zhuang, Bo Zhao, Guohao Dai, Yu Wang
https://arxiv.org/abs/2507.17728

Megrez2 Technical Report
We present Megrez2, a novel lightweight and high-performance language model architecture optimized for device native deployment. Megrez2 introduces a novel cross-layer expert sharing mechanism, which significantly reduces total parameter count by reusing expert modules across adjacent transformer layers while maintaining most of the model's capacity. It also incorporates pre-gated routing, enabling memory-efficient expert loading and faster inference. As the first instantiation of the Megrez2 a…

Conformable Scaling and Critical Dynamics: A Unified Framework for Phase Transitions
Jos\'e Weberszpil
https://arxiv.org/abs/2507.11782 https://…

Conformable Scaling and Critical Dynamics: A Unified Framework for Phase Transitions
We investigate the application of conformable derivatives to model critical phenomena near continuous phase transitions. By incorporating a deformation parameter into the differential structure, we derive unified expressions for thermodynamic observables such as heat capacity, magnetization, susceptibility, and coherence length, each exhibiting power-law behavior near the critical temperature. The conformable derivative framework naturally embeds scale invariance and critical slowing down into …

This https://arxiv.org/abs/2505.08944 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csDC_…

Toward Cost-Efficient Serving of Mixture-of-Experts with Asynchrony
Mixture-of-Experts (MoE) architectures offer the promise of larger model capacity without the prohibitive costs of fully dense designs. However, in real-world inference serving, load skew across experts often leads to suboptimal device utilization and excessive synchronization overheads. This paper introduces Asynchronous Expert Parallelism (AEP), a new paradigm that decouples layer execution from barrier-style synchronization. By dynamically queuing tokens at each layer (referred to as $μ$-qu…

Improving Outbreak Forecasts Through Model Augmentation
Graham C. Gibson, Spencer J. Fox, Emily Javan, Susan E. Ptak, Oluwasegun M. Ibrahim, Michael Lachmann, Lauren Ancel Meyers
https://arxiv.org/abs/2506.16410

Improving Outbreak Forecasts Through Model Augmentation
Accurate forecasts of disease outbreaks are critical for effective public health responses, management of healthcare surge capacity, and communication of public risk. There are a growing number of powerful forecasting methods that fall into two broad categories -- empirical models that extrapolate from historical data, and mechanistic models based on fixed epidemiological assumptions. However, these methods often underperform precisely when reliable predictions are most urgently needed -- durin…

Adaptive Robust Optimization for European Electricity System Planning Considering Regional Dunkelflaute Events
Maximilian Bernecker, Smaranda Sgarciu, Xiaoming Kan, Mehrnaz Anvari, Iegor Riepin, Felix M\"usgens
https://arxiv.org/abs/2507.11361

Adaptive Robust Optimization for European Electricity System Planning Considering Regional Dunkelflaute Events
This study develops a capacity expansion model for a fully decarbonized European electricity system using an Adaptive Robust Optimization (ARO) framework. The model endogenously identifies the worst regional Dunkelflaute events, prolonged periods of low wind and solar availability, and incorporates multiple extreme weather realizations within a single optimization run. Results show that system costs rise nonlinearly with the geographic extent of these events: a single worst-case regional disrup…

Grids Often Outperform Implicit Neural Representations
Namhoon Kim, Sara Fridovich-Keil
https://arxiv.org/abs/2506.11139 https://arxi…

Grids Often Outperform Implicit Neural Representations
Implicit Neural Representations (INRs) have recently shown impressive results, but their fundamental capacity, implicit biases, and scaling behavior remain poorly understood. We investigate the performance of diverse INRs across a suite of 2D and 3D real and synthetic signals with varying effective bandwidth, as well as both overfitting and generalization tasks including tomography, super-resolution, and denoising. By stratifying performance according to model size as well as signal type and ba…

Can Mixture-of-Experts Surpass Dense LLMs Under Strictly Equal Resources?
Houyi Li, Ka Man Lo, Ziqi Wang, Zili Wang, Wenzhen Zheng, Shuigeng Zhou, Xiangyu Zhang, Daxin Jiang
https://arxiv.org/abs/2506.12119

Can Mixture-of-Experts Surpass Dense LLMs Under Strictly Equal Resources?
Mixture-of-Experts (MoE) language models dramatically expand model capacity and achieve remarkable performance without increasing per-token compute. However, can MoEs surpass dense architectures under strictly equal resource constraints - that is, when the total parameter count, training compute, and data budget are identical? This question remains under-explored despite its significant practical value and potential. In this paper, we propose a novel perspective and methodological framework to …

Dynamic Parameter Memory: Temporary LoRA-Enhanced LLM for Long-Sequence Emotion Recognition in Conversation
Jialong Mai, Xiaofen Xing, Yawei Li, Zhipeng Li, Jingyuan Xing, Xiangmin Xu
https://arxiv.org/abs/2507.09076

Dynamic Parameter Memory: Temporary LoRA-Enhanced LLM for Long-Sequence Emotion Recognition in Conversation
Recent research has focused on applying speech large language model (SLLM) to improve speech emotion recognition (SER). However, the inherently high frame rate in speech modality severely limits the signal processing and understanding capabilities of SLLM. For example, a SLLM with a 4K context window can only process 80 seconds of audio at 50Hz feature sampling rate before reaching its capacity limit. Input token compression methods used in SLLM overlook the continuity and inertia of emotions a…