Widespread adoption of AI systems hinges on their ability to generate economic value that outweighs their inference costs. Evaluating this tradeoff requires metrics accounting for both performance and costs. Building on production theory, we develop an economically grounded framework to evaluate language models' productivity by combining accuracy and inference cost. We formalize cost-of-pass: the expected monetary cost of generating a correct solution. We then define the frontier cost-of-pass: the minimum cost-of-pass achievable across available models or the human-expert(s), using the approx. cost of hiring an expert. Our analysis reveals distinct economic insights. First, lightweight models are most cost-effective for basic quantitative tasks, large models for knowledge-intensive ones, and reasoning models for complex quantitative problems, despite higher per-token costs. Second, tracking the frontier cost-of-pass over the past year reveals significant progress, particularly for complex quant. tasks where the cost roughly halved every few months. Third, to trace key innovations driving this progress, we examine counterfactual frontiers -- estimates of cost-efficiency without spec
Log data plays a critical role in observability, debugging, and performance monitoring in modern cloud-native systems. In small and early-stage cloud deployments, however, log retention policies are frequently configured far beyond operational requirements, often defaulting to 90 days or more, without explicit consideration of their financial and performance implications. As a result, excessive log retention becomes a hidden and recurring cost. This study examines the financial and operational impact of log retention window selection from a cost-aware perspective. Using synthetic log datasets designed to reflect real-world variability in log volume and access patterns, we evaluate retention windows of 7, 14, 30, and 90 days. The analysis focuses on three metrics: storage cost, operationally useful log ratio, and cost per useful log. Operational usefulness is defined as log data accessed during simulated debugging and incident analysis tasks. The results show that reducing log retention from 90 days to 14 days can lower log storage costs by up to 78 percent while preserving more than 97 percent of operationally useful logs. Longer retention windows provide diminishing operational re
Multi-party computation (MPC) based machine learning, referred to as multi-party learning (MPL), has become an important technology for utilizing data from multiple parties with privacy preservation. In recent years, in order to apply MPL in more practical scenarios, various MPC-friendly models have been proposedto reduce the extraordinary communication overhead of MPL. Within the optimization of MPC-friendly models, a critical element to tackle the challenge is profiling the communication cost of models. However, the current solutions mainly depend on manually establishing the profiles to identify communication bottlenecks of models, often involving burdensome human efforts in a monotonous procedure. In this paper, we propose HawkEye, a static model communication cost profiling framework, which enables model designers to get the accurate communication cost of models in MPL frameworks without dynamically running the secure model training or inference processes on a specific MPL framework. Firstly, to profile the communication cost of models with complex structures, we propose a static communication cost profiling method based on a prefix structure that records the function calling
Low-pressure radio-frequency capacitively coupled plasmas operated in Ar/O$_2$ gas mixtures are widely adopted in critical semiconductor manufacturing processes. O($^3$P) and O($^1$D) are key highly reactive species for oxidation or as oxygen sources for deposited thin films. Optimizing external parameters to realize efficient generation of these species under limited energy deposition is essential for improving process yield.Based on a one-dimensional (1D) fluid/electron Monte Carlo (EMC) hybrid model, this study investigates the energy cost of O($^1$D) and O($^3$P) generation driven by sawtooth up-type voltage waveforms at a fixed peak-to-peak voltage, focusing on the effects of the harmonic number ($N$) and the O$_2$ ratio. The results show that O($^3$P) generation is consistently more efficient than that of O($^1$D). The generation energy cost decreases with increasing O$_2$ ratio, yet increases as $N$ increases. However, in the specific scenario of 10% O$_2$, an inflection point can be observed at $N = 2$. As $N$ increases from 1 to 2, the discharge mode shifts from the DA mode to the $α$-DA hybrid mode, expanding the effective spatio-temporal range of the ionization rate and
Minimum cost homomorphism problems can be viewed as a generalization of list homomorphism problems. They also extend two well-known graph colouring problems: the minimum colour sum problem and the optimum cost chromatic partition problem. In both of these problems, the cost function meets an additional constraint: the cost of using a specific colour is the same for every vertex of the input graph. We study minimum cost homomorphism problems with cost functions constrained to have this property. Clearly, when the standard minimum cost homomorphism problem is polynomial, then the problem with constrained costs is also polynomial. We expect that the same may hold for the cases when the standard minimum cost homomorphism problem is NP-complete. We prove that this is the case for trees $H$: we obtain a dichotomy of minimum constrained cost homomorphism problems which coincides with the dichotomy of standard minimum cost homomorphism problems. For general graphs $H$, we prove a partial dichotomy: the problem is polynomial if $H$ is a proper interval graph and NP-complete when $H$ is not chordal bipartite.
Large language models (LLMs) such as GPT-4o and Claude Sonnet 4.5 have demonstrated strong capabilities in open-ended reasoning and generative language tasks, leading to their widespread adoption across a broad range of NLP applications. However, for structured text classification problems with fixed label spaces, model selection is often driven by predictive performance alone, overlooking operational constraints encountered in production systems. In this work, we present a systematic comparison of two contrasting paradigms for text classification: zero- and few-shot prompt-based large language models, and fully fine-tuned encoder-only architectures. We evaluate these approaches across four canonical benchmarks (IMDB, SST-2, AG News, and DBPedia), measuring predictive quality (macro F1), inference latency, and monetary cost. We frame model evaluation as a multi-objective decision problem and analyze trade-offs using Pareto frontier projections and a parameterized utility function reflecting different deployment regimes. Our results show that fine-tuned encoder-based models from the BERT family achieve competitive, and often superior, classification performance while operating at on
Microservices architecture, known for its agility and efficiency, is an ideal framework for cloud-based software development and deployment. When integrated with containerization and orchestration systems, resource management becomes more streamlined. However, cloud computing costs remain a critical concern, necessitating effective strategies to minimize expenses without compromising performance. Cloud platforms like AWS offer transient pricing options, such as Spot Pricing, to reduce operational costs. However, unpredictable demand and abrupt termination of spot VMs introduce challenges. By leveraging containerization and intelligent orchestration, microservices deployment costs can be optimized while maintaining performance requirements. We present SpotKube, an open-source, Kubernetes-based solution that employs a genetic algorithm for cost optimization. Designed to dynamically scale clusters for microservice applications on public clouds using spot pricing, SpotKube analyzes application characteristics to recommend optimal resource allocations. This ensures cost-effective deployments without sacrificing performance. Its elastic cluster autoscaler adapts to changing demands, grac
The \emph{file caching} problem is defined as follows. Given a cache of size $k$ (a positive integer), the goal is to minimize the total retrieval cost for the given sequence of requests to files. A file $f$ has size $size(f)$ (a positive integer) and retrieval cost $cost(f)$ (a non-negative number) for bringing the file into the cache. A \emph{miss} or \emph{fault} occurs when the requested file is not in the cache and the file has to be retrieved into the cache by paying the retrieval cost, and some other file may have to be removed (\emph{evicted}) from the cache so that the total size of the files in the cache does not exceed $k$. We study the following variants of the online file caching problem. \textbf{\emph{Caching with Rental Cost} (or \emph{Rental Caching})}: There is a rental cost $λ$ (a positive number) for each file in the cache at each time unit. The goal is to minimize the sum of the retrieval costs and the rental costs. \textbf{\emph{Caching with Zapping}}: A file can be \emph{zapped} by paying a zapping cost $N \ge 1$. Once a file is zapped, all future requests of the file don't incur any cost. The goal is to minimize the sum of the retrieval costs and the zapping
The inception of AI-based fraud detection systems has presented the banking sector across the globe the opportunity to enhance fraud prevention mechanisms. However, the extent of adoption in Nigeria has been slow, fragmented, and inconsistent due to high cost of implementation and lack of technical expertise. This study seeks to investigate extent of adoption and determinants of AI-driven fraud detection systems in Nigerian banks. This study adopted a cross-sectional survey research design. Data were extracted from primary sources through structured questionnaire based on 5-point Likert scale. The population of the study consist of 24 licensed banks in Nigeria. A purposive sampling technique was used to select 5 biggest banks based on market capitalization and customer base. The Ordered Logistic Regression (OLR) model was used to estimate the data. The results showed that top management support, IT infrastructure, regulatory compliance, staff competency and perceived effectiveness accelerate the uptake of AI-driven fraud detection systems adoption. However, high implementation cost discourages it. Therefore, the study recommended that banks should invest in modern and scalable IT s
Particle Image Velocimetry (PIV) is fundamental to fluid dynamics, yet deep learning applications face significant hurdles. A critical gap exists: the lack of comprehensive evaluation of how diverse optical flow models perform specifically on PIV data, largely due to limitations in available datasets and the absence of a standardized benchmark. This prevents fair comparison and hinders progress. To address this, our primary contribution is a novel, large-scale synthetic PIV benchmark dataset generated from diverse CFD simulations (JHTDB and Blasius). It features unprecedented variety in particle densities, flow velocities, and continuous motion, enabling, for the first time, a standardized and rigorous evaluation of various optical flow and PIV algorithms. Complementing this, we propose Multi Cost Volume PIV (MCFormer), a new deep network architecture leveraging multi-frame temporal information and multiple cost volumes, specifically designed for PIV's sparse nature. Our comprehensive benchmark evaluation, the first of its kind, reveals significant performance variations among adapted optical flow models and demonstrates that MCFormer significantly outperforms existing methods, ach
Enforcing local consistencies in cost function networks is performed by applying so-called Equivalent Preserving Transformations (EPTs) to the cost functions. As EPTs transform the cost functions, they may break the property that was making local consistency enforcement tractable on a global cost function. A global cost function is called tractable projection-safe when applying an EPT to it is tractable and does not break the tractability property. In this paper, we prove that depending on the size r of the smallest scopes used for performing EPTs, the tractability of global cost functions can be preserved (r = 0) or destroyed (r > 1). When r = 1, the answer is indefinite. We show that on a large family of cost functions, EPTs can be computed via dynamic programming-based algorithms, leading to tractable projection-safety. We also show that when a global cost function can be decomposed into a Berge acyclic network of bounded arity cost functions, soft local consistencies such as soft Directed or Virtual Arc Consistency can directly emulate dynamic programming. These different approaches to decomposable cost functions are then embedded in a solver for extensive experiments that c
Feature Selection is a crucial procedure in Data Science tasks such as Classification, since it identifies the relevant variables, making thus the classification procedures more interpretable, cheaper in terms of measurement and more effective by reducing noise and data overfit. The relevance of features in a classification procedure is linked to the fact that misclassifications costs are frequently asymmetric, since false positive and false negative cases may have very different consequences. However, off-the-shelf Feature Selection procedures seldom take into account such cost-sensitivity of errors. In this paper we propose a mathematical-optimization-based Feature Selection procedure embedded in one of the most popular classification procedures, namely, Support Vector Machines, accommodating asymmetric misclassification costs. The key idea is to replace the traditional margin maximization by minimizing the number of features selected, but imposing upper bounds on the false positive and negative rates. The problem is written as an integer linear problem plus a quadratic convex problem for Support Vector Machines with both linear and radial kernels. The reported numerical experien
Cloud computing has revolutionized the way organizations manage their IT infrastructure, but it has also introduced new challenges, such as managing cloud costs. The rapid adoption of artificial intelligence (AI) and machine learning (ML) workloads has further amplified these challenges, with GPU compute now representing 40-60\% of technical budgets for AI-focused organizations. This paper provides a comprehensive review of cloud and AI infrastructure cost optimization techniques, covering traditional cloud pricing models, resource allocation strategies, and emerging approaches for managing AI/ML workloads. We examine the dramatic cost reductions in large language model (LLM) inference which has decreased by approximately 10x annually since 2021 and explore techniques such as model quantization, GPU instance selection, and inference optimization. Real-world case studies from Amazon Prime Video, Pinterest, Cloudflare, and Netflix showcase practical application of these techniques. Our analysis reveals that organizations can achieve 50-90% cost savings through strategic optimization approaches. Future research directions in automated optimization, sustainability, and AI-specific cost
This paper is a preliminary report of the research plan and a digest of the results and discussions. On research note explores the complex dynamics of fake news dissemination and fact-checking costs within the framework of information markets and analyzes the equilibrium between supply and demand using the concepts of droop quotas, Meek's method, and marginal contributions. By adopting a two-sided matching market perspective, we delve into scenarios in which markets are stable under the influence of fake news perceived as truth and those in which credibility prevails. Through the application of iterated dilemma game theory, we investigate the strategic choices of news providers affected by the costs associated with spreading fake news and fact-checking efforts. We further examine the maximum reward problem and strategies to minimize the cost path for spreading fake news, and consider a nuanced understanding of market segmentation into "cheap" and "premium" segments based on the nature of the information being spread. Our analysis uses mathematical models and computational processes to identify stable equilibrium points that ensure market stability in the face of deceptive informati
Many solutions to cost-sensitive classification (and regression) rely on some or all of the following assumptions: we have complete knowledge about the cost context at training time, we can easily re-train whenever the cost context changes, and we have technique-specific methods (such as cost-sensitive decision trees) that can take advantage of that information. In this paper we address the problem of selecting models and minimising joint cost (integrating both misclassification cost and test costs) without any of the above assumptions. We introduce methods and plots (such as the so-called JROC plots) that can work with any off-the-shelf predictive technique, including ensembles, such that we reframe the model to use the appropriate subset of attributes (the feature configuration) during deployment time. In other words, models are trained with the available attributes (once and for all) and then deployed by setting missing values on the attributes that are deemed ineffective for reducing the joint cost. As the number of feature configuration combinations grows exponentially with the number of features we introduce quadratic methods that are able to approximate the optimal configura
The digitalisation of the modern schooling system has led to multiple schools and organisations buying similar hardware. Electronic equipment like wireless microphones, projectors, touchscreen displays etc., have been almost standardised with a few well-known brands leading the market. This has led to the adoption of common frequency ranges between brands with many sticking between 600-670 MHz. The popularity of low-cost computing devices like the Raspberry Pi which has been used in a plethora of applications has also taken the path of being used as low-cost transmitters. There have been many implementations where the Raspberry Pi has been used as the target device but few cases where the PI is the actual threat. In this paper, we explore the use of the Raspberry Pi as a stealth radio frequency jamming device to disable wireless conference microphones. Harmonics were used to achieve frequencies outside the Pi's transmission frequency by taking advantage of its unfiltered transmission.
Quantum heat engines are commonly believed to achieve their optimal efficiency only when operated quasi-statically. When running at finite power, however, they suffer effective friction due to the generation of coherences and transitions between energy eigenstates. It was noted that it is possible to increase the power of a quantum heat engine using external control schemes or suitable dephasing noise. Here, we investigate the thermodynamic cost associated with dephasing noise schemes using both numerical and analytical methods. Our findings unveil that the observed gain in power is generally not free of thermodynamic costs, as it involves energy costs of the control fields or heat flows between thermal and dephasing baths. These contributions must be duly accounted for when determining the engine's overall efficiency. Interestingly, we identify a particular working regime where these costs become negligible, demonstrating that quantum heat engines can be operated at any power with an efficiency per cycle that approaches arbitrarily closely that under quasistatic operation.
This research note is organized with respect to a novel approach to solving problems related to the spread of fake news and effective fact-checking. Focusing on the least-cost routing problem, the discussion is organized with respect to the use of Metzler functions and Metzler matrices to model the dynamics of information propagation among news providers. With this approach, we designed a strategy to minimize the spread of fake news, which is detrimental to informational health, while at the same time maximizing the spread of credible information. In particular, through the punitive dominance problem and the maximum compensation problem, we developed and examined a path to reassess the incentives of news providers to act and to analyze their impact on the equilibrium of the information market. By applying the concept of entanglement to the context of information propagation, we shed light on the complexity of interactions among news providers and contribute to the formulation of more effective information management strategies. This study provides new theoretical and practical insights into issues related to fake news and fact-checking, and will be examined against improving inform
The intersection of quantum theory and accounting presents a novel and intriguing frontier in exploring financial valuation and accounting practices. This paper applies quantum theory to cost accounting, specifically Work in Progress (WIP) valuation. WIP is conceptualized as materials in a quantum superposition state whose financial value remains uncertain until observed or measured. This work comprehensively reviews the seminal works that explored the overlap between quantum theory and accounting. The primary contribution of this work is a more nuanced understanding of the uncertainties involved, which emerges by applying quantum phenomena to model the complexities and uncertainties inherent in managerial accounting. In contrast, previous works focus more on financial accounting or general accountancy.
The construction of quantum computers is based on the synthesis of low-cost quantum circuits. The quantum circuit of any Boolean function expressed in a Positive Polarity Reed-Muller $PPRM$ expansion can be synthesized using Multiple-Control Toffoli ($MCT$) gates. This paper proposes two algorithms to construct a quantum circuit for any Boolean function expressed in a Positive Polarity Reed-Muller $PPRM$ expansion. The Boolean function can be expressed with various algebraic forms, so there are different quantum circuits can be synthesized for the Boolean function based on its algebraic form. The proposed algorithms aim to map the $MCT$ gates into the $NCV$ gates for any quantum circuit by generating a simple algebraic form for the Boolean function. The first algorithm generates a special algebraic form for any Boolean function by rearrangement of terms of the Boolean function according to a predefined degree of term $d_{term}$, then synthesizes the corresponding quantum circuit. The second algorithm applies the decomposition methods to decompose $MCT$ circuit into its elementary gates followed by applying a set of simplification rules to simplify and optimize the synthesized quant