Francisco Ribeiro, Claudio Spiess, Prem Devanbu
et al.
Despite the effectiveness of large language models (LLMs) for code generation, they often output incorrect code. One reason is that model output probabilities are often not well-correlated with correctness, and reflect only the final output of the generation process. Inspired by findings that LLMs internally encode concepts like truthfulness, this paper explores if LLMs similarly represent code correctness. Specifically, we identify a correctness representation inside LLMs by contrasting the hidden states between pairs of correct and incorrect code for the same programming tasks. By experimenting on four LLMs, we show that exploiting this extracted correctness representation outperforms standard log-likelihood ranking, as well as verbalized model confidence. Furthermore, we explore how this internal correctness signal can be used to select higher-quality code samples, without requiring test execution. Ultimately, this work demonstrates how leveraging internal representations can enhance code generation systems and make LLMs more reliable, thus improving confidence in automatically generated code.
he demand for electric vehicles and plug-in hybrids, supported by green technologies and the IT-sector, is making significant changes in the structural transformation of the economy. The coverage of energy and raw materials markets bythe reduction in the consumption of petroleum products does not allow us to assert confidently that tax losses will be compensated by revenues from the increase in electricity consumption, including from the transport infrastructure. This is partly due to different levels of taxation of electricity and oil products. Removing obstacles hindering the development of electric vehicles through fiscal policy is an important prerogative of the state. On the part of the state, timely measures are needed regarding the revision of taxes and tax rates in support of the development of mechanisms for the introduction of environmentally friendly electric transport, plug-in hybrids, adaptive expansion of infrastructure for vehicle categories M, MG, N, NG, as well as minimizing losses from tax maneuvers of foreign trade relations. Investors are more likely to respond to positive market signals, where in many cases government regulation has a significant impact on the sustainable and economically safe formation of the investment climate. The development of electric transport narrows the fleet of cars on internal combustion engines, or on a fossil energy source. The fall in demand for oil products will hold back the replenishment of the budget revenue savings by reducing tax revenues on fuel. In order to avoid loss of economic benefits with the renewal of the car fleet with electric vehicles, the paper examines the mechanism for adapting the motor vehicle service to modern challenges of the global economy.
International institutions may have an important role to play in ensuring advanced AI systems benefit humanity. International collaborations can unlock AI's ability to further sustainable development, and coordination of regulatory efforts can reduce obstacles to innovation and the spread of benefits. Conversely, the potential dangerous capabilities of powerful and general-purpose AI systems create global externalities in their development and deployment, and international efforts to further responsible AI practices could help manage the risks they pose. This paper identifies a set of governance functions that could be performed at an international level to address these challenges, ranging from supporting access to frontier AI systems to setting international safety standards. It groups these functions into four institutional models that exhibit internal synergies and have precedents in existing organizations: 1) a Commission on Frontier AI that facilitates expert consensus on opportunities and risks from advanced AI, 2) an Advanced AI Governance Organization that sets international standards to manage global threats from advanced models, supports their implementation, and possibly monitors compliance with a future governance regime, 3) a Frontier AI Collaborative that promotes access to cutting-edge AI, and 4) an AI Safety Project that brings together leading researchers and engineers to further AI safety research. We explore the utility of these models and identify open questions about their viability.
Arif Triwinarso, S. Susianti, Maria Giovani Putri Leda
Hotel tax is one source of local taxes that will contribute to the potential local revenue (PAD) if managed effectively and can be a source of funding for regional development. The purpose of this study was to determine the trend of realization, the level of hotel tax effectiveness, and its contribution to local taxes and PAD in Ende Regency for the period 2017 - 2020. The research method used was descriptive quantitative with secondary data as the data source. The data was obtained from the Regional Revenue Agency of Ende Regency. The results of the study show that in general, the trend of hotel tax revenue realization and effectiveness has decreased. This is caused by internal factors such as a very high target increase without taking into account the existing potential as well as several external factors such as a low level of public awareness and knowledge. The effectiveness of hotel tax revenue in Ende Regency 3 out of 4 years of observation is in the ineffective category (< 60%). The contribution of hotel taxes to local taxes and Ende Regency's Original Revenue is relatively very small. The factors that cause the hotel tax to be ineffective are the lack of knowledge and awareness of citizens towards taxation, tax management accountability that is not optimal, payments are not by the value they should be, submission of SPTDP not on time, and some taxpayers close their business without notification.
Anastasiia Tkalich, Nils Brede Moe, Rasmus Ulfsnes
With the increasing availability of software usage and the influence of the Lean Startup mindset, more and more companies choose to innovate through internal software startups. Such startups aim at developing new business models while at the same time relying on the resources from the companies where they emerged. The evidence from both researchers and practitioners indicates that driving internal software startups is challenging. This paper seeks to address this problem by asking the research question: how to make internal software startups work? We examined a unique case of a venture builder, a company primarily focusing on building internal software startups and launching them as independent companies. Applying a Grounded Theory approach, we analyzed data on four internal software startups at the case company. The results suggest that four strategies drive the examined startups, cultural, financial, personnel, and venture arrangement. We interpret our results by drawing on earlier literature on intrapreneurship and internal ventures and suggest four recommendations to succeed with internal software startups 1 establish shared arenas for the employees, 2 provide necessary resources for experimentation in the initial phase and increase them incrementally, 3 build up in-house product management competence through coaching, and 4 harness employees own motivation to develop their own ideas.
This paper aims to understand the international practices of tax planning. International companies choose their capital structure according to differences in international taxation, in order to minimize the tax burden of the whole company group. This paper reviews the literature that deals with international tax avoidance techniques by highlighting tax planning measurements in the empirical literature. The methodology used is the narrative approach of literature review, which consists on assembling and synthesizing previously published research. The paper concludes that there are several approaches of international tax planning including transfers of revenues by geographical area, redevelopment of the company, haven and loopholes in tax legislation. Moreover, finding more precise measures of tax planning techniques would be of great value to studies in this respect.,The authors follow the guideline provided by Templier and Pare (2015) in order to select the type of the literature review to use in this paper. Accordingly, this paper employs the narrative approach of literature review, which consists on assembling and synthesizing previously published research on international tax planning. This narrative review will serve as a starting point for future investigations and research developments. The authors rely on a logic of configuration in order to analyze data. This logic consists on addressing then organizing various aspects of international practices of tax planning.,The paper concludes that there are many aspects of international tax planning that need to be covered by future researchers, especially finding more precise measures of tax planning techniques would be of great value to studies in this respect.,The literature survey reveals the following issues. First, few studies have been conducted to date. Second, several approaches remain unexplored, and studies rely only on surveys' results collected from the annual report of companies (microeconomics variables), while macroeconomic variables can better explain the phenomenon of international tax planning. In this context, studies containing proposals to estimate more accurate international companies' tax planning techniques would also be welcome. Previous literature supposes premises on this issue th:at limit the accurateness of the analysis. Particularly, empirical literature is short of the proper measurement to evaluate corporate tax avoidance. This would explain the various interpretations of research findings. Hence, finding more precise measures of tax planning techniques would be of great value to studies in this respect.,This literature survey highlights recent studies dealing with tax planning theories within the framework of corporate governance. This theoretical framework particularly specifies which key variables are the most suitable for measuring tax planning methods and highlights the need to examine how those key variables might differ and under what circumstances. In addition, it underlines limits on tax planning measurements by addressing the comparison of the empirical measurements.,The paper contributes to the literature on internal tax planning in several ways. First, this study is unique in that it constitutes the only literature review that provides a comprehensive overview of research on international tax planning. Especially, it extends previous studies by considering the specific new trend of empirical literature dealing with the techniques of international tax planning. This literature review identifies two categories of tax planning approaches including techniques related to company internal management practices and international tax planning techniques. In addition, the literature survey helps to determine various strategies used by multinationals for tax planning, through an in-depth review of the existing studies. Finally, it provides researchers with a starting point to further explore issues related to tax avoidance techniques.
The “tax gap”— the difference between the amount of taxes owed and the amount of tax actually paid — has garnered widespread attention in recent months. Much of the commentary on the subject equates the tax gap with “tax evasion,” a term broadly understood to connote intentional (and potentially criminal) underreporting . This paper cautions against conflating the tax gap with tax evasion. The tax gap includes substantial gray areas where the law is ambiguous and the Internal Revenue Service’s (IRS’s ) determination of true tax is debatable. T he IRS’s methodology for measu ring the tax gap also includes upward adjustments that are recommended by frontline examiners but reversed on administrative appeal or judicial review. Moreover, a substantial portion of the estimated tax gap is derived from a statistical technique called “detection controlled estimation” that potentially magnifies the impact of later-reversed recommendations on the ultimate tax gap measure. Weighing in the opposite direction, the IRS’s approach to measuring the tax gap excludes some amounts that clearly constitute tax evasion (most significantly, underreporting of tax on illegal-source income). Understanding the tax gap’s shades of gray can inform discussions of tax law and policy. We explain how proposals to use the tax gap as a performance target may produce perverse incentives for the IRS. We further explain how additional IRS funding —though necessary to improve the agency’s ability to enforce tax laws— may have counterintuitive effects on tax gap estimates. We also illustrate — using examples from the taxation of passthrough entities — how legislative reforms can reduce the size and scope of gray areas in the tax code that
This study analyses intergovernmental revenue policies and its practices in Nepal, specifically focusing on the role of local governments (LGs) in tax administration. Both qualitative and quantitative data were used in this study that concluded LGs are gradually institutionalising internal revenue bases, and rates at the local level. The constitution 2015 has provisioned that the local governments can autonomously execute various functions including legislative, administrative, financial management, tax administration and so on. Nonetheless, imperfect experience, inadequate technical skills, limited legal and procedural grounds, and weak institutional capacity, LGs are collecting tax. Result shows tax enforcement process has been implemented with limited consultation with the taxpayers and minimal coordination with the provincial and federal government. This has created number of criticisms to the LGs about their unaccountability to the taxpaying citizens. Thus, some recommendations such as integration of technical management capacity with local, provincial, and federal level for efficient fiscal administration systems; and digitization and automation of taxation for intergovernmental cooperation have been provided for improving local economic and effective tax administration at the local level.
Turbulent mixing processes in deep alpine Lake Garda (I) have not extensively been observed. Knowledge about drivers of turbulent fluxes are important for insights in the transport of matter, nutrients and pollutants, in the lake and in natural water bodies in general. In this paper, the occurrence of internal wave induced turbulent convection is addressed as opposed to the more common shear-induced turbulence in a density stratified environment. Observations are analyzed from a dedicated yearlong mooring holding 100 high-resolution temperature sensors at 1.5 m intervals under a single current meter in the deeper half of the 344 m deep lake-center. Episodically, the weakly density stratified waters in the lower 50 m above the lake floor show spectral slope and coherence evidence of short-term (15 to 30 minutes) convective motions under internal waves that are supported by the stronger stratified waters above. The near-homogeneous conditions are not attributable to frictional Ekman dynamics, but to large-scale internal wave crests.
Fluids with internal microstructure like dense suspensions, biological and polymer added fluids, are commonly found in the turbulent regime in many applications. Their flow is extremely difficult to be studied as microstructure complexity and Reynolds number increase, even nowadays. This bottleneck is novelty and efficiently treated here by the micropolar theory. Our findings support that when denser microstructure occurs turbulence is intensified by the chaotic rotation of internal fluid's elements. Unlike in Newtonian fluids, shear stress is now decreased, and viscous and microstructure interrotational stresses increase near the walls and mark the turbulence intensification.
The internal validity of observational study is often subject to debate. In this study, we define the counterfactuals as the unobserved sample and intend to quantify its relationship with the null hypothesis statistical testing (NHST). We propose the probability of a causal inference is robust for internal validity, i.e., the PIV, as a robustness index of causal inference. Formally, the PIV is the probability of rejecting the null hypothesis again based on both the observed sample and the counterfactuals, provided the same null hypothesis has already been rejected based on the observed sample. Under either frequentist or Bayesian framework, one can bound the PIV of an inference based on his bounded belief about the counterfactuals, which is often needed when the unconfoundedness assumption is dubious. The PIV is equivalent to statistical power when the NHST is thought to be based on both the observed sample and the counterfactuals. We summarize the process of evaluating internal validity with the PIV into an eight-step procedure and illustrate it with an empirical example (i.e., Hong and Raudenbush (2005)).
The article presents the issues taxation of employee revenue from the fringe benefits by Personal Income Tax. On the one hand, the employee fringe benefits are identified with a defined benefit and on the other with a non-monetary incentive system. However, the benefits give rise to many doubts and controversies. The reasons for this are the lack of legal definition and the lack of their legal directory. Therefore, the views of the judiciary and decisions of tax authorities indicate for example what can be classified fringe benefits.
To date, axisymmetric internal wave fields, which have relevance to atmospheric internal wave fields generated by storm cells and oceanic near-inertial wave fields generated by surface storms, have been experimentally realized using an oscillating sphere or torus as the source. Here, we use a novel wave generator configuration capable of exciting axisymmetric internal wave fields of arbitrary radial form to generate axisymmetric internal wave modes. After establishing the theoretical background for axisymmetric mode propagation, taking into account lateral and vertical confinement, and also accounting for the effects of weak viscosity, we experimentally generate and study modes of different order. We characterize the efficiency of the wave generator through careful measurement of the wave amplitude based upon group velocity arguments. This established, we investigate the ability of vertical confinement to induce resonance, identifying a series of experimental resonant peaks that agree well with theoretical predictions. In the vicinity of resonance, the wave fields undergo a transition to non-linear behaviour that is initiated on the central axis of the domain and proceeds to erode the wave field throughout the domain.
We consider a discrete dynamical system with internal degrees of freedom (DOF). Due to the symmetry between the internal DOFs, certain internal modes cannot be excited by external forcing (in a case of linear interactions) and thus are considered "hidden". If such a system is weakly asymmetric, the internal modes remain approximately "hidden" from the external excitation, given that small damping is taken into account. However, already in the case of weak cubic nonlinearity, these hidden modes can be excited, even as the exact symmetry is preserved. This excitation occurs through parametric resonance. Floquet analysis reveals instability patterns for the explored modes. To perform this analysis with the required accuracy, we suggest a special method for obtaining the Fourier series of the unperturbed solution for the nonlinear normal mode. This method does not require explicit integration of the arising quadratures. Instead, it employs expansion of the solution at the stage of the implicit quadrature in terms of Chebyshev polynomials. The emerging implicit equations are solved by using a fixed-point iteration scheme. Poincaré sections help to clarify the correspondence between the loss of stability of the modes and the global structure of the dynamical flow. In particular, the conditions for intensive energy exchange in the system are characterized.
Rad Niazadeh, Jason Hartline, Nicole Immorlica
et al.
Standard ad auction formats do not immediately extend to settings where multiple size configurations and layouts are available to advertisers. In these settings, the sale of web advertising space increasingly resembles a combinatorial auction with complementarities, where truthful auctions such as the Vickrey-Clarke-Groves (VCG) can yield unacceptably low revenue. We therefore study core selecting auctions, which boost revenue by setting payments so that no group of agents, including the auctioneer, can jointly improve their utilities by switching to a different outcome. Our main result is a combinatorial algorithm that finds an approximate bidder optimal core point with almost linear number of calls to the welfare maximization oracle. Our algorithm is faster than previously-proposed heuristics in the literature and has theoretical guarantees. We conclude that core pricing is implementable even for very time sensitive practical use cases such as realtime auctions for online advertising and can yield more revenue. We justify this claim experimentally using the Microsoft Bing Ad Auction data, through which we show our core pricing algorithm generates almost 26% more revenue than VCG on average, about 9% more revenue than other core pricing rules known in the literature, and almost matches the revenue of the standard Generalized Second Price (GSP) auction.