📑

학술

arXiv 등 학술 논문. CC-BY 라이선스로 자유 재사용 가능 — 출처표시 시 상업 사용 OK.

총 29,279건

Self-Employment as a Signal: Career Concerns with Hidden Firm Performance

arXiv:2509.01265v4 Announce Type: replace Abstract: We study a stationary labour market in which risk-averse workers privately know their permanent talent and choose between risky self-employment, which produces portable public outcomes, and firm employment, which pays a competitive wage but keeps individual performance hidden. Because workers decide whether to generate another public outcome or apply to a firm, both public records and applicant pools are endogenous. We construct market beliefs from the stationary flow of types through all histories leading to each record and then condition them on the current application decision. If the effective continuation factor is below one half, a stationary sequential competitive equilibrium exists and occupational choice follows a talent cutoff at every record. Firm employment is persistent whenever it is strictly optimal for a given worker type at a given record. At any on-path record where both occupations are chosen, higher-talent workers select into self-employment, while the applicant wage lies below mean talent among workers holding that record. This wage discount decomposes exactly into the self-employed share and the talent gap between self-employed workers and applicants. The model yields within-record predictions for occupational choice, wages, subsequent performance, and the duration of opaque employment spells.

arXiv Econ

Two Motives for Verification in Information Cascades

arXiv:2508.20538v5 Announce Type: replace Abstract: We study sequential social learning when agents can pay to conduct a publicly observed investigation before acting. The baseline test is one-sided: success conclusively establishes one state, whereas failure is ordinarily inconclusive; success also gives the investigator a discovery reward. Investigation therefore has two private returns. Its diagnostic value is highest near the action threshold and vanishes at sufficiently optimistic beliefs, while the expected discovery reward is weakly increasing. The equilibrium investigation set has at most two components and can be disconnected, with a central diagnostic region and a separate high-belief, reward-driven region. Because investigation is selected on private information, the attempt itself is informative. The exact public transition map shows that a failed investigation can raise public belief when favorable selection outweighs the adverse outcome. Nevertheless, a positive chance of proof at a given history does not guarantee eventual discovery: with strictly positive cost, the total number of attempts is finite almost surely under either state, and a failure can move beliefs into an absorbing cascade trap. The two-component geometry persists for sufficiently small false-positive rates and can also arise with heterogeneous private costs. The stopping result, however, relies on conclusive evidence and a positive lower bound on costs.

arXiv Econ

Platform Choice, Trust, and Privacy in the Consumer AI Assistant Market

arXiv:2607.15134v1 Announce Type: cross Abstract: We study how a representative sample of United States adult AI-assistant users (n=1,999; June 2026) choose among platforms, allocate tasks across them, evaluate provider trustworthiness, and value data-handling features. Estimates are weighted to the AI-user population using external adoption benchmarks. Four patterns emerge. The market is concentrated but internally differentiated: ChatGPT is the primary assistant for 58% of users and Gemini for 25%, yet smaller platforms hold defensible task niches--Claude captures a third of coding tasks despite a 7% overall share. Task allocation is thus organized by platform far more than by user, and technical use falls steeply with age. Trust is earned through use rather than reputation: Claude is ranked most trustworthy in every head-to-head among users of both platforms, and shows by far the largest gap between how its users and non-users rate it. Finally, privacy concern is near-universal but action is gated by knowledge, not concern; in a choice experiment users pay most to keep humans--not models--out of their conversations ($11.20/month), with valuations rising in task sensitivity.

arXiv Econ

Adaptive Ad Load Design for Sponsored Search Markets: Evidence, Theory, and Deployment

arXiv:2607.14418v1 Announce Type: cross Abstract: Ad-load design is a central supply-side decision in sponsored search: more sponsored slots can raise revenue, but may crowd out organic results and degrade user outcomes. We study this trade-off using a large-scale randomized field experiment on an Android app store, where over five million users are exposed to one through six sponsored slots. Increasing ad load raises revenue by up to 43%, but reduces total search conversions by up to 5% and daily engagement by up to 2.2%. These average effects mask substantial heterogeneity: additional slots generate large revenue gains for high-ad-conversion queries, but little or negative marginal revenue for low-conversion queries. The trade-off also shifts within query as advertiser composition changes, such as brand-advertiser presence. Motivated by these findings, we design and deploy a novel adaptive algorithm -- exploration-augmented Locally Adaptive Ad Load (e-LAAL). e-LAAL combines LAAL, a model-free query-level decision rule that updates ad-load recommendations using recent outcomes, with static exploration arms that maintain support and provide fixed-policy counterfactual benchmarks. We provide a finite-time dynamic-regret guarantee for the e-LAAL architecture. In a platform-level production deployment serving 22.3 million users and 77.6 million searches, e-LAAL improves the empirical revenue--conversion trade-off relative to deployed static benchmarks and outperforms uniform and historical query-dependent static benchmarks.

arXiv Econ

Supervised Fine-Tuning vs. In-Context Learning: An Equilibrium Analysis of LLM Personalization under Congestion

arXiv:2607.14371v1 Announce Type: cross Abstract: Large Language Models (LLMs) have revolutionized AI services, but a critical tension emerges: while personalization improves model performance, it consumes scarce computational resources that users must share. When should a user invest in expensive Supervised Fine-Tuning (SFT) versus lightweight In-Context Learning (ICL)? How does congestion from other users' personalization choices reshape these incentives? And what strategies should platforms adopt when offering multiple personalization algorithms? We develop a tractable framework for LLM serving that captures the statistical-economic trade-offs users face. Our analysis yields several surprising insights. First, we show that ICL and SFT dominate in different regimes, determined by an interplay between pretraining coverage and data signal-to-noise ratios, but congestion can flip these rankings. Second, equilibrium resource consumption exhibits pronounced non-monotonicity: improving pretraining precision reduces the congestion, while broader pretraining coverage and harder tasks sometimes increase it. Third, we prove that offering both personalization methods never hurts the platform's maximal profits, despite potentially increasing computational load. Experiments with GPT-2 on linear regression tasks validate our theoretical predictions about algorithm performance. Complementing these results, our review of documentation from 21 major AI platforms shows that the share offering both SFT and ICL increased from 9.5% in 2021 to 71.4% in 2025, consistent with our platform-design implications.

arXiv Econ

When Is Delegated Play Truthful? Within-Range Regret and the Trilemma of Aligned Delegation

arXiv:2607.14357v1 Announce Type: cross Abstract: Advertisers delegate bidding to autobidders; users delegate tasks to language-model agents. A person describes what they want to an automated proxy that acts in a mechanism on their behalf. This is the revelation principle in production, and it forces a question classical theory assumes away: when is it optimal to describe yourself honestly to your own proxy? We show the answer turns on one quantity, the proxy's within-range regret. The most a principal can gain by misreporting equals the regret of the proxy's honest-report action against those the principal could have steered it to take. Honest self-description is optimal exactly when the proxy already plays the best action it can reach, that is, when it is loyal (Theorem 1). The identity unifies auction-specific autobidding results and pins down when the faithful-communication assumption behind language-model elicitation proxies (Huang et al.) holds. The identity constrains guardrails placed on proxies, from bid caps to a model's alignment layer. No guardrail can be at once binding (it displaces the truthful action from the proxy's best reachable outcome), truthful (honest reporting stays optimal), and capability-preserving (that outcome stays reachable through some report); any two preclude the third (Theorem 2). A safety constraint that alters what a model does while leaving its best output reachable makes honest description of intent suboptimal, so a sharper report can gain. This is the incentive behind prompt-engineering and jailbreaking. Because within-range regret is #P-hard to compute exactly, we estimate it from samples and maintain it as a model is updated, at a cost set by how far the model drifts, not how often it changes. Running it on production language models from five providers under an alignment-style cap, we find honest reporting leaves surplus unclaimed on every model, recovered by inflating the report.

arXiv Econ

Indirect Variational Inference: Applications to Earnings Dynamics

arXiv:2607.15168v1 Announce Type: new Abstract: Latent-variable models are central to economics but often entail intractable integration. Variational inference (VI), widely used in machine learning, turns this integration into tractable, differentiable optimization by replacing the likelihood with a variational objective. However, guarantees of recovering the true parameters remain limited when the variational family is insufficiently flexible -- a key obstacle to the adoption of VI in economics. We first evaluate VI in models of earnings dynamics and show that the choice of variational posterior is crucial. We then introduce indirect variational inference (IVI), which treats VI as an auxiliary model and corrects the bias induced by the variational approximation. IVI retains much of VI's tractability because it does not require computing the likelihood. We apply these methods to models allowing for nonlinear persistence, non-Gaussian and serially correlated transitory shocks, and latent heterogeneity. Across simulated and empirical applications, flexible variational families combined with IVI deliver reliable estimates.

arXiv Econ

Aggregation Bias in Proxy Measurement: Nighttime Lights and Local Economic Activity

arXiv:2607.14825v1 Announce Type: new Abstract: This paper studies when high-resolution signals aggregated to administrative units can recover unobserved local economic activity. We develop a reverse-regression framework for signals generated by activity but used to predict it at coarser spatial supports. The main theorem decomposes predictive elasticity into elementary elasticity, reverse-regression attenuation, and a spatial aggregation term driven by unit size and within-unit dispersion, showing aggregation pulls elasticities toward one. Monte Carlo evidence confirms the decomposition and clarifies transferability conditions. Applications to VIIRS nighttime lights and local GDP or income in Brazil, Italy, the United States, Indonesia, and Kenya support local calibration mainly in richer contexts.

arXiv Econ

Does Multi-Agent Debate Improve AI Feedback on Research Papers?

arXiv:2607.14713v1 Announce Type: new Abstract: Probably not, at least for meta-analyses in economics. In a pre-registered, identity-masked, within-paper experiment, the authors of 44 meta-analyses ranked three AI reports on their own paper by usefulness for improving it: a single pass by a frontier model against two multi-agent debate tools we built and expected to win. All reports were held to a common length and template. The authors preferred the single pass, by 0.66 rank points over mad-research (95% CI 0.32 to 1.00) and 0.57 over paper-workshop (0.16 to 0.95), though paper-workshop spent roughly thirty times the tokens. Authors who recalled their journal referee report usually placed it first and never last; in a separate exercise, three AI judges almost always placed the real journal referee report last. Among the three AI reports, Gemini (the judge whose model family wrote none of the reports) would have ranked paper-workshop first in the authors' place, reversing the single-pass preference. The reversal warns against substituting an AI judge for the author. We measure perceived usefulness for finished papers; whether AI should referee papers is a separate question.

arXiv Econ

Which Green Technology to Subsidize? Evidence from Electric Vehicles in South Korea

arXiv:2607.14446v1 Announce Type: new Abstract: We develop a framework to compare the relative effectiveness of subsidizing alternative emission-reducing technologies. We show that an intermediate technology may reduce emissions more effectively than the cleanest technology if it induces sufficiently greater substitution away from the prevailing high-emission technology. We apply the framework to the South Korean passenger vehicle market using a demand model that incorporates mileage heterogeneity, an important determinant of fuel-type choice. First, reallocating existing subsidies from battery electric vehicles (BEVs), the cleanest technology, to hybrid electric vehicles (HEVs), an intermediate technology, would reduce total greenhouse gas emissions by an additional 47%. Second, for a BEV-focused subsidy policy to outperform an HEV-focused policy, the carbon intensity of electricity generation would need to fall by approximately 45%. Our findings suggest that HEV subsidies remain more effective than BEV subsidies until consumers become sufficiently willing to switch to BEVs or electricity generation becomes sufficiently decarbonized.

arXiv Econ

Probability of worthwhile effect of monotone-response treatments

arXiv:2607.14414v1 Announce Type: new Abstract: Experiments may, by design, prevent one from observing on a single subject both the response to a treatment and to its absence. Because of this, marginal distributions for both cases may be observable but not their joint distribution, thus obscuring the distribution of the treatment effect. We examine the case where we impose that the treatment effect is nonnegative, also called monotone treatment response, a common assumption relevant to many practical applications. We solve the problems of best- and worst-case probabilities that the treatment effect exceeds a given value, using an explicit construction for the dependence scheme in each case. Such problems can equivalently be described, in different contexts, as risk aggregation under dependence uncertainty and an order constraint, and as optimal transport with a particular cost function.

arXiv Econ

From Vector Autoregressions to AI-based Time Series Forecasting: A Review

arXiv:2607.14279v1 Announce Type: new Abstract: Forecasting is a central goal of time-series analysis. This review centers on three major developments in recent AI-based time-series forecasting: transformers, large pretrained models for zero-shot forecasting, and diffusion-based generative forecasters. We connect these methods to the econometric tradition built around the vector autoregression (VAR) through a common object: the conditional distribution of the future given the past. The review is organized around three long-standing challenges: \emph{high dimensionality}, \emph{nonstationarity}, and \emph{nonlinearity}. We argue that modern methods make progress by expanding the classical forecasting template: they allow more flexible dynamics, use larger information sets and training corpora, and represent richer predictive distributions. Yet they often lack the inferential and structural tools that make classical models useful for testing, explanation, and policy analysis. We close by outlining open problems where econometric tools remain important.

arXiv Econ

Model Uncertainty under Non-Gaussian Errors: Bayesian Model Averaging and Selection in Stochastic Frontier Models

arXiv:2607.14274v1 Announce Type: new Abstract: The paper investigates Bayesian Model Averaging and Selection (BMA/S) under non-standard stochastic assumptions, focusing on stochastic frontier analysis (SFA). We propose fast, reliable procedures for inference in the normal-exponential stochastic frontier model and examine whether accounting for asymmetric disturbances affects model averaging and/or selection outcomes relative to the conventional Gaussian-error BMA/S. Particular attention is given to moderate-dimensional covariate selection problems typical in SFA applications. We demonstrate that, with appropriate search strategies and parallelization techniques, exhaustive model search can be computationally feasible and, in some cases, more practical than stochastic search alternatives. A Monte Carlo simulation study is used to compare the proposed SF-BMA/S procedure with standard Gaussian-error BMA/S under varying levels of inefficiency-to-noise ratio and signal strength with respect to the data generating process. The results show that accounting for stochastic frontier structures may affect posterior inference and model averaging outcomes, especially in scenarios where efficiency analysis is most sensible.

arXiv Q-Bio

Higher-Order Hit-&-Run Samplers for Linearly Constrained Densities

arXiv:2602.14616v2 Announce Type: replace-cross Abstract: Markov chain Monte Carlo (MCMC) sampling of densities restricted to linearly constrained domains is an important task arising in Bayesian treatment of inverse problems in the natural sciences. While efficient algorithms for uniform polytope sampling exist, much less work has dealt with more complex constrained densities. In particular, gradient information as used in unconstrained MCMC is not necessarily helpful in the constrained case, where the gradient may push the proposal's density out of the polytope. In this work, we propose a novel constrained sampling algorithm, which combines strengths of higher-order information, like the target's log-density's gradients and curvature, with the Hit-&-Run proposal, a simple mechanism which guarantees the generation of feasible proposals, fulfilling the linear constraints. Our extensive experiments demonstrate improved sampling efficiency on complex constrained densities over various constrained and unconstrained samplers.

arXiv Q-Bio

Training a force field for proteins and small molecules from scratch

arXiv:2603.16770v2 Announce Type: replace Abstract: Force fields for molecular dynamics are usually developed manually, limiting their transferability and making systematic exploration of functional forms challenging. We developed a graph neural network that assigns all force field parameters for diverse molecules using continuous atom typing. The freely-available model, called Garnet, was trained on quantum mechanical, condensed phase and protein nuclear magnetic resonance data without the use of existing parameters. The resulting force field shows comparable performance to current force fields on small molecules, folded proteins, protein complexes and disordered proteins. It shows similar results to popular approaches for relative binding free energy predictions across a range of targets. Assessing different functional forms shows that the double exponential potential is a flexible and accurate alternative to the Lennard-Jones potential. Garnet provides a platform for automated, reproducible force field discovery that brings the benefits of machine learning to classical force fields.

arXiv Q-Bio

TRAECR: A Tool for Preprocessing Positron Emission Tomography Imaging for Statistical Modeling

arXiv:2511.04458v2 Announce Type: replace Abstract: Positron emission tomography (PET) imaging is widely used in a number of clinical applications, including cancer and Alzheimer's disease (AD) diagnosis, monitoring of disease development, and treatment effect evaluation. Statistical modeling of PET imaging is essential to address continually emerging scientific questions in these research fields, including hypotheses related to evaluation of effects of disease modifying treatments on amyloid reduction in AD and associations between amyloid reduction and cognitive function, among many others. In this paper, we provide background information and tools for statisticians interested in developing statistical models for PET imaging to pre-process and prepare data for analysis. We introduce our novel pre-processing and visualization tool TRAECR (Template registration, MRI-PET co-Registration, Anatomical brain Extraction and COMBAT/RAVEL harmonization) to facilitate data preparation for statistical analysis.

arXiv Q-Bio

A Machine Learning Benchmarking Framework for Lipid Nanoparticle Transfection Efficiency Prediction

arXiv:2507.03209v2 Announce Type: replace Abstract: The discovery of new ionizable lipids for efficient lipid nanoparticle (LNP)-mediated RNA delivery remains a major bottleneck in RNA therapeutics development. Recent advances demonstrate the potential of machine learning (ML) models to predict transfection efficiency directly from lipid structure, enabling high-throughput virtual screening and accelerating lead identification. However, as new models for LNP transfection prediction continue to emerge, the lack of rigorous and standardized benchmarking poses a significant risk and may undermine confidence in their reliability for discovery. Here, we present a robust ML benchmarking framework for evaluating transfection prediction models based on ionizable lipid structures. This framework systematically benchmarks diverse molecular representations paired with a broad range of ML architectures spanning traditional models, feedforward neural networks, and state-of-the-art graph-based methods. In addition, the presented framework supports assessment of model generalization and evaluates prediction reliability beyond standard regression metrics. Using a curated dataset of 1,100 unique ionizable lipid structures derived from the HeLa transfection dataset originally reported by Xu et al., we show that within this framework, models leveraging explicit molecular substructure encoding consistently achieve the highest predictive accuracy and should serve as essential baselines for the development of new, more sophisticated models. In contrast, some current graph-based models, including AGILE, Chemprop, and KPGT, tend to show comparatively lower accuracy. The presented framework provides a standardized, transparent, and comprehensive benchmarking resource that enables meaningful comparison of emerging architectures and establishes strong baselines for future development of predictive models in lipid-based RNA delivery.

arXiv Q-Bio

Multimodal Semantic-Aware Contrastive Learning For False Negative Mitigation in 3D Medical Imaging

arXiv:2607.14995v1 Announce Type: cross Abstract: Multimodal Contrastive Learning (CL) has shown significant performance in aligning representations across various data modalities and improving downstream tasks, especially in healthcare. It works by minimizing the distance between matched (positive) data modalities, while maximizing the distance between mismatched (negative) samples. Traditional CL frameworks typically assume instance-based correspondence within data batches, treating all non-paired samples as negatives. However, this assumption often fails in medical settings, where samples may share high-level semantic attributes, leading to false negatives that degrade representation quality. In this paper, we propose Multimodal Semantic-Aware Contrastive Learning (MseaCL), a CL framework trained on a pediatric cohort of 3D brain magnetic resonance imaging (MRI) scans and radiology reports. The goal of this framework is to mitigate the impact of semantically similar false negative samples by incorporating semantic similarity between radiology reports, as a guiding signal during the learning process. Our results indicate that applying this framework as a pretraining stage can achieve notable improvements in downstream tasks, e.g., at least a 22.6\% increase in the area under the receiver operating characteristic curve (AUC) of pediatric brain tumor molecular classification, demonstrating its potential for more robust and semantically aligned multimodal representations in clinical applications.

arXiv Q-Bio

LATTICE: Graph Self-Supervised Learning for Multimodal Spatial Omics Integration

arXiv:2607.14410v1 Announce Type: cross Abstract: Spatially resolved omics studies increasingly combine transcriptomic and epigenomic assays, yet downstream analysis is often still performed using single-modality pipelines. We present LATTICE (Latent Alignment of Tissue-level and Transcriptomic Information for Cross-modal Embedding), a graph-based self-supervised framework that learns spot-level representations from harmonized multimodal features. LATTICE integrates five aligned modality blocks per Visium spot: Visium RNA, scMultiome RNA, scMultiome ATAC, spatial ATAC, and spatial CUT\&Tag. These modalities capture spatial transcriptomic measurements, single-cell inferred regulatory activity, and in situ chromatin and histone states within a unified lattice representation. LATTICE constructs a spatial neighborhood graph and trains a TransformerConv encoder using masked reconstruction, cross-modal alignment, and spatial smoothness objectives. On a private 11-sample melanoma cohort from an anonymized clinical collaborator comprising 54{,}912 total spots, LATTICE demonstrated stable optimization behavior, reproducible embeddings across analysis seeds, and complete multimodal integration across all samples. Adding scMultiome RNA to Visium RNA alone substantially improved concordance with Space Ranger clusters across 11 runs (adjusted Rand index [ARI] +0.157, normalized mutual information [NMI] +0.143, and spatial contiguity +0.174). Additional modalities further improved spatial contiguity and multimodal utility score (MUS), although they sometimes reduced agreement with RNA-derived reference labels, likely because the learned embeddings captured chromatin and regulatory structure beyond transcriptomic similarity alone. These results position LATTICE as a practical and empirically grounded framework for multimodal spatial omics integration, while also highlighting the need for stronger supervision and broader external benchmarking.

arXiv Q-Bio

A model with exposure in the epidemiological sense. Part 1 -- Base model

arXiv:2607.15104v1 Announce Type: new Abstract: We explore a model of infectious disease spread that incorporates exposure to the pathogen in the classic epidemiological acceptation of the term, i.e., a contact with an infectious individual has taken place but the infection has not necessarily been acquired. The model also includes a (discrete) age of infection structure, allowing to implicitly describe the viral load of infected individuals and in turn, to describe the probability of developing an infection as a function of the viral load of the infectious contacts.

← 이전3 / 1464다음 →

arXiv Econ