arXiv subjects

Connor T. Jerzak

Publications and source records attributed to Connor T. Jerzak.

At least 19 recordsLinked to original sources

arXiv2026

Correcting Bias When Using Latent Regressors

Many core concepts in political science are latent and therefore can only be measured with error. Measurement error in a predictor attenuates slope coefficient estimates in regression, often biasing them toward zero. We show that widely used strategies for correcting attenuation bias -- including instrumental variables and the method of composition -- are themselves biased when applied to latent regressors, often performing worse than simple regression ignoring the measurement error altogether. We derive a correlation-based correction using split-sample measurement strategies. Rather than assuming a particular estimation strategy for the latent trait, our approach is modular and can be easily deployed with a wide variety of latent variable measurement strategies, including additive score, factor, or machine learning models; it requires no joint estimation while yielding consistent slopes under standard assumptions. Simulations and applications show stronger relationships after our correction, sometimes by as much as 50%. Open-source software implements the procedure. Results underscore that latent predictors demand tailored error correction; otherwise, conventional practice can exacerbate bias

stat.AP↗

arXiv2026

Platonic Representations for Poverty Mapping: Unified Vision-Language Codes or Agent-Induced Novelty?

We investigate whether socioeconomic indicators, like household wealth, leave recoverable informational imprints in both satellite imagery (capturing features like buildings and roads) and Internet-sourced text (reflecting historical, cultural, and narratives of neighborhoods). Using DHS data from African neighborhoods (clusters), we pair high-resolution Landsat images with textual descriptions generated by LLMs conditioned on location/year, plus text retrieved by an LLM-driven AI Search Agent from web sources. We develop a multimodal framework that predicts household wealth (International Wealth Index; IWI) via five pipelines: (i) a vision model on satellite images, (ii) an LLM using only location and year, (iii) an AI agent that searches and synthesizes web text, (iv) a joint image-text encoder, and (v) an ensemble of all signals. Our framework yields three contributions. First, evaluations show that fusing vision and agent/LLM-generated text improves on vision-only baselines in wealth prediction (e.g., R-squared of 0.77 vs. 0.63 on out-of-sample splits), with LLM-internal knowledge (artificial neural memory) proving surprisingly predictive in out-of-country/time generalization. Second, we find suggestive evidence of partial representational alignment: fused embeddings from vision and language modalities correlate moderately (median cosine similarity across modalities of about 0.60 after alignment). This pattern is broadly consistent with the Platonic Representation Hypothesis, but does not by itself establish convergence to a single shared latent representation. Because agent-retrieved data yields only marginal and unstable gains across splits, our evidence for the Agent-Induced Novelty Hypothesis is limited. Third, we release a large-scale multimodal dataset of about 60,000 DHS clusters, each linked to satellite images, LLM-generated descriptions, and AI-agent-retrieved texts.

cs.AI↗

arXiv2026

Temporal Dynamics of Development Aid in Africa: Evidence from a Staggered Difference-in-Differences Study of China and World Bank Projects

Subnational studies of aid effectiveness often rely on repeated cross-sections or nighttime lights, making it difficult to separate local treatment effects from baseline differences and potentially favoring infrastructure-heavy projects. We address these limitations by studying World Bank and Chinese development projects in Africa with a balanced panel of 2,166 DHS clusters across 35 countries from 2002 to 2013. Geocoded AidData projects are linked to satellite-imputed International Wealth Index estimates, a household-centered measure of material living standards. We compare a conventional two-way fixed effects (TWFE) event-study with the switcher--stayer estimator of de Chaisemartin and D'Haultfoeuille (dCdH), which avoids contaminated comparisons under staggered treatment timing. Pre-treatment diagnostics show that project placement is frequently selective: clusters that later receive projects often begin from weaker relative positions before treatment onset. Consequently, TWFE often implies larger post-treatment gains than the preferred staggered-treatment design supports. Under dCdH, the evidence becomes more selective and sector-specific. For the World Bank, positive evidence is strongest in Health, while Education shows positive but less cleanly identified gains. For China, Water Supply and Sanitation and Other Social Infrastructure and Services show positive associations with local wealth, although residual selection concerns remain. By contrast, Chinese Energy Generation and Supply appears strongly positive under TWFE but falls close to zero under dCdH. Overall, the results do not support a donor-wide claim that either the World Bank or China uniformly improves local wealth. Instead, estimated effects are concentrated in a limited set of donor--sector panels and depend strongly on how treatment timing, selection, and outcome measurement are handled.

econ.GN↗

arXiv2026

MiniMax Learning of Interpretable Factored Stochastic Policies from Conjoint Data, with Uncertainty Quantification

We study offline policy optimization over exponentially large factorial action spaces from randomized preference data, showing how conjoint experiments can estimate interpretable stochastic policies with asymptotically valid uncertainty under regularity conditions. Conjoint analyses typically report Average Marginal Component Effects (AMCEs) by averaging over opponent attributes and thus ignore strategic interdependence. We instead learn stochastic interventions -- product-of-Categorical policies over factor levels -- that (i) optimize expected outcomes in an average-case setting and (ii) extend to a two-player minimax (adversarial) setting that realistically captures simultaneous strategic candidate selection. Methodologically, we derive a closed-form optimizer for a tractable two-way interaction regime with L2 variance regularization, and provide a general gradient-based procedure for richer model classes. Uncertainty from the outcome model propagates asymptotically to both the optimal policy and its value via a Delta method approximation. We further model institutional details (e.g., primaries) inside the minimax objective and introduce a data-driven measure of strategic divergence between parties. On synthetic data, we empirically characterize finite-sample error and coverage as dimensionality and $n$ vary. On a U.S. presidential conjoint, adversarially learned policies produce restricted-equilibrium vote shares that align with historical election ranges in our data, in stark contrast to non-adversarial (averaging) optimizers.

stat.ME↗

arXiv2026

Queryable LoRA: Instruction-Regularized Routing Over Shared Low-Rank Update Atoms

We present a data-adaptive method for parameter-efficient fine-tuning of large neural networks. Standard low-rank adaptation methods improve efficiency by restricting each layer update to a fixed low-rank form, but this static parameterization can be too rigid when the appropriate correction depends on the input and on the evolving depth-wise computation of the network. Our approach replaces a purely layer-local adapter with a shared queryable memory of low-rank update atoms. For each block of layers, the model forms a query from the current low-rank state and a running summary of previous blocks, uses this query to retrieve a content-dependent combination of shared update components via attention, and applies the resulting routed operator within the low-rank bottleneck. In this way, the method retains the efficiency and scalability of low-rank adaptation while allowing the effective update to vary across inputs and to share reusable structure across layers. The resulting architecture provides a principled middle ground between static LoRA-style updates and fully generated parameter updates: it remains compact and parameter-efficient while supporting dynamic, context-sensitive adaptation. Further, we incorporate instruction-regularization by augmenting routing logits with a language-induced prior over update atoms, thereby biasing the selection of low-rank transformations toward semantically relevant directions without generating unconstrained parameter updates. Experiments on noisy non-linear regression tasks and LLM fine-tuning suggest that this queryable update-memory formulation can improve final test performance and training stability compared to standard low-rank adaptation, while using a comparable number of trainable parameters.

cs.LG↗

arXiv2026

Multiplication in Multimodal LLMs: Computation with Text, Image, and Audio Inputs

Multimodal LLMs can accurately perceive numerical content across modalities yet fail to perform exact multi-digit multiplication when the identical underlying arithmetic problem is presented as numerals, number words, images, or in audio form. Because existing benchmarks often lack systematically paired instances across modalities, it remains difficult to compare genuine arithmetic limits within and across model families. We therefore introduce a controlled multimodal multiplication benchmark that factorially varies digit length, digit sparsity, representation (e.g., numerals vs. number words), and modality (text, rendered images, audio), with paired instances from a reproducible generator. We also define arithmetic load, C, as the product of the total and non-zero digit count as a compact, mechanistically motivated proxy for operation count. Across evaluations, accuracy falls sharply as C grows, often nearing zero by C > 100. Indeed, C remains predictive of performance across modalities and models, with R-squared often > 0.5, nearing the value from more complex measures of arithmetic load that count the number of intermediate arithmetic steps. A separate perception-versus-computation decomposition shows that multimodal degradation is primarily computational rather than perceptual: on matched-perception checks, models are near-perfect (> 99%) across modalities, even when multiplication accuracy drops. Beyond measuring when models fail, we ask which procedures they are predisposed to follow. We introduce a forced-completion loss probe that scores heuristic-specific reasoning prefixes--including columnar multiplication, distributive decomposition, and rounding/compensation. Here, decomposition is favored in both text and vision modalities; heuristic-specific LoRA adapters produce near-orthogonal updates yet degrade accuracy, indicating the base model maintains a well-tuned internal router.

cs.CL↗

arXiv2026

Chinese vs. World Bank Development Projects: Insights from Earth Observation and Computer Vision on Wealth Gains in Africa, 2002-2013

Debates about whether development projects improve living conditions persist, partly because observational estimates can be biased by incomplete adjustment and because reliable outcome data are scarce at the neighborhood level. We address both issues in a continent-scale, sector-specific evaluation of Chinese and World Bank projects across 9,899 neighborhoods in 36 African countries (2002-2013), representative of ~88% of the population. First, we use a recent dataset that measures living conditions with a machine-learned wealth index derived from contemporaneous satellite imagery, yielding a consistent panel of 6.7 km square mosaics. Second, to strengthen identification, we proxy officials' map-based placement criteria using pre-treatment daytime satellite images and fuse these with tabular covariates to estimate funder- and sector-specific ATEs via inverse-probability weighting. Incorporating imagery often shrinks effects relative to tabular-only models. On average, both donors raise wealth, with larger and more consistent gains for China; sector extremes in our sample include Trade and Tourism (330) for the World Bank (+12.29 IWI points), and Emergency Response (700) for China (+15.15). Assignment-mechanism analyses also show World Bank placement is often more predictable from imagery alone (as well as from tabular covariates). This suggests that Chinese project placements are more driven by non-visible, political, or event-driven factors than World Bank placements. To probe residual concerns about selection on observables, we also estimate within-neighborhood (unit) fixed-effects models at a spatial resolution about 67 times finer than prior fixed-effects analyses, leveraging the computer-vision-imputed IWI panels; these deliver smaller but, for Chinese projects, directionally consistent effects.

stat.AP↗

arXiv2026

FastRerandomize: Fast Rerandomization Using Accelerated Computing

We present fastrerandomize, an R package for fast, scalable rerandomization in experimental design. Rerandomization improves precision by discarding treatment assignments that fail a prespecified covariate-balance criterion, but existing implementations can become computationally prohibitive as the number of units or covariates grows. fastrerandomize introduces three complementary advances: (i) optional GPU/TPU acceleration to parallelize balance checks, (ii) memory-efficient key-only storage that avoids retaining full assignment matrices, and (iii) auto-vectorized, just-in-time compiled kernels for batched candidate generation and inference. This approach enables exact or Monte Carlo rerandomization at previously intractable scales, making it practical to adopt the tighter balance thresholds required in modern high-dimensional experiments while simultaneously quantifying the resulting gains in precision and power for a given covariate set. Our approach also supports randomization-based testing conditioned on acceptance. In controlled benchmarks, we observe order-of-magnitude speedups over baseline workflows, with larger gains as the sample size or dimensionality grows, translating into improved precision of causal estimates.

stat.CO↗

arXiv2025

Detecting and Mitigating Treatment Leakage in Text-Based Causal Inference: Distillation and Sensitivity Analysis

Text-based causal inference increasingly employs textual data as proxies for unobserved confounders, yet this approach introduces a previously undertheorized source of bias: treatment leakage. Treatment leakage occurs when text intended to capture confounding information also contains signals predictive of treatment status, thereby inducing post-treatment bias in causal estimates. Critically, this problem can arise even when documents precede treatment assignment, as authors may employ future-referencing language that anticipates subsequent interventions. Despite growing recognition of this issue, no systematic methods exist for identifying and mitigating treatment leakage in text-as-confounder applications. This paper addresses this gap through three contributions. First, we provide formal statistical and set-theoretic definitions of treatment leakage that clarify when and why bias occurs. Second, we propose four text distillation methods -- similarity-based passage removal, distant supervision classification, salient feature removal, and iterative nullspace projection -- designed to eliminate treatment-predictive content while preserving confounder information. Third, we validate these methods through simulations using synthetic text and an empirical application examining International Monetary Fund structural adjustment programs and child mortality. Our findings indicate that moderate distillation optimally balances bias reduction against confounder retention, whereas overly stringent approaches degrade estimate precision.

econ.EM↗

arXiv2025

Debiasing Machine Learning Predictions for Causal Inference Without Additional Ground Truth Data: "One Map, Many Trials" in Satellite-Driven Poverty Analysis

Machine learning models trained on Earth observation data, such as satellite imagery, have demonstrated significant promise in predicting household-level wealth indices, enabling the creation of high-resolution wealth maps that can be leveraged across multiple causal trials while addressing chronic data scarcity in global development research. However, because standard training objectives prioritize overall predictive accuracy, these predictions often suffer from shrinkage toward the mean, leading to attenuated estimates of causal treatment effects and limiting their utility in policy evaluations. Existing debiasing methods, such as Prediction-Powered Inference (PPI), can handle this attenuation bias but require additional fresh ground-truth data at the downstream stage of causal inference, which restricts their applicability in data-scarce environments. We introduce and evaluate two post-hoc correction methods -- Linear Calibration Correction (LCC) and a Tweedie's correction approach -- that substantially reduce shrinkage-induced prediction bias without relying on newly collected labeled data. LCC applies a simple linear transformation estimated on a held-out calibration split; Tweedie's method locally de-shrink predictions using density score estimates and a noise scale learned upstream. We provide practical diagnostics for when a correction is warranted and discuss practical limitations. Across analytical results, simulations, and experiments with Demographic and Health Surveys (DHS) data, both approaches reduce attenuation; Tweedie's correction yields nearly unbiased treatment-effect estimates, enabling a "one map, many trials" paradigm. Although we demonstrate on EO-ML wealth mapping, the methods are not geospatial-specific: they apply to any setting where imputed outcomes are reused downstream (e.g., pollution indices, population density, or LLM-derived indicators).

stat.ML↗

arXiv2025

Remote Auditing: Design-based Tests of Randomization, Selection, and Missingness with Broadly Accessible Satellite Imagery

Randomized controlled trials (RCTs) are the benchmark for causal inference, yet field implementation can drift from the registered design or, by chance, yield imbalances. We introduce a remote audit -- a preregistrable, design-based diagnostic that uses strictly pre-treatment, publicly available satellite imagery to test whether assignment is independent of local conditions. The audit implements a conditional randomization test that asks whether treatment is more predictable from pre-treatment features than under the registered mechanism, delivering a finite-sample-valid, nonparametric check that honors blocks and clusters and controls multiplicity across image models, resolutions, and patch sizes via a max-statistic. The same preregistered procedure can be run before baseline data collection to guide implementation and, after assignments are realized, to audit the actual allocation. In two illustrations -- Uganda's Youth Opportunities Program (randomization corroborated) and a school-based experiment in Bangladesh (assignment predictable relative to the design, consistent with independent concerns) -- the audit can surface potential problems early, before costly scientific investments. We also provide descriptive diagnostics for selection into the study and for missingness. Because it is low-cost and can be implemented rapidly in a unified way across diverse global administrative jurisdictions, the remote audit complements balance tests, strengthens preregistration, and enables rapid design checks when conventional data collection is slow, expensive, or infeasible.

stat.ME↗

arXiv2025

Benchmarking Debiasing Methods for LLM-based Parameter Estimates

Large language models (LLMs) offer an inexpensive yet powerful way to annotate text, but are often inconsistent when compared with experts. These errors can bias downstream estimates of population parameters such as regression coefficients and causal effects. To mitigate this bias, researchers have developed debiasing methods such as Design-based Supervised Learning (DSL) and Prediction-Powered Inference (PPI), which promise valid estimation by combining LLM annotations with a limited number of expensive expert annotations. Although these methods produce consistent estimates under theoretical assumptions, it is unknown how they compare in finite samples of sizes encountered in applied research. We make two contributions. First, we study how each methods performance scales with the number of expert annotations, highlighting regimes where LLM bias or limited expert labels significantly affect results. Second, we compare DSL and PPI across a range of tasks, finding that although both achieve low bias with large datasets, DSL often outperforms PPI on bias reduction and empirical efficiency, but its performance is less consistent across datasets. Our findings indicate that there is a bias-variance tradeoff at the level of debiasing methods, calling for more research on developing metrics for quantifying their efficiency in finite samples.

cs.CL↗

arXiv2025

A Scoping Review of Earth Observation and Machine Learning for Causal Inference: Implications for the Geography of Poverty

Earth observation (EO) data such as satellite imagery can have far-reaching impacts on our understanding of the geography of poverty, especially when coupled with machine learning (ML) and computer vision. Early research used computer vision to predict living conditions in areas with limited data, but recent studies increasingly focus on causal analysis. Despite this shift, the use of EO-ML methods for causal inference lacks thorough documentation, and best practices are still developing. Through a comprehensive scoping review, we catalog the current literature on EO-ML methods in causal analysis. We synthesize five principal approaches to incorporating EO data in causal workflows: (1) outcome imputation for downstream causal analysis, (2) EO image deconfounding, (3) EO-based treatment effect heterogeneity, (4) EO-based transportability analysis, and (5) image-informed causal discovery. Building on these findings, we provide a detailed protocol guiding researchers in integrating EO data into causal analysis -- covering data requirements, computer vision model selection, and evaluation metrics. While our focus centers on health and living conditions outcomes, our protocol is adaptable to other sustainable development domains utilizing EO data.

cs.LG↗

arXiv2025

Optimizing Multi-Scale Representations to Detect Effect Heterogeneity Using Earth Observation and Computer Vision: Applications to Two Anti-Poverty RCTs

Earth Observation (EO) data are increasingly used in policy analysis by enabling granular estimation of conditional average treatment effects (CATE). However, a challenge in EO-based causal inference is determining the scale of the input satellite imagery -- balancing the trade-off between capturing fine-grained individual heterogeneity in smaller images and broader contextual information in larger ones. This paper introduces Multi-Scale Representation Concatenation, a set of composable procedures that transform arbitrary single-scale EO-based CATE estimation algorithms into multi-scale ones. We benchmark the performance of Multi-Scale Representation Concatenation on a CATE estimation pipeline that combines Vision Transformer (ViT) models (which encode images) with Causal Forests (CFs) to obtain CATE estimates from those encodings. We first perform simulation studies where the causal mechanism is known, showing that our multi-scale approach captures information relevant to effect heterogeneity that single-scale ViT models fail to capture as measured by $R^2$. We then apply the multi-scale method to two randomized controlled trials (RCTs) conducted in Peru and Uganda using Landsat satellite imagery. As we do not have access to ground truth CATEs in the RCT analysis, the Rank Average Treatment Effect Ratio (RATE Ratio) measure is employed to assess performance. Results indicate that Multi-Scale Representation Concatenation improves the performance of deep learning models in EO-based CATE estimation without the complexity of designing new multi-scale architectures for a specific use case. The application of Multi-Scale Representation Concatenation could have meaningful policy benefits -- e.g., potentially increasing the impact of poverty alleviation programs without additional resource expenditure.

stat.ML↗

arXiv2024

Linking Datasets on Organizations Using Half A Billion Open-Collaborated Records

Scholars studying organizations often work with multiple datasets lacking shared identifiers or covariates. In such situations, researchers usually use approximate string ("fuzzy") matching methods to combine datasets. String matching, although useful, faces fundamental challenges. Even where two strings appear similar to humans, fuzzy matching often struggles because it fails to adapt to the informativeness of the character combinations. In response, a number of machine learning methods have been developed to refine string matching. Yet, the effectiveness of these methods is limited by the size and diversity of training data. This paper introduces data from a prominent employment networking site (LinkedIn) as a massive training corpus to address these limitations. By leveraging information from the LinkedIn corpus regarding organizational name-to-name links, we incorporate trillions of name pair examples into various methods to enhance existing matching benchmarks and performance by explicitly maximizing match probabilities. We also show how relationships between organization names can be modeled using a network representation of the LinkedIn data. In illustrative merging tasks involving lobbying firms, we document improvements when using the LinkedIn corpus in matching calibration and make all data and methods open source.

cs.SI↗

arXiv2024

Effect Heterogeneity with Earth Observation in Randomized Controlled Trials: Exploring the Role of Data, Model, and Evaluation Metric Choice

Many social and environmental phenomena are associated with macroscopic changes in the built environment, captured by satellite imagery on a global scale and with daily temporal resolution. While widely used for prediction, these images and especially image sequences remain underutilized for causal inference, especially in the context of randomized controlled trials (RCTs), where causal identification is established by design. In this paper, we develop and compare a set of general tools for analyzing Conditional Average Treatment Effects (CATEs) from temporal satellite data that can be applied to any RCT where geographical identifiers are available. Through a simulation study, we analyze different modeling strategies for estimating CATE in sequences of satellite images. We find that image sequence representation models with more parameters generally yield a greater ability to detect heterogeneity. To explore the role of model and data choice in practice, we apply the approaches to two influential RCTs -- Banerjee et al. (2015), a poverty study in Cusco, Peru, and Bolsen et al. (2014), a water conservation experiment in Georgia, USA. We benchmark our image sequence models against image-only, tabular-only, and combined image-tabular data sources, summarizing practical implications for investigators in a multivariate analysis. Land cover classifications over satellite images facilitate interpretation of what image features drive heterogeneity. We also show robustness to data and model choice of satellite-based generalization of the RCT results to larger geographical areas outside the original. Overall, this paper shows how satellite sequence data can be incorporated into the analysis of RCTs, and provides evidence about the implications of data, model, and evaluation metric choice for causal analysis.

stat.ME↗

arXiv2023

CausalImages: An R Package for Causal Inference with Earth Observation, Bio-medical, and Social Science Images

The causalimages R package enables causal inference with image and image sequence data, providing new tools for integrating novel data sources like satellite and bio-medical imagery into the study of cause and effect. One set of functions enables image-based causal inference analyses. For example, one key function decomposes treatment effect heterogeneity by images using an interpretable Bayesian framework. This allows for determining which types of images or image sequences are most responsive to interventions. A second modeling function allows researchers to control for confounding using images. The package also allows investigators to produce embeddings that serve as vector summaries of the image or video content. Finally, infrastructural functions are also provided, such as tools for writing large-scale image and image sequence data as sequentialized byte strings for more rapid image analysis. causalimages therefore opens new capabilities for causal inference in R, letting researchers use informative imagery in substantive analyses in a fast and accessible manner.

cs.LG↗

arXiv2023

Degrees of Randomness in Rerandomization Procedures

Randomized controlled trials are susceptible to imbalance on covariates predictive of the outcome. Rerandomization and deterministic treatment assignment are two proposed solutions. This paper explores the relationship between rerandomization and deterministic assignment, showing how deterministic assignment is an extreme case of rerandomization. The paper argues that in small experiments, both fully randomized and fully deterministic assignment have limitations. Instead, the researcher should consider setting the rerandomization acceptance probability based on an analysis of covariates and assumptions about the data structure to achieve an optimal alignment between randomness and balance. This allows for the calculation of minimum p-values along with valid permutation tests and fiducial intervals. The paper also introduces tools, including a new, open-source R package named fastrerandomize, to implement rerandomization and explore options for optimal rerandomization acceptance thresholds.

stat.ME↗