The paper introduces D‑VAE, a Deconfounder Variational Autoencoder that estimates causal effects of expert‑generated marketing labels on LLM‑generated advertisements. By integrating a multi‑treatment deconfounder with causal representation learning, D‑VAE disentangles expert‑labeling logic and provides PAC‑based bounds on omitted‑variable bias. Applied to over 29,000 foundation‑makeup ad creatives, the method outperforms existing benchmarks and a field experiment shows that its recommended label configurations boost engagement by 32.7% while matching human expert performance.
arXiv:2604. 14575v3 Announce Type: replace-cross Abstract: Marketing research often relies on parameters estimated from costly human-generated data, such as conjoint survey responses, purchase decisions, and field experiment outcomes.
arXiv:2605. 29928v2 Announce Type: replace-cross Abstract: As AI-generated and AI-assisted content floods online spaces, source labels attached to such content can distort human reasoning judgments, with downstream consequences for moderation, evaluation, and decision-making.
arXiv:2605.27043v2 Announce Type: replace Abstract: Predictive models trained on observational data often fail to generalise when deployed, especially when the training data is a product of the syste...
arXiv:2602. 12972v2 Announce Type: replace-cross Abstract: In online advertising, marketing interventions such as coupons introduce significant confounding bias into Click-Through Rate (CTR) prediction.
arXiv:2608. 09019v1 Announce Type: cross Abstract: As generative AI increasingly becomes a common source of daily decision-making, including financial choices, it is critical to understand how people evaluate AI-generated financial advice.
arXiv:2404.06349v3 Announce Type: replace Abstract: The ability to understand causality significantly impacts the competence of large language models (LLMs) in output explanation and counterfactual r...
The paper investigates how learned multimodal representations of high‑dimensional health data can fail to preserve causal information, even when trained to reconstruct the data. It establishes conditions under which a representation is sufficient for causal identification, shows that interventional mediation requires stricter preservation than counterfactual comparison, and proves that reconstruction‑based objectives cannot certify sufficiency. The authors decompose the error of cross‑fitted doubly robust estimators into a first‑order representation bias and a second‑order nuisance‑estimation remainder, demonstrating that coverage can deteriorate with larger samples if the representation is insufficient. They propose a practical workflow that combines modality‑specific variational autoencoders, cross‑fitted doubly robust estimation, locked phenotypes, selective pathway analysis, portable representations, and diagnostics for representation insufficiency, exposure leakage, and sensitivity, and validate it in simulations where a rare confounder’s omission leads to severely under‑covered indirect‑effect intervals.
ACTIVA is a transformer‑based conditional variational autoencoder designed to estimate full interventional distributions from observational data without requiring interventional data, a causal graph, or per‑task fitting. It learns a conditional latent prior that amortizes causal knowledge across diverse training tasks, enabling a single forward pass to handle new causal models drawn from the training distribution. Empirical results on synthetic and biologically realistic gene‑expression simulations show that ACTIVA outperforms correlational baselines, reduces spurious non‑descendant effects, and matches strong amortized baselines, while also validating its theoretical mixture‑of‑distributions property on observationally equivalent models.
The paper investigates how the definition of influence—specifically the behavior being attributed, the intervention on training data, and the counterfactual training process—affects rankings produced by influence estimators. It formalizes influence as a counterfactual estimand, distinguishes specification mismatch from approximation error, and categorizes existing estimators by their implied specifications. Experiments demonstrate that different specifications can lead to markedly different rankings, and that careful specification choice improves attribution quality in tasks such as noisy label detection and large‑language‑model attribution.
AI-generated covariates from notes, conversations, images, and wearable streams can change the causal question when their roles are left unspecified. A generated feature may represent a treatment vers...
arXiv:2510. 12957v4 Announce Type: replace-cross Abstract: We treat the internals of generative models as mechanistic objects rather than black boxes.
The paper introduces Integrated Policy Gradient (IPG), a training‑free framework that applies policy‑based attribution to large language model reasoning. IPG backpropagates outcome‑level, non‑differentiable signals through entire inference trajectories, using path integration to isolate outcome‑relevant mediators at the component level. Empirical results show that IPG precisely localizes sparse internal components linked to specific reasoning behaviors and enables modulation of reasoning capabilities.