arXiv subjects

Stefanus Jasin

Publications and source records attributed to Stefanus Jasin.

12 recordsLinked to original sources

arXiv2026

Designing Agentic AI Workflow Portfolios under Imperfect Selection and Compute Cost

Agentic AI systems often approach the same task through multiple workflows that differ in reasoning strategy, verification structure, and compute cost. A natural deployment policy is to use the workflow with the highest average performance, but this can be suboptimal because different workflows may succeed on different instances. We study a portfolio-and-selector paradigm in which a firm runs multiple workflow executions and selects the final answer after observing their outputs. Additional executions may uncover correct answers that the best standalone workflow misses, but they consume compute and introduce plausible distractors that complicate final selection. We formulate this as a workflow portfolio problem in which the firm jointly chooses run size and allocation across workflow types. We summarize selector quality through an odds-lift index and derive sharp bounds on the value of workflow variety. For finite workflow pools, we develop exact formulations, linear programming relaxations, randomized rounding procedures, and computable performance certificates. For large implicit workflow classes, we derive a finite-dimensional dual and an ellipsoid method using a pricing oracle to identify workflows with high weighted accuracy net of recurring compute cost. Under a weak condition, the method obtains a near-optimal solution to the relaxation with polynomially many oracle calls. We evaluate the framework on three datasets: ABCD, Schema-Guided Dialogue, and HotpotQA. Relative to the best standalone workflow, portfolio optimization improves held-out selector accuracy by 3.1, 7.5, and 0.9 percentage points, respectively. Dual-guided workflow generation adds 3.5 points on ABCD and 24.1 on HotpotQA, with no additional gain on Schema-Guided Dialogue.

cs.AI↗

arXiv2026

Automated Design of Inventory Policy with Large Language Models: An Exploratory Study

Firms making inventory decisions have access to operational data, optimization tools, and large language models (LLMs). Typically, data characterize the operating environment, optimization selects parameters within a prespecified inventory policy class, and LLMs support coding and decision analysis. We develop an integrated framework that combines these resources to automate inventory policy design. Given demand data, the framework iteratively uses an LLM to generate parameterized policy classes and an external solver to optimize its parameters within each class. Across 30 lost-sales inventory instances, the mean cost reduction relative to optimized base-stock benchmarks increases from 17.5% after one generation to 30.0% after ten generations. Parameter optimization is central to this performance: an LLM-only variant performs substantially worse, whereas optimization-guided feedback improves policy quality, accelerates search, and directs the LLM toward better policy classes rather than merely better parameter values within a fixed class. The strongest discovered policies are also interpretable: they combine recognizable inventory-control motifs, including capped orders, discounted or weighted pipeline inventory, and threshold-based replenishment logic. The search thereby produces new policy-class functional forms that, to our knowledge, have not previously been studied in the lost-sales inventory literature. These functional forms are not specified ex ante but emerge from the search process. Moreover, after their parameters are re-optimized, three discovered policy classes achieve average cost reductions of 21.75% to 22.60% across 10,064 new inventory instances. Overall, the results show that data-driven parameter optimization can guide LLM-based search over a broad space of inventory policy classes and identify high-performing, interpretable, and transferable decision rules.

cs.AI↗

arXiv2026

Prompt Injection in Automated Résumé Screening with Large Language Models: Single and Multi-Injection Settings

Large language models (LLMs) are increasingly used to screen and rank job applicants, creating incentives for candidates to strategically manipulate algorithmic hiring systems. We study prompt injection in automated résumé screening, defined as subtle self-promotional text that introduces no new qualifications but is designed to influence LLM evaluations. Using controlled experiments, we show that prompt injection reliably improves applicant rankings when résumé quality is homogeneous and few candidates inject. However, its effectiveness rapidly diminishes as more candidates inject, collapsing when manipulation becomes widespread. When candidate quality is heterogeneous, prompt injection is less effective on average, but can occasionally allow lower-quality candidates to outrank higher-quality ones, raising fairness concerns. Overall, LLM-based screening is most vulnerable when manipulation is rare and candidate quality differences are small. Code and resources are publicly available at: https://github.com/preetb1199/Prompt_Injection_ACL26

cs.AI↗

arXiv2026

From Small to Large: A Graph Convolutional Network Approach for Solving Assortment Optimization Problems

Assortment optimization seeks to select a subset of substitutable products, subject to constraints, to maximize expected revenue. The problem is NP-hard due to its combinatorial and nonlinear nature and arises frequently in industries such as e-commerce, where platforms must solve thousands of such problems each minute. We propose a graph convolutional network (GCN) framework to efficiently solve constrained assortment optimization problems. Our approach constructs a graph representation of the problem, trains a GCN to learn the mapping from problem parameters to optimal assortments, and develops three inference policies based on the GCN's output. Owing to the GCN's ability to generalize across instance sizes, patterns learned from small-scale samples can be transferred to large-scale problems. Theoretical results are established to show the expressive power of the proposed GCN, and explain the underlying mechanism of the size generalization ability. Numerical experiments show that a GCN trained on instances with 20 products achieves over 85% of the optimal revenue on problems with up to 2,000 products within seconds, outperforming existing heuristics in both accuracy and efficiency. We further extend the framework to settings with an unknown choice model using transaction data and demonstrate similar performance and scalability.

cs.LG↗

arXiv2026

A General Framework for Optimal Group Sequential Testing via Mixed-Integer Linear Programming

Sequential hypothesis tests are widely adopted as a principled way to perform multiple tests on data that arrives over time. In particular, researchers frequently utilize group sequential hypothesis tests (GST) to test the same hypotheses at K times or "groups" while data arrives sequentially. In this setting, many methods have been proposed to allow researchers to uniformly control type-1 error across K checks (often known as various alpha-spending budgets). Although these methods are all successfully valid in controlling uniform type-1 error, it is not clear which of these methods are optimal when trying to reject the null as soon as possible. In this paper, we directly optimize the rejection criterion in the GST setting under the same constraints of controlling type-1 and type-2 errors. We use a sample average approximation combined with mixed integer linear programming (S-MILP) approach for this problem and show how our S-MILP approach dominates classical GST procedures such as Lan-DeMets, Pocock, and O'Brien-Fleming methods. We also find that the optimal solution typically aggressively spends the alpha-budget early, shedding insight to the long-standing debate of which alpha-spending budgets are more efficient. We finally apply our optimal S-MILP approach to a recent study on acute kidney injury interventions and find our optimal S-MILP approach can reach the same statistically significant conclusion faster than the original study and other GST methods.

stat.ME↗

arXiv2026

Asymptotically Optimal Sequential Testing with Heterogeneous LLMs

We study a Bayesian binary sequential hypothesis testing problem with multiple large language models (LLMs). Each LLM $j$ has per-query cost $c_j>0$, random waiting time with mean $μ_j>0$ and sub-Gaussian tails, and \emph{asymmetric} accuracies: the probability of returning the correct label depends on the true hypothesis $θ\in\{A,B\}$ and needs not be the same under $A$ and $B$. This asymmetry induces two distinct information rates $(I_{j,A}, I_{j,B})$ per LLM, one under each hypothesis. The decision-maker chooses LLMs sequentially, observes their noisy binary answers, and stops when the posterior probability of one hypothesis exceeds $1-α$. The objective is to minimize the sum of expected query cost and expected waiting cost, $\mathbb{E}[C_π] + \mathbb{E}[g(W_π)]$, where $C_π$ is the total query cost, $W_π$ is the total waiting time and $g$ is a polynomial function (e.g., $g(x)=x^ρ$ with $ρ\ge 1$). We prove that as the error tolerance $α\to0$, the optimal policy is asymptotically equivalent to one that uses at most two LLMs. In this case, a single-LLM policy is \emph{not} generically optimal: optimality now requires exploiting a two-dimensional tradeoff between information under $A$ and information under $B$. Any admissible policy induces an expected information-allocation vector in $\mathbb{R}_+^2$, and we show that the optimal allocation lies at an extreme point of the associated convex set when $α$ is relatively small, and hence uses at most two LLMs. We construct belief-dependent policies that first mix between two LLMs when the posterior is ambiguous, and then switch to a single "specialist" LLM when the posterior is sufficiently close to one of the hypotheses. These policies match the universal lower bound up to a $(1+o(1))$ factor as $α\rightarrow 0$.

cs.DS↗

arXiv2026

Multi-LLM Query Optimization

Deploying multiple large language models (LLMs) in parallel to classify an unknown ground-truth label is a common practice, yet the problem of optimally allocating queries across heterogeneous models remains poorly understood. In this paper, we formulate a robust, offline query-planning problem that minimizes total query cost subject to statewise error constraints which guarantee reliability for every possible ground-truth label. We first establish that this problem is NP-hard via a reduction from the minimum-weight set cover problem. To overcome this intractability, we develop a surrogate by combining a union bound decomposition of the multi-class error into pairwise comparisons with Chernoff-type concentration bounds. The resulting surrogate admits a closed-form, multiplicatively separable expression in the query counts and is guaranteed to be feasibility-preserving. We further show that the surrogate is asymptotically tight at the optimization level: the ratio of surrogate-optimal cost to true optimal cost converges to one as error tolerances shrink, with an explicit rate of $O\left(\log\log(1/α_{\min}) / \log(1/α_{\min})\right)$. Finally, we design an asymptotic fully polynomial-time approximation scheme (AFPTAS) that returns a surrogate-feasible query plan within a $(1+\varepsilon)$ factor of the surrogate optimum.

cs.DS↗

arXiv2026

Minimizing Type 2 Errors in an Experiment-Rich Regime via Optimal Resource Allocation

Randomized experiments (often known as "A/B tests") are widely used to evaluate product and service innovations. We study how to allocate limited experimentation resources across M concurrent experiments in an experiment-rich regime. Existing work on allocation has predominantly focused on minimizing the worst-case mean squared error (MSE) of estimated treatment effects, which favors experiments with larger (and typically unknown) outcome variance. While appropriate for controlling estimation accuracy, this objective does not directly capture a common managerial priority in screening stages: detecting practically meaningful treatment effects with high probability. Motivated by this, we consider the objective of minimizing the worst-case Type II error across all experiments. When the standard deviations are known, we characterize the power-optimal allocation and show that MSE-based allocations can be highly inefficient for detection, even though the two objectives align asymptotically. When the standard deviations are unknown and must be learned from pilot data, we show that a naive plug-in approach, treating pilot standard deviations as truth, can suffer substantial power loss. We propose inflating pilot estimates via correction factors and develop three optimization-based frameworks for selecting them, each reflecting a different risk criterion with distinct managerial implications. Although the resulting stochastic programs are computationally challenging at scale, we derive tractable surrogate reformulations inspired by robust optimization and establish favorable theoretical properties. We further propose Surrogate-S, a fully data-dependent and implementable procedure that computes correction factors using only pilot variance estimates and achieves near-oracle performance in numerical experiments.

stat.ME↗

arXiv2026

Decentralized Multi-product Pricing: Diagonal Dominance, Nash Equilibrium, and Price of Anarchy

Decentralized decision making in multi--product firms can lead to efficiency losses when autonomous decision makers fail to internalize cross--product demand interactions. This paper quantifies the magnitude of such losses by analyzing the Price of Anarchy in a pricing game in which each decision maker independently sets prices to maximize its own product--level revenue. We model demand using a linear system that captures both substitution and complementarity effects across products. We first establish existence and uniqueness of a pure--strategy Nash equilibrium under economically standard diagonal dominance conditions. Our main contribution is the derivation of a tight worst--case lower bound on the ratio between decentralized revenue and the optimal centralized revenue. We show that this efficiency loss is governed by a single scalar parameter, denoted by $μ$, which measures the aggregate strength of cross--price effects relative to own--price sensitivities. In particular, we prove that the revenue ratio is bounded below by $4(1-μ)/(2-μ)^2$, and we demonstrate the tightness of this bound by constructing a symmetric market topology in which the bound is exactly attained. We further refine the analysis by providing an instance--exact characterization of efficiency loss based on the spectral properties of the demand interaction matrix. Together, these results offer a quantitative framework for assessing the trade--off between centralized pricing and decentralized autonomy in multi--product firms.

cs.GT↗

arXiv2025

Adaptive Risk Mitigation in Demand Learning

We study dynamic pricing of a product with an unknown demand distribution over a finite horizon. Departing from the standard no-regret learning environment in which prices can be adjusted at any time, we restrict price changes to predetermined points in time to reflect common retail practice. This constraint, coupled with demand model ambiguity and an unknown customer arrival pattern, imposes a high risk of revenue loss, as a price based on a misestimated demand model may be applied to many customers before it can be revised. We develop an adaptive risk learning (ARL) framework that embeds a data-driven ambiguity set (DAS) to quantify demand model ambiguity by adapting to the unknown arrival pattern. Initially, when arrivals are few, the DAS includes a broad set of plausible demand models, reflecting high ambiguity and revenue risk. As new data is collected through pricing, the DAS progressively shrinks, capturing the reduction in model ambiguity and associated risk. We establish the probabilistic convergence of the DAS to the true demand model and derive a regret bound for the ARL policy that explicitly links revenue loss to the data required for the DAS to identify the true model with high probability. The dependence of our regret bound on the arrival pattern is unique to our constrained dynamic pricing problem and contrasts with no-regret learning environments, where regret is constant and arrival-pattern independent. Relaxing the constraint on infrequent price changes, we show that ARL attains the known constant regret bound. Numerical experiments further demonstrate that ARL outperforms benchmarks that prioritize either regret or risk alone by adaptively balancing both without knowledge of the arrival pattern. This adaptive risk adjustment is crucial for achieving high revenues and low downside risk when prices are sticky and both demand and arrival patterns are unknown.

cs.LG↗

arXiv2022

Phase Transitions in Learning and Earning under Price Protection Guarantee

Motivated by the prevalence of ``price protection guarantee", which allows a customer who purchased a product in the past to receive a refund from the seller during the so-called price protection period (typically defined as a certain time window after the purchase date) in case the seller decides to lower the price, we study the impact of such policy on the design of online learning algorithm for data-driven dynamic pricing with initially unknown customer demand. We consider a setting where a firm sells a product over a horizon of $T$ time steps. For this setting, we characterize how the value of $M$, the length of price protection period, can affect the optimal regret of the learning process. We show that the optimal regret is $\tildeΘ(\sqrt{T}+\min\{M,\,T^{2/3}\})$ by first establishing a fundamental impossible regime with novel regret lower bound instances. Then, we propose LEAP, a phased exploration type algorithm for \underline{L}earning and \underline{EA}rning under \underline{P}rice Protection to match this lower bound up to logarithmic factors or even doubly logarithmic factors (when there are only two prices available to the seller). Our results reveal the surprising phase transitions of the optimal regret with respect to $M$. Specifically, when $M$ is not too large, the optimal regret has no major difference when compared to that of the classic setting with no price protection guarantee. We also show that there exists an upper limit on how much the optimal regret can deteriorate when $M$ grows large. Finally, we conduct extensive numerical experiments to show the benefit of LEAP over other heuristic methods for this problem.

stat.ML↗

arXiv2020

Algorithmic Transparency with Strategic Users

Should firms that apply machine learning algorithms in their decision-making make their algorithms transparent to the users they affect? Despite growing calls for algorithmic transparency, most firms have kept their algorithms opaque, citing potential gaming by users that may negatively affect the algorithm's predictive power. We develop an analytical model to compare firm and user surplus with and without algorithmic transparency in the presence of strategic users and present novel insights. We identify a broad set of conditions under which making the algorithm transparent benefits the firm. We show that, in some cases, even the predictive power of machine learning algorithms may increase if the firm makes them transparent. By contrast, users may not always be better off under algorithmic transparency. The results hold even when the predictive power of the opaque algorithm comes largely from correlational features and the cost for users to improve on them is close to zero. Overall, our results show that firms should not view manipulation by users as bad. Rather, they should use algorithmic transparency as a lever to motivate users to invest in more desirable features.

cs.GT↗