arXiv subjects

Tinghan Ye

Publications and source records attributed to Tinghan Ye.

14 recordsLinked to original sources

arXiv2026

LLM-Derived Preference Judgments Are Not Self-Consistent

Agents increasingly interpret a person's natural-language preferences by querying an LLM for numerical preference judgments, e.g., by asking how much the person would be willing to pay for an item. A growing body of work estimates a utility function from these judgments and then chooses actions based on their estimated utility. This pipeline assumes the judgments are approximately self-consistent: that a single utility function can reproduce them. But are they? To study this question, we measure the self-consistency of cardinal LLM preference judgments. For example, the difference in stated willingness-to-pay between two items should match the stated payment that makes a person indifferent to exchanging them. We develop statistical tests and interpretable measures of how far observed responses depart from the best-fitting self-consistent utility function. Experiments with flight, apartment, and hotel examples across six LLMs reveal large persistent inconsistencies. This suggests that LLM-derived preference judgments cannot be faithfully summarized by a single utility function.

cs.AI↗

arXiv2026

Learning Optimization Proxies for Sequential Contextual Stochastic Programs: An Order Fulfillment Application

Sequential contextual stochastic programs model real-time decision systems in which each time epoch commits to an action under uncertainty whose consequences propagate into future decisions. In many practical contexts, these programs require obtaining solutions rapidly as new information becomes available. These problems can be represented through scenario approximations to be solved by off-the-shelf optimization solvers, which achieve high decision quality offline but typically run in seconds to minutes per instance, falling short of the sub-second responses that peak periods of planning require. This paper develops a learning-based optimization proxy: a scenario-embedded neural network trained offline on solver-generated labels, paired online with a decoder that enforces feasibility, replacing the per-epoch solve with a single forward pass. The framework is specialized to omnichannel order fulfillment, where each arriving order requires a sub-second assignment of products to distribution centers and carrier services under stochastic delivery times and future demand. A two-stage contextual stochastic program is introduced to formulate this problem, and its contextual sample average approximation (C-SAA) supplies the offline labels, while a composite training loss combines label imitation, a constraint-violation penalty, and self-supervised cost alignment. In a calibrated simulator built from JD.com transactional records, a detailed computational study is provided. The proxy reduces decision latency by roughly 2800x relative to the online finite-sample C-SAA reference and improves over it by 3.3% in realized fulfillment cost. Relative to established fulfillment policies, the proxy lowers total realized cost by at least 10.7% and roughly halves the late-delivery rate.

math.OC↗

arXiv2026

Decision-Focused On-Policy Learning for Contextual Linear Optimization with Partial Feedback

Decision-focused learning (DFL) trains predictive models by optimizing downstream decision quality rather than standalone prediction accuracy. For contextual linear optimization, most existing DFL methods assume offline data and full observations of the objective cost vector. We develop an on-policy learning method for sequential contextual linear optimization under partial feedback, generalizing the standard bandit feedback setting. Our method learns a stochastic predict-then-optimize policy that samples a cost-vector prediction from a conditional distribution and solves the resulting downstream linear optimization problem. To update this distributional model, we introduce a two-component hybrid gradient estimator. The first component is a score function estimator, which provides an unbiased but potentially high-variance policy gradient estimate. The second is a decision-focused plug-in component that uses an auxiliary nuisance estimate of the latent cost vector to exploit the downstream optimization structure, becoming more informative as the estimate improves. We prove an $\mathcal{O}(T^{-1/2})$ bound on the average squared policy-gradient norm, matching the standard non-convex SGD rate. Experiments on top-$k$ selection, shortest path, combinatorial pricing, and a real-data energy-scheduling benchmark show that the hybrid gradient approach achieves lower cumulative regret than contextual-bandit-style baselines across all benchmarks, using both Gaussian and richer conditional generative models. Code is available at https://github.com/Joeyetinghan/on-policy-bandit-dfl.

cs.LG↗

arXiv2026

HawkesLLM: Semantic Uncertainty Propagation in Agentic Text Simulation

Agentic text-simulation systems write in sequence, with each item becoming possible context for later steps. That makes uncertainty path-dependent: an early ambiguity can affect later outputs. This paper studies this problem with HawkesLLM, a framework that separates temporal influence modeling from text generation. We represent the cascade as a network whose nodes are text-generating agents. A multivariate Hawkes process models how these nodes activate over time and which earlier node outputs should influence later prompts. A language model then writes each new event from the compact memory selected by this temporal model. We evaluate the framework on a held-out Global Database of Events, Language, and Tone (GDELT) news-cascade case study. The diagnostics track semantic alignment with local held-out references and separate local drift from global drift. In this setting, HawkesLLM improves late-stage semantic alignment under a compact prompt-memory budget.

cs.CL↗

arXiv2026

Democratizing Large-Scale Re-Optimization with LLM-Guided Model Patches

Optimization models developed by operations research (OR) experts are often deployed as decision-support systems in industrial settings. However, real-world environments are dynamic, with evolving business rules and unforeseen perturbations. In such contexts, end users should ideally re-optimize models to recover feasible and implementable solutions, often without access to the original model developers. This paper introduces an agentic re-optimization framework in which a large language model (LLM) acts as an OR expert, dynamically supporting end users through natural-language interaction. The LLM translates user prompts into structured updates of the underlying optimization model, selects suitable re-optimization techniques from an optimization toolbox, and solves the resulting instance to return implementable solutions. The toolbox leverages primal information, including historical solutions, valid inequalities, solver configurations, and metaheuristics, to accelerate re-optimization while preserving solution quality. The proposed framework enables interactive and continuous adaptation of deployed optimization models, reducing dependence on OR experts, and improving the sustainability of decision-support systems. Extensive experiments on two complementary large-scale real-world case studies demonstrate the effectiveness and scalability of the proposed framework. The first considers online supply chain re-optimization, where solutions must be generated rapidly while remaining close to the deployed plan, whereas the second focuses on offline university exam scheduling, where solution quality is prioritized over runtime. Results show that the toolbox-driven architecture significantly improves computational efficiency through primal-based and solver-aware re-optimization techniques, while the structured patch-based updates improve interpretability and traceability of model modifications.

cs.AI↗

arXiv2026

SkillsBench: Benchmarking How Well Agent Skills Work Across Diverse Tasks

Agent Skills are structured packages of procedural knowledge that augment large language model (LLM) agents at inference time. Despite rapid adoption, there is no standard way to measure whether they actually help. We present SkillsBench, a benchmark whose current inventory contains 87 tasks across 8 domains paired with curated Skills and deterministic verifiers. Our latest aggregate evaluation runs the 87-task benchmark under matched no-Skills and curated-Skills conditions for 18 model-harness configurations. Curated Skills raise the average pass rate from 33.9% to 50.5% (+16.6 percentage points; 25.5% normalized gain), with configuration-level gains ranging from +4.1 to +25.7 pp. Focused Skills with at most three modules outperform larger or exhaustive bundles, and smaller models with Skills can match larger models without them. SkillsBench establishes paired evaluation as the foundation for rigorous measurement of Skill efficacy on agentic, expertise-heavy work.

cs.AI↗

arXiv2026

Plan, Verify and Fill: A Structured Parallel Decoding Approach for Diffusion Language Models

Diffusion Language Models (DLMs) present a promising non-sequential paradigm for text generation, distinct from standard autoregressive (AR) approaches. However, current decoding strategies often adopt a reactive stance, underutilizing the global bidirectional context to dictate global trajectories. To address this, we propose Plan-Verify-Fill (PVF), a training-free paradigm that grounds planning via quantitative validation. PVF actively constructs a hierarchical skeleton by prioritizing high-leverage semantic anchors and employs a verification protocol to operationalize pragmatic structural stopping where further deliberation yields diminishing returns. Extensive evaluations on LLaDA-8B-Instruct and Dream-7B-Instruct demonstrate that PVF reduces the Number of Function Evaluations (NFE) by up to 65% compared to confidence-based parallel decoding across benchmark datasets, unlocking superior efficiency without compromising accuracy.

cs.CL↗

arXiv2025

LISTEN to Your Preferences: An LLM Framework for Multi-Objective Selection

Human experts often struggle to select the best option from a large set of items with multiple competing objectives, a process bottlenecked by the difficulty of formalizing complex, implicit preferences. To address this, we introduce LISTEN (LLM-based Iterative Selection with Trade-off Evaluation from Natural-language), an agentic LLM-based framework that treats the LLM as a decision-making agent capable of iteratively refining its internal preference model and taking actions (e.g., proposing utilities or selecting candidates) to maximize alignment with a user's implicit goals. To operate within LLM constraints like context windows and inference costs, we propose two iterative algorithms: LISTEN-U, which uses the LLM to refine a parametric utility function, and LISTEN-T, a non-parametric method that performs tournament-style selections over small batches of solutions. Evaluated on diverse tasks including flight booking, shopping, and exam scheduling, our results show LISTEN-U excels when preferences are parametrically aligned (a property we measure with a novel concordance metric), while LISTEN-T offers more robust performance overall. This work explores a promising direction for steering complex multi-objective decisions directly with natural language, reducing the cognitive burden of traditional preference elicitation. Code is available at https://github.com/AdamJovine/LISTEN; data is available at https://huggingface.co/datasets/AdamJovine/LISTEN-benchmark.

cs.CL↗

arXiv2025

Paratransit Optimization with Constraint Programming: A Case Study in Savannah, Georgia

Paratransit services are vital for individuals who cannot use fixed-route public transit, including those with disabilities. Optimizing these services is essential for transit agencies to deliver high-quality service efficiently. This paper introduces a Constraint Programming (CP) model to jointly optimize route planning and shift scheduling for paratransit operations, along with practical guidance for real-world implementation. A case study in Savannah, Georgia, demonstrates that the new approach is competitive with a recently proposed, highly effective AI-accelerated column generation framework, and significantly increases the number of requests served compared to current practices. The method is also easier to implement and provides an inherently practical solution for transportation planners. CP further provides the flexibility to optimize schedules without requiring shifts to start exactly on the hour, yielding an additional 5% improvement in the number of requests served.

math.OC↗

arXiv2025

Conformal Predictive Distributions for Order Fulfillment Time Forecasting

Accurate estimation of order fulfillment time is critical for e-commerce logistics, yet traditional rule-based approaches often fail to capture the inherent uncertainties in delivery operations. This paper introduces a novel framework for distributional forecasting of order fulfillment time, leveraging Conformal Predictive Systems and Cross Venn-Abers Predictors -- model-agnostic techniques that provide rigorous coverage or validity guarantees. The proposed machine learning methods integrate granular spatiotemporal features, capturing fulfillment location and carrier performance dynamics to enhance predictive accuracy. Additionally, a cost-sensitive decision rule is developed to convert probabilistic forecasts into reliable point predictions. Experimental evaluation on a large-scale industrial dataset demonstrates that the proposed methods generate competitive distributional forecasts, while machine learning-based point predictions significantly outperform the existing rule-based system -- achieving up to 14% higher prediction accuracy and up to 75% improvement in identifying late deliveries.

cs.LG↗

arXiv2024

Contextual Stochastic Optimization for Omnichannel Multi-Courier Order Fulfillment Under Delivery Time Uncertainty

The paper studies a large-scale order fulfillment problem for a leading e-commerce company in the United States. The challenge involves selecting fulfillment centers and shipping carriers with observational data only to efficiently process orders from a vast network of physical stores and warehouses. The company's current practice relies on heuristic rules that choose the cheapest fulfillment and shipping options for each unit, without considering opportunities for batching items or the reliability of carriers in meeting expected delivery dates. The paper develops a data-driven Contextual Stochastic Optimization (CSO) framework that integrates distributional forecasts of delivery time deviations with stochastic and robust order fulfillment optimization models. The framework optimizes the selection of fulfillment centers and carriers, accounting for item consolidation and delivery time uncertainty. Validated on a real-world data set containing tens of thousands of products, each with hundreds to thousands of fulfillment options, the proposed CSO framework significantly enhances the accuracy of meeting customer-expected delivery dates compared to current practices. It provides a flexible balance between reducing fulfillment costs and managing delivery time deviation risks, emphasizing the importance of contextual information and distributional forecasts in order fulfillment. This is the first paper that studies the omnichannel multi-courier order fulfillment problem with delivery time uncertainty through the lens of contextual optimization, fusing machine learning and optimization.

math.OC↗

arXiv2024

Cornell University Uses Integer Programming to Optimize Final Exam Scheduling

This paper presents an integer programming-based optimization framework designed to effectively address the complex final exam scheduling challenges encountered at Cornell University. With high flexibility, the framework is specifically tailored to accommodate a variety of different constraints, including the front-loading of large courses and the exclusion of specific time slots during the exam period. By generating multiple scheduling model variants and incorporating heuristic approaches, our framework enables comprehensive comparisons of different schedules. This empowers the University Registrar to make informed decisions, considering trade-offs in terms of schedule comfort measured by different levels of exam conflicts. Our results demonstrate significant advantage over the historical lecture time-based approach, providing time and effort savings for the university administration while enhancing student and faculty satisfaction.

math.OC↗

arXiv2024

Boosting Column Generation with Graph Neural Networks for Joint Rider Trip Planning and Crew Shift Scheduling

Optimizing service schedules is pivotal to the reliable, efficient, and inclusive on-demand mobility. This pressing challenge is further exacerbated by the increasing needs of an aging population, the oversubscription of existing services, and the lack of effective solution methods. This study addresses the intricacies of service scheduling, by jointly optimizing rider trip planning and crew scheduling for a complex dynamic mobility service. The resulting optimization problems are extremely challenging computationally for state-of-the-art methods. To address this fundamental gap, this paper introduces the Joint Rider Trip Planning and Crew Shift Scheduling Problem (JRTPCSSP) and a novel solution method, called Attention and Gated GNN-Informed Column Generation (AGGNNI-CG), that hybridizes column generation and machine learning to obtain near-optimal solutions to the JRTPCSSP with real-life constraints of the application. The key idea of the machine-learning component is to dramatically reduce the number of paths to explore in the pricing problem, accelerating the most time-consuming component of the column generation. The machine learning component is a graph neural network with an attention mechanism and a gated architecture, which is particularly suited to cater for the different input sizes coming from daily operations. AGGNNI-CG has been applied to a challenging, real-world dataset from the Paratransit system of Chatham County in Georgia. It produces substantial improvements compared to the baseline column generation approach, which typically cannot produce high-quality feasible solutions in reasonable time on large-scale complex instances. AGGNNI-CG also produces significant improvements in service quality compared to the existing system.

math.OC↗

arXiv2022

A min-max theorem for the minimum fleet-size problem

A retrospective fleet-sizing problem can be solved via bipartite matching, where a maximum cardinality matching corresponds to the minimum number of vehicles needed to cover all trips. We prove a min-max theorem on this minimum fleet-size problem: the maximum number of pairwise incompatible trips is equal to the minimum fleet size needed.

cs.DM↗