Chonghe Jiang, Bao Nguyen, Anthony Man-Cho So, Viet Anh Nguyen. Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers). 2025.

In recent years, the Vehicle Routing Problem, with its variants handling complex realistic constraints, continues to be a key area of research, reflecting its enduring importance in modernized societies. Nevertheless, less attention has been paid to the variant with the flexibility of customer demands. This paper introduces the problem of addressing the integration of drones into the existing transportation system to maximize the profit of distributing freight during a limited working time by allowing flexibility in the delivered quantity. We propose a Mixed Integer Linear Programming (MILP) formulation and a genetic algorithm (GA) to solve the problem. To assess the proposed algorithm, we create the benchmark instances of the problem from published data sources. Experimental results on the instances show a significantly better performance of our proposal than MILP. Moreover, we evaluate our approach by comparing its performance with existing solutions in published closely related works.

Flow matching is a powerful framework for generating high-quality samples in various applications, especially image synthesis. However, the intensive computational demands of these models, especially during the finetuning process and sampling processes, pose significant challenges for low-resource scenarios. This paper introduces Bellman Optimal Stepsize Straightening (BOSS) technique for distilling flow-matching generative models: it aims specifically for a few-step efficient image sampling while adhering to a computational budget constraint. First, this technique involves a dynamic programming algorithm that optimizes the stepsizes of the pretrained network. Then, it refines the velocity network to match the optimal step sizes, aiming to straighten the generation paths. Extensive experimental evaluations across image generation tasks demonstrate the efficacy of BOSS in terms of both resource utilization and image quality. Our results reveal that BOSS achieves substantial gains in efficiency while maintaining competitive sample quality, effectively bridging the gap between low-resource constraints and the demanding requirements of flow-matching generative models. Our paper also fortifies the responsible development of artificial intelligence, offering a more sustainable generative model that reduces computational costs and environmental footprints. Our code can be found at https://github.com/nguyenngocbaocmt02/BOSS.

Learning conditional distributions is challenging because the desired outcome is not a single distribution but multiple distributions that correspond to multiple instances of the covariates. We introduce a novel neural entropic optimal transport method designed to effectively learn generative models of conditional distributions, particularly in scenarios characterized by limited sample sizes. Our method relies on the minimax training of two neural networks: a generative network parametrizing the inverse cumulative distribution functions of the conditional distributions and another network parametrizing the conditional Kantorovich potential. To prevent overfitting, we regularize the objective function by penalizing the Lipschitz constant of the network output. Our experiments on real-world datasets show the effectiveness of our algorithm compared to state-of-the-art conditional distribution learning techniques. Our implementation can be found at https://github.com/nguyenngocbaocmt02/GENTLE.

Given a (machine learning) classifier and a collection of unlabeled data, how can we efficiently identify misclassification patterns presented in this dataset? To address this problem, we propose a human-machine collaborative framework that consists of a team of human annotators and a sequential recommendation algorithm. The recommendation algorithm is conceptualized as a stochastic sampler that, in each round, queries the annotators a subset of samples for their true labels and obtains the feedback information on whether the samples are misclassified. The sampling mechanism needs to balance between discovering new patterns of misclassification (exploration) and confirming the potential patterns of classification (exploitation). We construct a determinantal point process, whose intensity balances the exploration-exploitation trade-off through the weighted update of the posterior at each round to form the generator of the stochastic sampler. The numerical results empirically demonstrate the competitive performance of our framework on multiple datasets at various signal-to-noise ratios.

Adaptive mesh refinement based on octree data structures has enabled efficient simulations of complex physical phenomena. Existing meshing algorithms were proposed with the assumption that computer memory is volatile. Consequently, for failure recovery, in-core algorithms need to save memory states as snapshots with slow file I/O, while out-of-core algorithms store octants on disk for persistence. However, neither was designed to best exploit the unique characteristics of non-volatile byte-addressable memory (NVBM). We propose a novel data structure, the Distributed Persistent Merged octree (DPM-octree), for both meshing and in-memory storage of persistent octrees using NVBM. DPM-octree is a multi-version data structure that can recover from failures using an earlier persistent version stored in NVBM. In addition, we design a feature-directed sampling approach to help dynamically transform the DPM-octree layout for reducing NVBM-induced memory write latency. DPM-octree uses parity trees which are created using erasure coding and stored in NVBM to support low-latency in-memory octant recovery after data loss. DPM-octree has been successfully integrated with the Gerris software for simulation of fluid dynamics. Our experimental results with real-world scientific workloads show that DPM-octree scales up to 1.1 billion mesh elements with 1,000 processors on the Titan supercomputer.

Data-intensive knowledge discovery requires scientific applications to run concurrently with analytics and visualization codes, executing in situ for timely output inspection and knowledge extraction. Consequently, I/O pipelines of scientific workflows can be long and complex because they comprise many "stages" of analytics across different layers of the I/O stack of high-performance computing systems. Performance limitations at any I/O layer or stage can cause an I/O bottleneck resulting in longer than expected end-to-end I/O latency. The causes of such performance issues are missing a performance guarantee (e.g., lower bounds of I/O throughput) across stages of I/O pipelines and across layers of the I/O stacks. In this paper, we present the design and implementation of a novel data management infrastructure called Software-defined Storage Resource Enclaves (SIREN) at system levels to enforce end-to-end policies that dictate an I/O pipeline's performance. Our results demonstrate that SIREN provides performance isolation among scientific workflows sharing multiple storage servers across two I/O layers while maintaining high system scalability and resource utilization.

In-situ analytics have been increasingly adopted by leadership scientific applications to gain fast insights into massive output data of simulations. Current practice buffers the output data in DRAM for analytics processing, constraining it to DRAM capacity un-used by the simulation. The rapid growth of data size requires alternative approaches to accommodating data-rich analytics, such as using solid-state disks (SSDs) to increase effective memory capacity. For this purpose, this paper explores software solutions for exploring the deep memory hierarchies expected on future high-end machines. Leveraging the fact that many analytics are sensitive to data features (regions- of-interest) hidden in the data being processed, the approach incorporates the knowledge of the data features into in-situ data management. It uses adaptive index creation/refinement to reduce the overhead of index management. In addition, it uses data features to predict data skew and improve load balance through controlling data distribution and placement on distributed staging servers. The experimental results show that such feature-guided optimizations achieve substantial improvements over state-of-the-art approaches for managing output data in-situ.

Octree-based mesh adaptation has enabled simulations of complex physical phenomena. Existing meshing algorithms were proposed with the assumption that computer memory is volatile. Consequently, for failure recovery, the in-core algorithms need to save memory states as snapshots with slow file I/Os. The out-of-core algorithms store octants on disks for persistence. However, neither of them was designed to leverage unique characteristics of non-volatile byte-addressable memory (NVBM). In this paper, we propose a novel data structure Persistent Merged octree (PM-octree) for both meshing and in-memory storage of persistent octrees using NVBM. It is a multi-version data structure and can recover from failures using its earlier persistent version stored in NVBM. In addition, we design a feature-directed sampling approach to help dynamically transform the PM-octree layout for reducing NVBM-induced memory write latency. PM-octree has been successfully integrated with Gerris software for simulation of fluid dynamics. Our experimental results with real-world scientific workloads show that PM-octree scales up to 1.1 billion mesh elements with 1000 processors on the Titan supercomputer.

By exploiting superimposed signals at relays, physicallayer network coding (PNC) could significantly improve the throughput of wireless communications systems. However, it is challenging to achieve perfect synchronization between superimposed signals at the relay. In this paper, we propose an approach to resolve the issue of frame-misalignment in PNC using type-I Euclidean geometry low-density parity-check (EG-LDPC) codes with cyclic prefix or zero padding. To estimate the arrival delay between two transmitted signals at the relay, Gold sequences are adopted as pilot sequences in our transmit frames. Simulation results show that our approach can effectively resolve the frame-misalignment issue in PNC.

Automatic Target Recognition (ATR) methods have been successfully applied to detect possible objects or regions of interest in sonar imagery. It is anticipated that the additional information obtained from additional views of an object should improve the classification performance over single-aspect classification. In this paper the detection of mine-like objects (MLO) on the seabed from multiple side-scan sonar views is considered. We transform the multiple-aspect classification problem into a multiple-instance learning problem and present a framework based upon the concepts of multiple-instance classifiers. Moreover, we present another framework based upon the Dempster-Shafer (DS) concept of fusion from single-view classifiers. Our experimental results indicate that both the presented frameworks can be successfully used in mine-like object classification.

A typical firing doctrine is the Shoot-Look-Shoot tactic. In this tactic, the defence launches a salvo of interceptors against the targets (Shoot), assesses the outcomes of the engagements (Shoot-Look), and launches another salvo (Shoot-Look-Shoot) if time and the inventory of interceptors permit. In the open literature, it is often assumed that the targets are identical. This is not always true as targets come in with different ranges, speeds, sizes, cross sections etc. In this paper, we consider two types of targets. Each type of target has a different number engagement opportunities due to their ranges and speeds. Through the use of dynamic programming, a genetic algorithm, and a recursive generating function, we determine the probability of raid annihilation (the probability of neutralizing all of the targets) for two different Shoot-Look-Shoot (SLS) tactics. The first SLS tactic is based on variable size salvos and maximizes the probability of raid annihilation (PRA) for heterogeneous targets. The second SLS tactic is based on fixed-size salvos and is robust as it is independent of the number and types of targets. Theoretical results are validated through some computer simulations.

In this paper, we examine scenarios of maritime air defence where the threats can be of different types i.e., characterized by varying different single shot probabilities of hit. We develop a methodology that integrates constrained optimization calculus, dynamic programming and a genetic algorithm to determine the optimal allocation of interceptors that globally maximize the probability of raid annihilation. We call this strategy the Heterogeneous tactic. We show that the improvement in the probability of raid annihilation could exceed thirty percent for one scenario examined, relative to a Shoot-Look-Shoot tactic with fixed size salvos.

Real world data mining applications such as Mine Countermeasure Missions (MCM) involve learning from imbalanced data sets, which contain very few instances of the minority classes and many instances of the majority class. For instance, the number of naturally occurring clutter objects (such as rocks) that are detected typically far outweighs the relatively rare event of detecting a mine. In this paper we propose support vector machine with adaptive asymmetric misclassification costs (instances weighted) to solve the skewed vector spaces problem in mine countermeasure missions. Experimental results show that the given algorithm could be used for imbalanced sonar image data sets and makes an improvement in prediction performance.

The IEEE RAS Ontologies for Robotics and Automation Working Group is dedicated to developing a methodology for knowledge representation and reasoning in robotics and automation. As part of this working group, the Autonomous Robots sub-group is tasked with developing ontology modules for autonomous robots. This paper describes the work in progress on the development of ontologies for autonomous systems. For autonomous systems, the focus is on the cooperation, coordination, and communication of multiple unmanned aerial vehicles (UAVs), unmanned ground vehicles (UGVs), and autonomous underwater vehicles (AUVs). The ontologies serve as a framework for working out concepts of employment with multiple vehicles for a variety of operational scenarios with emphasis on collaborative and cooperative missions.