arXiv subjects

Cody Fleming

Publications and source records attributed to Cody Fleming.

At least 37 records · Page 2Linked to original sources

arXiv2026

Toward Faithful and Complete Answer Construction from a Single Document

Modern large language models (LLMs) are powerful generators driven by statistical next-token prediction. While effective at producing fluent text, this design biases models toward high-probability continuations rather than exhaustive and faithful answers grounded in source content. As a result, directly applying LLMs lacks systematic mechanisms to ensure both completeness (avoiding omissions) and faithfulness (avoiding unsupported content), which fundamentally conflicts with core AI safety principles. To address this limitation, we present EVE, a structured framework for document-grounded reasoning. Unlike free-form prompting, EVE constrains generation to a structured, verifiable pipeline that decomposes high-rigor reasoning into extraction, validation, and enumeration. Empirically, this design enables consistent and simultaneous improvements in recall, precision, and F1-score: recall and precision increase by up to 24\% and 29\%, respectively, with a corresponding 31\% gain in F1-score. This effectively breaks the long-standing trade-off between coverage and accuracy typical of single-pass LLM generation, while also mitigating generation truncation caused by length limitations. At the same time, we emphasize that EVE exhibits performance saturation due to the inherent ambiguity of natural language, reflecting fundamental limits of language-based reasoning.

cs.LG↗

arXiv2025

HopCast: Calibration of Autoregressive Dynamics Models

Deep learning models are often trained to approximate dynamical systems that can be modeled using differential equations. Many of these models are optimized to predict one step ahead; such approaches produce calibrated one-step predictions if the predictive model can quantify uncertainty, such as Deep Ensembles. At inference time, multi-step predictions are generated via autoregression, which needs a sound uncertainty propagation method to produce calibrated multi-step predictions. This work introduces an alternative Predictor-Corrector approach named \hop{} that uses Modern Hopfield Networks (MHN) to learn the errors of a deterministic Predictor that approximates the dynamical system. The Corrector predicts a set of errors for the Predictor's output based on a context state at any timestep during autoregression. The set of errors creates sharper and well-calibrated prediction intervals with higher predictive accuracy compared to baselines without uncertainty propagation. The calibration and prediction performances are evaluated across a set of dynamical systems. This work is also the first to benchmark existing uncertainty propagation methods based on calibration errors.

cs.LG↗

arXiv2025

Certified Learning-Enabled Noise-Aware Motion Planning for Urban Air Mobility

Urban Air Mobility (UAM) has emerged as a promising solution to alleviate urban congestion and transportation challenges. Nevertheless, the noise generated by eVTOL aircrafts poses a significant barrier to public acceptance and regulatory approval, potentially limiting the operational scope and scalability of UAM systems. Hence, the successful adoption of UAM systems hinges on the ability to predict generated noise levels, and further develop motion planning strategies that comply with community-level noise regulations while maintaining operational efficiency. To this end, this paper proposes a novel noise-aware motion planning framework for UAM systems that ensures compliance with noise regulations. We first develop a certifiable neural network model to accurately predict eVTOL noise propagation patterns in urban environments, providing provable bounds on its correctness. To achieve a desired level of accuracy, we propose an active sampling strategy to efficiently build the dataset used to train and test the noise model. Next, we develop a noise-aware motion planning algorithm that utilizes the noise model to generate eVTOL trajectories that guarantee compliance with community noise regulations. The algorithm exploits the monotonic structure of the noise model to efficiently sample the configuration space, ensuring that the generated trajectories are both noise-compliant and operationally efficient. We demonstrate the effectiveness of the proposed framework through a number of experiments for Vahana eVTOLs. The results show that the framework can generate noise-compliant flight plans for a fleet of eVTOLs that adhere to community noise regulations while optimizing operational efficiency.

eess.SY↗

arXiv2025

Can Pretrained Vision-Language Embeddings Alone Guide Robot Navigation?

Foundation models have revolutionized robotics by providing rich semantic representations without task-specific training. While many approaches integrate pretrained vision-language models (VLMs) with specialized navigation architectures, the fundamental question remains: can these pretrained embeddings alone successfully guide navigation without additional fine-tuning or specialized modules? We present a minimalist framework that decouples this question by training a behavior cloning policy directly on frozen vision-language embeddings from demonstrations collected by a privileged expert. Our approach achieves a 74% success rate in navigation to language-specified targets, compared to 100% for the state-aware expert, though requiring 3.2 times more steps on average. This performance gap reveals that pretrained embeddings effectively support basic language grounding but struggle with long-horizon planning and spatial reasoning. By providing this empirical baseline, we highlight both the capabilities and limitations of using foundation models as drop-in representations for embodied tasks, offering critical insights for robotics researchers facing practical design tradeoffs between system complexity and performance in resource-constrained scenarios. Our code is available at https://github.com/oadamharoon/text2nav

cs.RO↗

arXiv2025

Latent Safety-Constrained Policy Approach for Safe Offline Reinforcement Learning

In safe offline reinforcement learning (RL), the objective is to develop a policy that maximizes cumulative rewards while strictly adhering to safety constraints, utilizing only offline data. Traditional methods often face difficulties in balancing these constraints, leading to either diminished performance or increased safety risks. We address these issues with a novel approach that begins by learning a conservatively safe policy through the use of Conditional Variational Autoencoders, which model the latent safety constraints. Subsequently, we frame this as a Constrained Reward-Return Maximization problem, wherein the policy aims to optimize rewards while complying with the inferred latent safety constraints. This is achieved by training an encoder with a reward-Advantage Weighted Regression objective within the latent constraint space. Our methodology is supported by theoretical analysis, including bounds on policy performance and sample complexity. Extensive empirical evaluation on benchmark datasets, including challenging autonomous driving scenarios, demonstrates that our approach not only maintains safety compliance but also excels in cumulative reward optimization, surpassing existing methods. Additional visualizations provide further insights into the effectiveness and underlying mechanisms of our approach.

cs.LG↗

arXiv2025

Modular Diffusion Policy Training: Decoupling and Recombining Guidance and Diffusion for Offline RL

Classifier free guidance has shown strong potential in diffusion-based reinforcement learning. However, existing methods rely on joint training of the guidance module and the diffusion model, which can be suboptimal during the early stages when the guidance is inaccurate and provides noisy learning signals. In offline RL, guidance depends solely on offline data: observations, actions, and rewards, and is independent of the policy module's behavior, suggesting that joint training is not required. This paper proposes modular training methods that decouple the guidance module from the diffusion model, based on three key findings: Guidance Necessity: We explore how the effectiveness of guidance varies with the training stage and algorithm choice, uncovering the roles of guidance and diffusion. A lack of good guidance in the early stage presents an opportunity for optimization. Guidance-First Diffusion Training: We introduce a method where the guidance module is first trained independently as a value estimator, then frozen to guide the diffusion model using classifier-free reward guidance. This modularization reduces memory usage, improves computational efficiency, and enhances both sample efficiency and final performance. Cross-Module Transferability: Applying two independently trained guidance models, one during training and the other during inference, can significantly reduce normalized score variance (e.g., reducing IQR by 86%). We show that guidance modules trained with one algorithm (e.g., IDQL) can be directly reused with another (e.g., DQL), with no additional training required, demonstrating baseline-level performance as well as strong modularity and transferability. We provide theoretical justification and empirical validation on bullet D4RL benchmarks. Our findings suggest a new paradigm for offline RL: modular, reusable, and composable training pipelines.

cs.LG↗

arXiv2025

Correct-by-construction requirement decomposition

In systems engineering, accurately decomposing requirements is crucial for creating well-defined and manageable system components, particularly in safety-critical domains. Despite the critical need, rigorous, top-down methodologies for effectively breaking down complex requirements into precise, actionable sub-requirements are scarce, especially compared to the wealth of bottom-up verification techniques. Addressing this gap, we introduce a formal decomposition for contract-based design that guarantees the correctness of decomposed requirements if specific conditions are met. Our (semi-)automated methodology augments contract-based design with reachability analysis and constraint programming to systematically identify, verify, and validate sub-requirements representable by continuous bounded sets -- continuous relations between real-valued inputs and outputs. We demonstrate the efficacy and practicality of a correct-by-construction approach through a comprehensive case study on a cruise control system, highlighting how our methodology improves the interpretability, tractability, and verifiability of system requirements.

eess.SY↗

arXiv2025

ShieldNN: A Provably Safe NN Filter for Unsafe NN Controllers

In this paper, we develop a novel closed-form Control Barrier Function (CBF) and associated controller shield for the Kinematic Bicycle Model (KBM) with respect to obstacle avoidance. The proposed CBF and shield -- designed by an algorithm we call ShieldNN -- provide two crucial advantages over existing methodologies. First, ShieldNN considers steering and velocity constraints directly with the non-affine KBM dynamics; this is in contrast to more general methods, which typically consider only affine dynamics and do not guarantee invariance properties under control constraints. Second, ShieldNN provides a closed-form set of safe controls for each state unlike more general methods, which typically rely on optimization algorithms to generate a single instantaneous for each state. Together, these advantages make ShieldNN uniquely suited as an efficient Multi-Obstacle Safe Actions (i.e. multiple-barrier-function shielding) during training time of a Reinforcement Learning (RL) enabled Neural Network controller. We show via experiments that ShieldNN dramatically increases the completion rate of RL training episodes in the presence of multiple obstacles, thus establishing the value of ShieldNN in training RL-based controllers.

cs.RO↗

arXiv2024

FAWAC: Feasibility Informed Advantage Weighted Regression for Persistent Safety in Offline Reinforcement Learning

Safe offline reinforcement learning aims to learn policies that maximize cumulative rewards while adhering to safety constraints, using only offline data for training. A key challenge is balancing safety and performance, particularly when the policy encounters out-of-distribution (OOD) states and actions, which can lead to safety violations or overly conservative behavior during deployment. To address these challenges, we introduce Feasibility Informed Advantage Weighted Actor-Critic (FAWAC), a method that prioritizes persistent safety in constrained Markov decision processes (CMDPs). FAWAC formulates policy optimization with feasibility conditions derived specifically for offline datasets, enabling safe policy updates in non-parametric policy space, followed by projection into parametric space for constrained actor training. By incorporating a cost-advantage term into Advantage Weighted Regression (AWR), FAWAC ensures that the safety constraints are respected while maximizing performance. Additionally, we propose a strategy to address a more challenging class of problems that involves tempting datasets where trajectories are predominantly high-rewarded but unsafe. Empirical evaluations on standard benchmarks demonstrate that FAWAC achieves strong results, effectively balancing safety and performance in learning policies from the static datasets.

cs.LG↗

arXiv2024

Solving Offline Reinforcement Learning with Decision Tree Regression

This study presents a novel approach to addressing offline reinforcement learning (RL) problems by reframing them as regression tasks that can be effectively solved using Decision Trees. Mainly, we introduce two distinct frameworks: return-conditioned and return-weighted decision tree policies (RCDTP and RWDTP), both of which achieve notable speed in agent training as well as inference, with training typically lasting less than a few minutes. Despite the simplification inherent in this reformulated approach to offline RL, our agents demonstrate performance that is at least on par with the established methods. We evaluate our methods on D4RL datasets for locomotion and manipulation, as well as other robotic tasks involving wheeled and flying robots. Additionally, we assess performance in delayed/sparse reward scenarios and highlight the explainability of these policies through action distribution and feature importance.

cs.LG↗

arXiv2024

F1tenth Autonomous Racing With Offline Reinforcement Learning Methods

Autonomous racing serves as a critical platform for evaluating automated driving systems and enhancing vehicle mobility intelligence. This work investigates offline reinforcement learning methods to train agents within the dynamic F1tenth racing environment. The study begins by exploring the challenges of online training in the Austria race track environment, where agents consistently fail to complete the laps. Consequently, this research pivots towards an offline strategy, leveraging `expert' demonstration dataset to facilitate agent training. A waypoint-based suboptimal controller is developed to gather data with successful lap episodes. This data is then employed to train offline learning-based algorithms, with a subsequent analysis of the agents' cross-track performance, evaluating their zero-shot transferability from seen to unseen scenarios and their capacity to adapt to changes in environment dynamics. Beyond mere algorithm benchmarking in autonomous racing scenarios, this study also introduces and describes the machinery of our return-conditioned decision tree-based policy, comparing its performance with methods that employ fully connected neural networks, Transformers, and Diffusion Policies and highlighting some insights into method selection for training autonomous agents in driving interactions.

cs.RO↗

arXiv2024

Towards Robust Car Following Dynamics Modeling via Blackbox Models: Methodology, Analysis, and Recommendations

The selection of the target variable is important while learning parameters of the classical car following models like GIPPS, IDM, etc. There is a vast body of literature on which target variable is optimal for classical car following models, but there is no study that empirically evaluates the selection of optimal target variables for black-box models, such as LSTM, etc. The black-box models, like LSTM and Gaussian Process (GP) are increasingly being used to model car following behavior without wise selection of target variables. The current work tests different target variables, like acceleration, velocity, and headway, for three black-box models, i.e., GP, LSTM, and Kernel Ridge Regression. These models have different objective functions and work in different vector spaces, e.g., GP works in function space, and LSTM works in parameter space. The experiments show that the optimal target variable recommendations for black-box models differ from classical car following models depending on the objective function and the vector space. It is worth mentioning that models and datasets used during evaluation are diverse in nature: the datasets contained both automated and human-driven vehicle trajectories; the black-box models belong to both parametric and non-parametric classes of models. This diversity is important during the analysis of variance, wherein we try to find the interaction between datasets, models, and target variables. It is shown that the models and target variables interact and recommended target variables don't depend on the dataset under consideration.

cs.LG↗

arXiv2022

Characterizing the Identity of Model-based Safety Assessment: A Systematic Analysis

Model-based safety assessment has been one of the leading research thrusts of the System Safety Engineering community for over two decades. However, there is still a lack of consensus on what MBSA is. The ambiguity in the identity of MBSA impedes the advancement of MBSA as an active research area. For this reason, this paper aims to investigate the identity of MBSA to help achieve a consensus across the community. Towards this end, we first reason about the core activities that an MBSA approach must conduct. Second, we characterize the core patterns in which the core activities must be conducted for an approach to be considered MBSA. Finally, a recently published MBSA paper is reviewed to test the effectiveness of our characterization of MBSA.

cs.SE↗

arXiv2021

Cyberphysical Security Through Resiliency: A Systems-centric Approach

Cyber-physical systems (CPS) are often defended in the same manner as information technology (IT) systems -- by using perimeter security. Multiple factors make such defenses insufficient for CPS. Resiliency shows potential in overcoming these shortfalls. Techniques for achieving resilience exist; however, methods and theory for evaluating resilience in CPS are lacking. We argue that such methods and theory should assist stakeholders in deciding where and how to apply design patterns for resilience. Such a problem potentially involves tradeoffs between different objectives and criteria, and such decisions need to be driven by traceable, defensible, repeatable engineering evidence. Multi-criteria resiliency problems require a system-oriented approach that evaluates systems in the presence of threats as well as potential design solutions once vulnerabilities have been identified. We present a systems-oriented view of cyber-physical security, termed Mission Aware, that is based on a holistic understanding of mission goals, system dynamics, and risk.

cs.CR↗

arXiv2021

Gaussian Process-based Model Predictive Controller for Connected Vehicles with Uncertain Wireless Channel

In this paper, we present a data-driven Model Predictive Controller that leverages a Gaussian Process to generate optimal motion policies for connected autonomous vehicles in regions with uncertainty in the wireless channel. The communication channel between the vehicles of a platoon can be easily influenced by numerous factors, e.g. the surrounding environment, and the relative states of the connected vehicles, etc. In addition, the trajectories of the vehicles depend significantly on the motion policies of the preceding vehicle shared via the wireless channel and any delay can impact the safety and optimality of its performance. In the presented algorithm, Gaussian Process learns the wireless channel model and is involved in the Model Predictive Controller to generate a control sequence that not only minimizes the conventional motion costs, but also minimizes the estimated delay of the wireless channel in the future. This results in a farsighted controller that maximizes the amount of transferred information beyond the controller's time horizon, which in turn guarantees the safety and optimality of the generated trajectories in the future. To decrease computational cost, the algorithm finds the reachable set from the current state and focuses on that region to minimize the size of the kernel matrix and related calculations. In addition, we present an efficient recursive approach to decrease the time complexity of developing the data-driven model and involving it in Model Predictive Control. We demonstrate the capability of the presented algorithm in a simulated scenario.

eess.SY↗

arXiv2021

SCAN: A Spatial Context Attentive Network for Joint Multi-Agent Intent Prediction

Safe navigation of autonomous agents in human centric environments requires the ability to understand and predict motion of neighboring pedestrians. However, predicting pedestrian intent is a complex problem. Pedestrian motion is governed by complex social navigation norms, is dependent on neighbors' trajectories, and is multimodal in nature. In this work, we propose SCAN, a Spatial Context Attentive Network that can jointly predict socially-acceptable multiple future trajectories for all pedestrians in a scene. SCAN encodes the influence of spatially close neighbors using a novel spatial attention mechanism in a manner that relies on fewer assumptions, is parameter efficient, and is more interpretable compared to state-of-the-art spatial attention approaches. Through experiments on several datasets we demonstrate that our approach can also quantitatively outperform state of the art trajectory prediction methods in terms of accuracy of predicted intent.

cs.CV↗

arXiv2021

DMPC: A Data-and Model-Driven Approach to Predictive Control

This work presents DMPC (Data-and Model-Driven Predictive Control) to solve control problems in which some of the constraints or parts of the objective function are known, while others are entirely unknown to the controller. It is assumed that there is an exogenous ``black box'' system, e.g. a machine learning technique, that predicts the value of the unknown functions for a given trajectory. DMPC (1) provides an approach to merge both the model-based and black-box systems; (2) can cope with very little data and is sample efficient, building its solutions based on recently generated trajectories; and (3) improves its cost in each iteration until converging to an optimal trajectory, typically needing only a few trials even for nonlinear dynamics and objectives. Theoretical analysis of the algorithm is presented, proving that the quality of the trajectory does not worsen with each new iteration, as well as providing bounds on the complexity. We apply the DMPC algorithm to the motion planning of an autonomous vehicle with nonlinear dynamics.

eess.SY↗

arXiv2019

A Spatially and Temporally Attentive Joint Trajectory Prediction Framework for Modeling Vessel Intent

Ships, or vessels, often sail in and out of cluttered environments over the course of their trajectories. Safe navigation in such cluttered scenarios requires an accurate estimation of the intent of neighboring vessels and their effect on the self and vice-versa well into the future. In manned vessels, this is achieved by constant communication between people on board, nautical experience, and audio and visual signals. In this paper we propose a deep neural network based architecture to predict intent of neighboring vessels into the future for an unmanned vessel solely based on positional data.

eess.SP↗