Source author record

Dan Negrut

Dan Negrut appears in the imported research catalog. Authorship, coauthor and topic links are available while profile ownership is still unclaimed.

ResearcherUnclaimed source record

Artificial Intelligence Robotics Computational Engineering, Finance, and Science Cryptography and Security Distributed, Parallel, and Cluster Computing Machine Learning Mathematical Software Multiagent Systems Numerical Analysis

Catalog footprint

What is connected

7works

9topics

4close collaborators

Actions

Connect this record

Open graph Browse works

Inspect adjacent papers, topics, institutions and collaborators without losing the researcher page.

Building this map preview

BZPEER is loading the nearby papers, people, topics and institutions for this page.

preprint2026arXiv

Active Learning for Communication Structure Optimization in LLM-Based Multi-Agent Systems

Optimizing the communication structure of large language model based multi-agent systems (LLM-MAS) has been shown to improve downstream performance and reduce token usage. Existing methods typically rely on randomly sampled training tasks. However, tasks may differ substantially in difficulty and domain, and thus they are not equally informative for updating communication structure, making optimization under limited training budgets often unstable and highly sensitive to the particular training set. To actively identify the most valuable tasks for communication-structure optimization, we propose an ensemble-based information-theoretic task selection framework. The proposed method estimates task informativeness by how much a candidate task changes the distribution over graph parameters, using ensemble Kalman inversion as an efficient and derivative-free approximation of the corresponding Bayesian update. The resulting estimator is especially suitable for black-box and noisy multi-agent systems. To enhance scalability, we construct a compact candidate pool through embedding-based representative selection and combine the informative selection with surrogate modeling and batch Thompson sampling. We validate our method in both benign settings and settings with agent attacks, demonstrating its effectiveness for communication-structure optimization under constrained computational budgets.

preprint2026arXiv

Coding Agent Is Good As World Simulator

World models have emerged as a powerful paradigm for building interactive simulation environments, with recent video-based approaches demonstrating impressive progress in generating visually plausible dynamics. However, because these models typically infer dynamics from video and represent them in latent states, they do not explicitly enforce physical constraints. As a result, the generated video rollouts are not physically plausible, exhibiting unstable contacts, distorted shapes, or inconsistent motion. In this paper, we present an agentic framework constructing physics-based world models through executable simulation code. The framework coordinates planning, code generation, visual review, and physics analysis agents. The planning agent converts the natural language prompt into a structured scene plan, the code agent implements it as executable simulation code, and the visual review agent provide visual feedback while the physics analysis agent checks physical consistency. The code is iteratively revised based on the feedback until the simulation matches the prompt reqirements and physical constraints. Experimental results show that our framework outperforms advanced video-based models in physical accuracy, instruction fidelity and visual quality, which could be applied to various scenarios including driving simulation and embodied robot tasks.

preprint2022arXiv

A performance contextualization approach to validating camera models for robot simulation

The focus of this contribution is on camera simulation as it comes into play in simulating autonomous robots for their virtual prototyping. We propose a camera model validation methodology based on the performance of a perception algorithm and the context in which the performance is measured. This approach is different than traditional validation of synthetic images, which is often done at a pixel or feature level, and tends to require matching pairs of synthetic and real images. Due to the high cost and constraints of acquiring paired images, the proposed approach is based on datasets that are not necessarily paired. Within a real and a simulated dataset, A and B, respectively, we find subsets Ac and Bc of similar content and judge, statistically, the perception algorithm's response to these similar subsets. This validation approach obtains a statistical measure of performance similarity, as well as a measure of similarity between the content of A and B. The methodology is demonstrated using images generated with Chrono::Sensor and a scaled autonomous vehicle, using an object detector as the perception algorithm. The results demonstrate the ability to quantify (i) differences between simulated and real data; (ii) the propensity of training methods to mitigate the sim-to-real gap; and (iii) the context overlap between two datasets.

preprint2022arXiv

Evaluating a GAN for enhancing camera simulation for robotics

Given the versatility of generative adversarial networks (GANs), we seek to understand the benefits gained from using an existing GAN to enhance simulated images and reduce the sim-to-real gap. We conduct an analysis in the context of simulating robot performance and image-based perception. Specifically, we quantify the GAN's ability to reduce the sim-to-real difference in image perception in robotics. Using semantic segmentation, we analyze the sim-to-real difference in training and testing, using nominal and enhanced simulation of a city environment. As a secondary application, we consider use of the GAN in enhancing an indoor environment. For this application, object detection is used to analyze the enhancement in training and testing. The results presented quantify the reduction in the sim-to-real gap when using the GAN, and illustrate the benefits of its use.

preprint2021arXiv

CryptoEmu: An Instruction Set Emulator for Computation Over Ciphers

Fully homomorphic encryption (FHE) allows computations over encrypted data. This technique makes privacy-preserving cloud computing a reality. Users can send their encrypted sensitive data to a cloud server, get encrypted results returned and decrypt them, without worrying about data breaches. This project report presents a homomorphic instruction set emulator, CryptoEmu, that enables fully homomorphic computation over encrypted data. The software-based instruction set emulator is built upon an open-source, state-of-the-art homomorphic encryption library that supports gate-level homomorphic evaluation. The instruction set architecture supports multiple instructions that belong to the subset of ARMv8 instruction set architecture. The instruction set emulator utilizes parallel computing techniques to emulate every functional unit for minimum latency. This project report includes details on design considerations, instruction set emulator architecture, and datapath and control unit implementation. We evaluated and demonstrated the instruction set emulator's performance and scalability on a 48-core workstation. CryptoEmu has shown a significant speedup in homomorphic computation performance when compared with HELib, a state-of-the-art homomorphic encryption library.

preprint2020arXiv

Producing 3D Friction Loads by Tracking the Motion of the Contact Point on Bodies in Mutual Contact

We outline a phenomenological model to assess friction at the interface between two bodies in mutual contact. Although the approach is general, the application inspiring the approach is the Discrete Element Method. The kinematics of the friction process is expressed in terms of the relative 3D motion of the contact point on the two surfaces in mutual contact. The model produces expressions for three friction loads: slide force, roll torque, and spin torque. The cornerstone of the methodology is the process of tracking the evolution/path of the contact point on the surface of the two bodies in mutual contact. The salient attribute of the model lies with its ability to simultaneously compute, in a 3D setup, the slide, roll, and spin friction loads for smooth bodies of arbitrary geometry while accounting for both static and kinematic friction coefficients.

preprint2015arXiv

Analysis of A Splitting Approach for the Parallel Solution of Linear Systems on GPU Cards

We discuss an approach for solving sparse or dense banded linear systems ${\bf A} {\bf x} = {\bf b}$ on a Graphics Processing Unit (GPU) card. The matrix ${\bf A} \in {\mathbb{R}}^{N \times N}$ is possibly nonsymmetric and moderately large; i.e., $10000 \leq N \leq 500000$. The ${\it split\ and\ parallelize}$ (${\tt SaP}$) approach seeks to partition the matrix ${\bf A}$ into diagonal sub-blocks ${\bf A}_i$, $i=1,\ldots,P$, which are independently factored in parallel. The solution may choose to consider or to ignore the matrices that couple the diagonal sub-blocks ${\bf A}_i$. This approach, along with the Krylov subspace-based iterative method that it preconditions, are implemented in a solver called ${\tt SaP::GPU}$, which is compared in terms of efficiency with three commonly used sparse direct solvers: ${\tt PARDISO}$, ${\tt SuperLU}$, and ${\tt MUMPS}$. ${\tt SaP::GPU}$, which runs entirely on the GPU except several stages involved in preliminary row-column permutations, is robust and compares well in terms of efficiency with the aforementioned direct solvers. In a comparison against Intel's ${\tt MKL}$, ${\tt SaP::GPU}$ also fares well when used to solve dense banded systems that are close to being diagonally dominant. ${\tt SaP::GPU}$ is publicly available and distributed as open source under a permissive BSD3 license.