15 citations · 53 across the 26 of their papers we have counts for
26 papers
Foundation Model-Assisted Full Waveform Inversion
Mustafa Alfarhan, Matteo Ravasi, Fuqiang Chen +2
Full waveform inversion (FWI) can recover high-resolution subsurface velocity models. Conventional waveform-difference objectives, however, are vulnerable to cycle skipping when th…
Whittle-Matérn Fields with Variable Smoothness
Hamza Ruzayqat, Wenyu Lei, David Bolin +2
We introduce and analyze a nonlocal generalization of Whittle--Matérn Gaussian fields in which the smoothness parameter varies in space through the fractional order, $s=s(x)\in[\un…
A Deep Learning-based time shift objective function for Full Waveform Inversion
Mustafa Alfarhan, Fuqiang Chen, George Turkiyyah +3
Full Waveform Inversion (FWI) is a powerful technique for estimating high-resolution subsurface velocity models by minimizing the discrepancy between modeled and observed seismic d…
ALARB: An Arabic Legal Argument Reasoning Benchmark
Harethah Abu Shairah, Somayah AlHarbi, Abdulaziz AlHussein +5
We introduce ALARB, a dataset and suite of tasks designed to evaluate the reasoning capabilities of large language models (LLMs) within the Arabic legal domain. While existing Arab…
Linear Complexity Direct Solver for Fine-Grained Parallel Architectures
Wajih Boukaram, David Keyes, Sherry Li +2
We present factorization and solution phases for a new linear complexity direct solver designed for concurrent batch operations on fine-grained parallel architectures, for matrices…
Cooperative Grasping for Collective Object Transport in Constrained Environments
David Alvear, George Turkiyyah, Shinkyu Park
We propose a novel framework for decision-making in cooperative grasping for two-robot object transport in constrained environments. The core of the framework is a Conditional Embe…
Turning the Spell Around: Lightweight Alignment Amplification via Rank-One Safety Injection
Harethah Abu Shairah, Hasan Abed Al Kader Hammoud, George Turkiyyah +1
Safety alignment in Large Language Models (LLMs) often involves mediating internal representations to refuse harmful requests. Recent research has demonstrated that these safety me…
Synthetic Geology: Structural Geology Meets Deep Learning
Simon Ghyselincks, Valeriia Okhmak, Stefano Zampini +3
Reconstructing the structural geology and mineral composition of the first few kilometers of the Earth's subsurface from sparse or indirect surface observations remains a long-stan…
An Embarrassingly Simple Defense Against LLM Abliteration Attacks
Harethah Abu Shairah, Hasan Abed Al Kader Hammoud, Bernard Ghanem +1
Large language models (LLMs) are typically aligned to refuse harmful instructions through safety fine-tuning. A recent attack, termed abliteration, identifies and suppresses the si…
H2-MG: A multigrid method for hierarchical rank structured matrices
Daria Sushnikova, George Turkiyyah, Edmond Chow +1
This paper presents a new fast iterative solver for large systems involving kernel matrices. Advantageous aspects of H2 matrix approximations and the multigrid method are hybridize…
ArabLegalEval: A Multitask Benchmark for Assessing Arabic Legal Knowledge in Large Language Models
Faris Hijazi, Somayah AlHarbi, Abdulaziz AlHussein +5
The rapid advancements in Large Language Models (LLMs) have led to significant improvements in various natural language processing tasks. However, the evaluation of LLMs' legal kno…
Finite element discretizations for variable-order fractional diffusion problems
Wenyu Lei, George Turkiyyah, Omar Knio
We present a finite element scheme for fractional diffusion problems with varying diffusivity and fractional order. We consider a symmetric integral form of these nonlocal equation…
H2Opus: A distributed-memory multi-GPU software package for non-local operators
Stefano Zampini, Wajih Boukaram, George Turkiyyah +2
Hierarchical -matrices are asymptotically optimal representations for the discretizations of non-local operators such as those arising in integral equations or from…
Space-Fractional Diffusion with Variable Order and Diffusivity: Discretization and Direct Solution Strategies
Hasnaa Alzahrani, George Turkiyyah, Omar Knio +1
We consider the multidimensional space-fractional diffusion equations with spatially varying diffusivity and fractional order. Significant computational challenges are encountered…
H2OPUS-TLR: High Performance Tile Low Rank Symmetric Factorizations using Adaptive Randomized Approximation
Wajih Boukaram, Stefano Zampini, George Turkiyyah +1
Tile low rank representations of dense matrices partition them into blocks of roughly uniform size, where each off-diagonal tile is compressed and stored as its own low rank factor…
Hierarchical Matrix Approximations of Hessians Arising in Inverse Problems Governed by PDEs
Ilona Ambartsumyan, Wajih Boukaram, Tan Bui-Thanh +5
Hessian operators arising in inverse problems governed by partial differential equations (PDEs) play a critical role in delivering efficient, dimension-independent convergence for…
Exploiting Low Rank Covariance Structures for Computing High-Dimensional Normal and Student- Probabilities
Jian Cao, Marc G. Genton, David E. Keyes +1
We present a preconditioned Monte Carlo method for computing high-dimensional multivariate normal and Student- probabilities arising in spatial statistics. The approach combines…
Hierarchical Matrix Operations on GPUs: Matrix-Vector Multiplication and Compression
Wajih Halim Boukaram, George Turkiyyah, David E. Keyes
Hierarchical matrices are space and time efficient representations of dense matrices that exploit the low rank structure of matrix blocks at different levels of granularity. The hi…
O(N) Hierarchical algorithm for computing the expectations of truncated multi-variate normal distributions in N dimensions
Jingfang Huang, Fuhui Fang, George Turkiyyah +3
In this paper, we study the -dimensional integral representing the expectation of a function where is the t…
CUDACLAW: A high-performance programmable GPU framework for the solution of hyperbolic PDEs
H. Gorune Ohannessian, George Turkiyyah, Aron Ahmadia +1
We present cudaclaw, a CUDA-based high performance data-parallel framework for the solution of multidimensional hyperbolic partial differential equation (PDE) systems, equations de…
Parallel accelerated cyclic reduction preconditioner for three-dimensional elliptic PDEs with variable coefficients
Gustavo Chávez, George Turkiyyah, Stefano Zampini +1
We present a robust and scalable preconditioner for the solution of large-scale linear systems that arise from the discretization of elliptic PDEs amenable to rank compression. The…
Batched QR and SVD Algorithms on GPUs with Applications in Hierarchical Matrix Compression
Wajih Halim Boukaram, George Turkiyyah, Hatem Ltaief +1
We present high performance implementations of the QR and the singular value decomposition of a batch of small matrices hosted on the GPU with applications in the compression of hi…
Accelerated Cyclic Reduction: A Distributed-Memory Fast Solver for Structured Linear Systems
Gustavo Chávez, George Turkiyyah, Stefano Zampini +2
We present Accelerated Cyclic Reduction (ACR), a distributed-memory fast direct solver for rank-compressible block tridiagonal linear systems arising from the discretization of ell…
A Direct Elliptic Solver Based on Hierarchically Low-rank Schur Complements
Gustavo Chávez, George Turkiyyah, David Keyes
A parallel fast direct solver for rank-compressible block tridiagonal linear systems is presented. Algorithmic synergies between Cyclic Reduction and Hierarchical matrix arithmetic…