Publications
My research develops execution-grounded AI for scientific and computational systems, with AI for Science and AI for Systems tied together by validation from compilation, profiling, runtime behavior, biological structure, measurement, and scientific computing. This page groups the publication record into execution-grounded AI for code and systems; AI for biology and RNA editing; AI for materials, microscopy, and scientific imaging; and scientific computing, HPC, and distributed foundations.
Publication map
Execution-grounded AI for code and systems
Research on AI systems that generate, translate, and improve parallel code under compiler, runtime, and performance feedback.
Journal and main-conference papers
-
ParaCodex: A Profiling-Guided Autonomous Coding Agent for Reliable Parallel Code Generation and Translation
-
PragFormer: Data-driven Parallel Source Code Classification with Transformers
-
UniPar: A Unified LLM-Based Framework for Parallel and Accelerated Code Translation in HPC
-
MonoCoder: Domain-Specific Code Language Model for HPC Codes and Tasks
-
OMPGPT: A Generative Pre-trained Transformer Model for OpenMP
-
Quantifying OpenMP: Statistical Insights into Usage and Adoption
Workshops and co-located papers
-
Manual, Translated, or LLM-Generated? Comparing Paths to OpenMP Target Offload for LULESH
-
ParCoder: Harnessing Transformer Models for Code Translation to Shared-Memory Computing Paradigms
-
From OpenACC to OpenMP5 GPU Offloading: Performance Evaluation on NAS Parallel Benchmarks
-
Can large language models predict parallel code performance?
-
OMPar: Automatic Parallelization with AI-Driven Source-to-Source Compilation
-
MPIrigen: MPI Code Generation through Domain-Specific Language Models
-
MPI-rical: Data-Driven MPI Distributed Parallelism Assistance with Transformers
-
Advising OpenMP Parallelization via a Graph-Based Approach with Transformers
-
Portability and Scalability of OpenMP Offloading on State-of-the-art Accelerators
-
ComPar: optimized multi-compiler for automatic OpenMP S2S parallelization
Submitted preprints and extended abstracts
-
Latent Reasoning Guidance for Parallel Code Translation
-
ParBench: A Benchmark for Reliable Evaluation of LLM Parallel Code Translation
-
Counting Without Running: Evaluating LLMs' Reasoning About Code Complexity
-
The Landscape and Challenges of HPC Research and LLMs
-
Scope is all you need: Transforming LLMs for HPC Code
-
CompCodeVet: A Compiler-guided Validation and Enhancement Approach for Code Dataset
-
Learning to Parallelize in a Shared-Memory Environment with Transformers
AI for biology and RNA editing
Research on language models and graph learning for RNA editability prediction and ADAR biology.
Journal and main-conference papers
-
ADAR-GPT: A continually fine-tuned language model for predicting A-to-I RNA editing sites
Workshops and co-located papers
-
Structure-aware Graph Learning Predicts RNA Editability Across Tissues and Species
-
Leveraging GPT Continual Fine-Tuning for Improved RNA Editing Site Prediction
-
Detection of RNA Editing Sites by GPT Fine-tuning
Submitted preprints
-
Structure-aware Graph Learning Predicts RNA Editability Across Tissues and Species
AI for materials, microscopy, and scientific imaging
Research on vision models for materials and microscopy where outputs must support quantitative measurement, not just visual plausibility.
Journal and main-conference papers
-
Context-aware learning for reliable grain-boundary analysis in quantitative metallography
-
Decoupling Shape and Texture in SAM-2 via Controlled Texture Replacement
-
An End-to-End Computer Vision Methodology for Quantitative Metallography
-
ChangeChip: A Reference-Based Unsupervised Change Detection for PCB Defect Detection
Workshops and co-located papers
-
Learning 4D Material-Interface Dynamics From Few X-RAY Projections
-
Evaluating Diffusion-based Super-Resolution for Trustworthy Quantitative Metallography
-
Avoiding Post-Processing with Context: Texture Boundary Detection in Metallography
-
ClariPhy: Physics-Informed Image Deblurring with Transformers for Hydrodynamic Instability Analysis
-
Explainable Deep Multi-View Methodology for HEDP Foam Quality Assessment
-
Universal Semantic-less Texture Boundary Detection for Microscopy (and Metallography)
-
Determining HEDP Foams' Quality with Multi-View Deep Learning Classification
-
Complete Computer Vision Deep Learning Methodology for Investigating Hydrodynamic Instabilities
-
MLography: An Automated Quantitative Metallography Model for Impurities Anomaly Detection using Novel Data Mining and Deep Learning Approach
Submitted preprints
-
Where Does Texture Evidence Live in SAM? Features, Proposal Masks, and Texture Segmentation
-
Sub-Semantic Image Segmentation
Scientific computing, HPC, and distributed foundations
Research on scientific-computing and distributed-systems foundations for execution, recovery, simulation, and applied validation.
Journal and main-conference papers
-
Reclassification and weighting of multiple causes of death: US death certificates 2003-2023
-
Reactor Optimization Benchmark by Reinforcement Learning
-
Distributed Backup Placement
-
Borehole Breakout Modeling in Arkose and Granite Rocks
-
Source-to-Source Parallelization Compilers for Scientific Shared-Memory Multi-core and Accelerated Multiprocessing: Analysis, Pitfalls, Enhancement and Potential
-
GraphiX: A Fast Human-Computer Interaction Symmetric Multiprocessing Parallel Scientific Visualization Tool
-
Acceleration of Hydro Poro-elastic Damage Simulation in a Shared-Memory Environment
-
TradeMarker - Artificial Intelligence-Based Trademarks Similarity Search Engine
-
Distributed Fault-Tolerant Backup-Placement in Overloaded Wireless Sensor Networks
-
CalCul: A Python-based Workspace for High-Performance Parameter-Survey in Scientific Legacy Codes
-
Adaptive Distributed Hierarchical Sensing Algorithm for Reduction of Wireless Sensor Network Cluster-Heads Energy Consumption
-
Load-Balancing Adaptive Clustering Refinement Algorithm for Maximization of Wireless Sensor Network Clusters Lifetime
-
The Looking-Glass System: A Unidirectional System for Secure Data Transfer using an Optic Medium
Workshops and co-located papers
-
SCALSALE: A Scalable MPI+X Benchmark for Lagrangian Hydrodynamics on Heterogeneous Systems
-
CXL Memory as Persistent Memory for Disaggregated HPC: A Practical Approach
-
Recovery of Distributed Iterative Solvers for Linear Systems Using Non-Volatile RAM
-
Assessing the Use Cases of Persistent Memory in High-Performance Scientific Computing
-
Backup Placement in WSNs in the Network Management Distributed Setting
-
Optimized Memoryless Fair-Share HPC Resources Scheduling using Transparent Checkpoint-Restart Preemption
-
Distributed Backup K-Placement and Applications to Virtual Memory in Wireless Networks
-
Distributed Backup Placement in One Round and Its Applications to Maximum Matching Approximation and Self-Stabilization
-
Fast Distributed Backup Placement in Sparse and Dense Networks
-
Memory-Aware Management for Heterogeneous Main Memory Complex using an Optimization of the Aging Paging Algorithm
Extended abstracts
-
Distributed Fault-Tolerant Backup-Placement in Overloaded Wireless Sensor Networks
-
Memory-Aware Management for Heterogeneous Main Memory Complex Using an Optimization of the Aging Paging Algorithm
External records
External indexes and institutional pages remain linked here for live citation data, profile context, and publication pages outside this static site.
Google Scholar
Live citation record and broad publication index.
DBLP
Computer-science bibliography and conference spine.
ResearchGate
Publication pages and full-text availability where provided.
Stanford page
Institutional profile linked from this site.
Technion page
Institutional page linked from this site.
