News

on why reviewer precision does not translate into critique uptake in multi-agent math reasoning

on agentic CUDA-to-CSL kernel translation and optimization for wafer-scale systems

on an agentic framework for evaluating AI-generated scientific code in PETSc

on latent reasoning guidance for parallel code translation

on generating test-grounded data for AI-assisted scientific software modernization

on latent reasoning guidance for parallel code translation — now accepted at EMNLP'26 Findings

on a benchmark for reliable evaluation of LLM parallel code translation

on agent-based workflows that improve PETSc usability and productivity

on an agents-evaluating-agents framework for AI-generated scientific code

on a profiling-guided autonomous agent for parallel code, and on dialogue-based data generation for code translation

on transformer-based auto-parallelization for shared-memory paradigms

on how test-time compute constraints shape the safety and skill reasoning of large reasoning models

on a unified LLM framework for parallel and accelerated code translation in HPC

on federated inference scheduling for scientific AI models, and on test-time compute constraints in large reasoning models

on AI assistants that enhance and exploit the PETSc knowledge base

on a multi-dimensional benchmark for LLM parallel code generation

on GNN-guided context generation for zero-shot code parallelization

on Fortran-to-C++ migration via multi-turn dialogue and dual-agent integration

on unsupervised code translation for parallel programming

on a generative pre-trained transformer tailored to OpenMP

on the landscape and open challenges of LLMs for HPC research

on swift inference for text-to-video generation

on compiler-guided validation and enhancement of code datasets

on building an OpenMP Fortran-to-C++ translation dataset with LLMs

on learning to parallelize with OpenMP via heterogeneous AST representation