ALL-FEM fine-tunes LLMs on a corpus of verified FEniCS scripts and uses multi-agent workflows to automate finite element code generation, achieving 71.79% success on 39 benchmarks across elasticity, flow, and coupled problems.
URL https://arxiv.org/abs/2504.19338
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
verdicts
UNVERDICTED 2representative citing papers
Experiments with local LLMs and RAG on closed-source simulation software show promising but incomplete results, with prompt-specific information retrieval providing the clearest gains.
citing papers explorer
-
ALL-FEM: Agentic Large Language models Fine-tuned for Finite Element Methods
ALL-FEM fine-tunes LLMs on a corpus of verified FEniCS scripts and uses multi-agent workflows to automate finite element code generation, achieving 71.79% success on 39 benchmarks across elasticity, flow, and coupled problems.
-
Experiments with Large Language Models on Retrieval-Augmented Generation for Closed-Source Simulation Software
Experiments with local LLMs and RAG on closed-source simulation software show promising but incomplete results, with prompt-specific information retrieval providing the clearest gains.