The field of programming and design is experiencing a significant shift with the integration of large language models (LLMs). Recent developments indicate a move towards leveraging LLMs to improve the efficiency and effectiveness of various tasks, such as code generation, compiler optimization, and design processes. LLMs are being explored for their potential to automate and enhance tasks that were previously manual or required specialized expertise. This trend is expected to continue, with LLMs playing an increasingly important role in shaping the future of programming and design. Noteworthy papers in this area include CompilerGPT, which demonstrates the potential of LLMs in analyzing and acting on compiler optimization reports, and CUDA-LLM, which showcases the ability of LLMs to generate efficient CUDA kernels. Execution Guided Line-by-Line Code Generation is also a notable contribution, as it presents a novel approach to neural code generation that incorporates real-time execution signals into the language model generation process.
Large Language Models in Programming and Design
Sources
Conversational Interfaces for Parametric Conceptual Architectural Design: Integrating Mixed Reality with LLM-driven Interaction
CompilerGPT: Leveraging Large Language Models for Analyzing and Acting on Compiler Optimization Reports
CUDA-LLM: LLMs Can Write Efficient CUDA Kernels
Multi-GPU Acceleration of PALABOS Fluid Solver using C++ Standard Parallelism
From Tool Calling to Symbolic Thinking: LLMs in a Persistent Lisp Metaprogramming Loop
HPCTransCompile: An AI Compiler Generated Dataset for High-Performance CUDA Transpilation and LLM Preliminary Exploration
Execution Guided Line-by-Line Code Generation