The field of large language models (LLMs) is rapidly advancing, with significant developments in their application to programming and scientific research. Recent studies have demonstrated the effectiveness of LLMs in solving complex problems, such as calculus and advanced computer science assignments, although their ability to provide conceptual understanding and human-like reasoning remains limited. The use of LLMs for library migration, code generation, and API testing has also shown promise, with some models achieving high accuracy and efficiency in these tasks. Furthermore, the integration of LLMs with other tools and techniques, such as static and dynamic analysis, has led to improved results in areas like RESTful API testing and code snippet generation. Noteworthy papers in this area include 'Benchmarking Large Language Models for Calculus Problem-Solving' and 'Paper2Code: Automating Code Generation from Scientific Papers in Machine Learning', which demonstrate the potential of LLMs to revolutionize programming and scientific research. Overall, the field is moving towards increased adoption of LLMs in various applications, with a focus on improving their accuracy, efficiency, and usability.
Advancements in Large Language Models for Programming and Science
Sources
Benchmarking Large Language Models for Calculus Problem-Solving: A Comparative Analysis
Using LLMs for Library Migration
Do Prompt Patterns Affect Code Quality? A First Empirical Assessment of ChatGPT-Generated Code
The Evolving Role of Programming and LLMs in the Development of Self-Driving Laboratories
Code2API: A Tool for Generating Reusable APIs from Stack Overflow Code Snippets
APIRAT: Integrating Multi-source API Knowledge for Enhanced Code Translation with LLMs
Evaluating Code Generation of LLMs in Advanced Computer Science Problems
EasyEdit2: An Easy-to-use Steering Framework for Editing Large Language Models
CUBETESTERAI: Automated JUnit Test Generation using the LLaMA Model
LLM-Assisted Translation of Legacy FORTRAN Codes to C++: A Cross-Platform Study
Agent for User: Testing Multi-User Interactive Features in TikTok
A Framework for Testing and Adapting REST APIs as LLM Tools
What's the Difference? Supporting Users in Identifying the Effects of Prompt and Model Changes Through Token Patterns
GENCNIPPET: Automated Generation of Code Snippets for Supporting Programming Questions
LRASGen: LLM-based RESTful API Specification Generation
LLM impact on BLV programming
Paper2Code: Automating Code Generation from Scientific Papers in Machine Learning
Combining Static and Dynamic Approaches for Mining and Testing Constraints for RESTful API Testing
Evaluating Grounded Reasoning by Code-Assisted Large Language Models for Mathematics