A Survey of Machine Learning for Big Code and Naturalness

2017/09/18 by Miltiadis Allamanis, Allamanis, Miltiadis, Earl T. Barr +5 · 41 citations
Computer Science · #FOS: Computer and information sciences #Machine Learning (cs.LG) #Programming Languages (cs.PL) #Software Engineering (cs.SE) #cs.LG #cs.PL #cs.SE

paper · pdf · doi:10.48550/arxiv.1709.06182

Website accompanying this survey paper can be found at https://ml4code.github.io

arxiv created 2018/05/05 · arxiv updated 2018/05/08

Abstract

Research at the intersection of machine learning, programming languages, and software engineering has recently taken important steps in proposing learnable probabilistic models of source code that exploit code's abundance of patterns. In this article, we survey this work. We contrast programming languages against natural languages and discuss how these similarities and differences drive the design of probabilistic models. We present a taxonomy based on the underlying design principles of each model and use it to navigate the literature. Then, we review how researchers have adapted these models to application areas and discuss cross-cutting and application-specific challenges and opportunities.

Cited by

• Alternative UX Extensions and Their Trade-offs for Code Completion in Pharo
• Larger Is Not Always Better: Leveraging Structured Code Diffs for Comment Inconsistency Detection
• Beyond Language Boundaries: Uncovering Programming Language Families for Code Language Models
• Completion by Comprehension: Guiding Code Generation with Multi-Granularity Understanding
• From Code Changes to Quality Gains: An Empirical Study in Python ML Systems with PyQu
• A Systematic Literature Review of Code Hallucinations in LLMs: Characterization, Mitigation Methods, Challenges, and Future Directions for Reliable AI
• Can Large Language Models Detect Real-World Android Software Compliance Violations?
• Teaching Machine Learning to Software Engineers
• Scaling Laws for Code: A More Data-Hungry Regime
• DeepCodeSeek: Real-Time API Retrieval for Context-Aware Code Generation
• RANGER -- Repository-Level Agent for Graph-Enhanced Retrieval
• Previously on... Automating Code Review
• On the Soundness and Consistency of LLM Agents for Executing Test Cases Written in Natural Language
• Discovering Software Parallelization Points Using Deep Neural Networks
• Automated Repair of C Programs Using Large Language Models
• The Fools are Certain; the Wise are Doubtful: Exploring LLM Confidence in Code Completion
• The Gold Medals in an Empty Room: Diagnosing Metalinguistic Reasoning in LLMs with Camlang
• Does AI Code Review Lead to Code Changes? A Case Study of GitHub Actions
• AI Agentic Programming: A Survey of Techniques, Challenges, and Opportunities
• RAG for Geoscience: What We Expect, Gaps and Opportunities
• Key-Augmented Neural Triggers for Knowledge Sharing
• A Pilot Study on LLM-Based Agentic Translation from Android to iOS: Pitfalls and Insights
• A Systematic Literature Review on Detecting Software Vulnerabilities with Large Language Models
• Vibe Coding as a Reconfiguration of Intent Mediation in Software Development: Definition, Implications, and Research Agenda
• Automated Code Review Using Large Language Models with Symbolic Reasoning
• AlgoTune: Can Language Models Speed Up General-Purpose Numerical Programs?
• Automated Test Data Generation for Enterprise Protobuf Systems: A Metaclass-Enhanced Statistical Approach
• Syntax Repair as Language Intersection
• PALM: Synergizing Program Analysis and LLMs to Enhance Rust Unit Test Coverage
• Formal Methods Meets Readability: Auto-Documenting JML Java Code
• Assessing Small Language Models for Code Generation: An Empirical Study with Benchmarks
• Computer Science Education in the Age of Generative AI
• SACL: Understanding and Combating Textual Bias in Code Retrieval with Semantic-Augmented Reranking and Localization
• LLM-based Multi-Agent System for Intelligent Refactoring of Haskell Code
• Directed Acyclic Graph Convolutional Networks
• Parameter-Efficient Fine-Tuning with Attributed Patch Semantic Graph for Automated Patch Correctness Assessment
• Deconstructing Obfuscation: A four-dimensional framework for evaluating Large Language Models assembly code deobfuscation capabilities
• Notation-level confounding: When inconsistent molecular notations mislead chemical language models
• Towards Leveraging Large Language Model Summaries for Topic Modeling in Source Code
• TD-Suite: All Batteries Included Framework for Technical Debt Classification
• Towards an Understanding of Context Utilization in Code Intelligence
• Bringing Structure to Naturalness: On the Naturalness of ASTs
• Deep Learning-based Intrusion Detection Systems: A Survey

Related

• Deep Probabilistic Programming Languages: A Qualitative Study
• CodeBERT-nt: code naturalness via CodeBERT
• Model-based machine learning
• Predicting Program Properties from "Big Code"
• Automated learning with a probabilistic programming language: Birch