This paper examines the instruction stacks of five coding agents to understand their composition and evolution.

The paper introduces the Last Secure Code Benchmark to evaluate secure code generation by coding agents.

This paper uses large language models to improve the best-known bounds for binary linear codes.

MATRIX is a novel, robust code watermarking framework that encodes watermarks using constrained parity-check matrix equations, achieving high detection accuracy and improved robustness for code proven…

The paper tests the premise that code-graph tools improve the performance of coding agents by providing a repository's structure.

This paper investigates code generation for no-resource programming languages using large language models.

The paper introduces Spine, an interface and runtime that improves the efficiency of long-horizon coding tasks by aligning model context with subtask nesting and completion.

This paper evaluates the use of rendered code as operational context for agentic coding, showing consistent reduction in prompt-token cost but limited increase in linear savings and end-to-end repair…

The study found that while multi-agent LLM code generation architectures significantly affect code complexity, the added complexity does not translate into better functional correctness, suggesting ar…

This paper analyzes various source-to-bytecode obfuscation techniques for Erlang, demonstrating that effective protection relies on exploiting the representational gaps between high-level semantics an…

This paper analyzes the efficiency of AI coding agents and identifies the opportunity for reducing unnecessary turns in their loop engineering process.

This paper presents MediaWiki Code2Code Search, a neural retrieval system for semantic code-to-code discovery in large-scale ecosystems, achieving superior performance over traditional Information Ret…

This paper introduces Recursive Agent Harness (RAH), a code-first extension to recursive language models, and evaluates its effectiveness in long-context reasoning.

The paper introduces Spaghetti Architect, a tool that generates controlled code datasets for machine learning models by deliberately adding redundancy, messiness, and difficulty.

This paper identifies the 'Format-Reliability Gap'—where LLMs know about code vulnerabilities but generate insecure code anyway—and proposes a localized, per-vulnerability steering vector fix that sig…

This paper introduces TRIM, an algorithm that minimizes redundant edits in AI-generated code, called CodeSlop, by minimizing agent trajectories, reducing CodeSlop by 17.9%-32.9% with negligible perfor…

The authors argue that coding agents can replace traditional human code review in software development.