Related papers: Generating Software for Well-Understood Domains
It has recently been shown that reinforcement learning can be used to train generators capable of producing high-quality game levels, with quality defined in terms of some user-specified heuristic. To ensure that these generators' output is…
The creation and deployment of software development processes for new domains (such as wireless Internet services) is a challenging task due to the lack of knowledge about adequate engineering techniques and their effects. In addition,…
Recent advancements in generative models have unlocked the capabilities to render photo-realistic data in a controllable fashion. Trained on the real data, these generative models are capable of producing realistic samples with minimal to…
Software Documentation plays a major role in the usage and development of a project. Widespread adoption of open source software projects contributes to larger and faster development of the projects, making it difficult to maintain the…
Recent domain generalized semantic segmentation (DGSS) studies have achieved notable improvements by distilling semantic knowledge from Vision-Language Models (VLMs). However, they overlook the semantic misalignment between visual and…
Distribution of software development is becoming more and more common in order to save the production cost and reduce the time to market. Large geographical distance, different time zones and cultural differences in distributed software…
Nowadays, software has become a complex piece of work that may be beyond our control. Understanding how software evolves over time plays an important role in controlling software development processes. Recently, a few researchers found the…
Intelligent coding systems are transforming software development by enabling users to specify code behavior in natural language. However, the opaque decision-making of AI-driven coders raises trust and usability concerns, particularly for…
Random graph models are frequently used as a controllable and versatile data source for experimental campaigns in various research fields. Generating such data-sets at scale is a non-trivial task as it requires design decisions typically…
Program comprehension is a fundamental task in software development and maintenance processes. Software developers often need to understand a large amount of existing code before they can develop new features or fix bugs in existing…
Code generation is increasingly critical for real-world applications. Still, diffusion-based large language models continue to struggle with this demand. Unlike free-form text, code requires syntactic precision; even minor structural…
Model-driven software development is a promising way to cope with the complexity of system integration in advanced robotics, as it already demonstrated its benefits in domains with comparably challenging system integration requirements.…
High-fidelity generative models are increasingly needed in privacy-sensitive scenarios, where access to data is severely restricted due to regulatory and copyright constraints. This scarcity hampers model development--ironically, in…
We propose Composition Sampling, a simple but effective method to generate diverse outputs for conditional generation of higher quality compared to previous stochastic decoding strategies. It builds on recently proposed plan-based neural…
Generative models for source code are an interesting structured prediction problem, requiring to reason about both hard syntactic and semantic constraints as well as about natural, likely programs. We present a novel model for this problem…
Software interfaces today generally fall at either end of a spectrum. On one end are programmable systems, which allow expert users (i.e. programmers) to write software artifacts that describe complex abstractions, but programs are…
Large language models (LLMs) have revolutionized code generation, automating programming with remarkable efficiency. However, these advancements challenge programming skills, ethics, and assessment integrity, making the detection of…
Search-based Software Testing (SBST) can automatically generate test cases to search for requirements violations. Unlike manual test case development, it can generate a substantial number of test cases in a limited time. However, SBST does…
The vast majority of the long tail of scientific software, the myriads of tools that implement the many analysis and visualization methods for different scientific fields, is highly specialized, purpose-built for a research project, and has…
Causality has been combined with machine learning to produce robust representations for domain generalization. Most existing methods of this type require massive data from multiple domains to identify causal features by cross-domain…