The 2nd International Workshop on AI for Software Modernization (AISM 2026)

Application modernization involves upgrading software systems to adopt modern technologies, architectures, and programming paradigms while meeting evolving business requirements. Studies suggest that modernization accounts for a significant portion of software maintenance costs, underscoring the need for scalable, automated, and AI-driven solutions.

Common modernization activities include :

• Transforming legacy systems (e.g., COBOL to Java, C++ to Rust)
• Upgrading languages, frameworks, and dependencies
• Migrating from on-premises systems to cloud-native architectures
• Refactoring monolithic applications into microservices
• Migrating and transforming legacy data models, particularly from file-based or hierarchical systems to relational or distributed data stores
• Extracting business rules and process specifications from legacy systems to configure Commercial Off-The-Shelf (COTS) products

Despite its importance, modernization remains challenging due to the need to preserve application semantics, accurately estimate effort, and ensure correctness after transformation and maintain data integrity during schema evolution and migration.

This workshop focuses on the role of Artificial Intelligence in software modernization, including application understanding, transformation, testing, and evaluation. We also welcome submissions that advance traditional (non-AI) techniques, particularly when combined with or evaluated alongside AI-driven approaches.

The workshop will feature :

• Peer-reviewed research and vision papers
• Invited talks from leading researchers and practitioners
• Interactive discussions on open challenges and future directions

Proceedings

All accepted papers will be included in the ASE 2026 conference proceedings. The proceedings will be made available online and indexed in the ACM/IEEE Digital Library.

Call for Papers

We invite high-quality, original research contributions, including but not limited to the following areas:

1. Application Understanding

• AI-driven functionality detection and classification
• Architecture extraction
• Business rule extraction from legacy codebases
• AI-powered question-answering and retrieval-based techniques for understanding application logic
• AI-based code search and summarization
• Defining and evaluating metrics for application understanding and summarization
• Understanding legacy data models (e.g., file-based, hierarchical, VSAM) and their relationships with application logic

2. Modernization Design and Effort Estimation

• AI-driven insights on the potential impact of modernization changes, including downtime, compatibility issues, and risk mitigation strategies
• Mapping and rearchitecting legacy applications (e.g., monolith to microservices)
• AI-based recommendation systems for modernization planning and estimation of modernization complexity, cost, and effort
• Effort estimation for data migration, including schema evolution, data quality challenges, and system interoperability constraints

3. Application Transformation

• Automated extraction, modularization, and migration of application functionality and database systems
• AI-generated transformation and refactoring rules
• Fine-tuning AI models for transformation-aware code generation
• AI-driven automated UI modernization
• Large-scale, multi-language migration frameworks
• Agentic approach to feedback-driven transformation
• AI-driven automated refactoring
• AI-assisted data model transformation (e.g., file-to-relational, relational-to-NoSQL, schema normalization/denormalization)
• Automated generation of data mapping and transformation logic between legacy and target systems

4. Data Migration and Validation

• AI-driven schema matching and mapping between legacy and modern data models
• Data extraction, cleansing, and transformation from legacy storage systems (e.g., flat files, VSAM, IMS)
• Automated detection of data anomalies, inconsistencies, and missing values during migration
• AI-assisted validation of migrated data for completeness, correctness, and consistency
• Techniques for preserving data semantics and business constraints across schema evolution
• Incremental and zero-downtime data migration strategies using AI support

5. Testing, Debugging, and Repair

• Ensuring semantic preservation in automated transformations
• AI-based testing strategies for modernized applications
• Coverage metrics for program transformation correctness
• Automated generation of functional test suites
• AI-driven defect detection and iterative repair of transformed code
• Defining and evaluating metrics for transformation quality
• Validation strategies that jointly verify code and data correctness post-migration

6. Case Studies and Applications

• Real-world applications of AI in modernization
• Development and adoption of AI-driven modernization frameworks and tools
• Empirical studies and lessons learned in large-scale migration projects

Submission Guidelines

All papers must be original and must not have been previously published or be under review for any other publication venue.

Submissions must follow the Formatting Guidelines of the ASE 2026 Research Track.

We accept two types of submissions:

• Full papers: 8 pages (including references), presenting mature research contributions or industry experience papers.
• Short papers: 4 pages (including references), presenting new ideas, preliminary results, or early-stage work.

No additional pages are permitted.

The workshop follows a single-blind review process.

The Microsoft CMT service was used for managing the peer-reviewing process for this conference. This service was provided for free by Microsoft and they bore all expenses, including costs for Azure cloud services as well as for software development and support.