ICSME 2025
Sun 7 - Fri 12 September 2025 Auckland, New Zealand

This program is tentative and subject to change.

Large Language Models (LLMs) are increasingly applied to data-centric tasks in software maintenance and evo- lution, such as quality assurance and migration. While recent methods constrain LLM outputs using grammars or regular expressions, these syntactic techniques fail to enforce deeper semantic constraints involving numeric dependencies, conditional logic, and checksums. We present ClauseBandit, a framework that combines LLMs with Satisfiability Modulo Theories (SMT) solvers to generate structured data satisfying such constraints from natural language descriptions. ClauseBandit introduces a Bayesian inference approach that selects the most plausible SMT formula using posterior probabilities derived from formula self- consistency and data likelihoods. Evaluated on 27 structured generation tasks inspired by industrial use-cases, ClauseBandit successfully selected valid SMT formulas for 74.1% of tasks. Our approach enables LLM-based structured generation that goes beyond syntax, producing semantically valid, reusable constraint specifications from natural language.

This program is tentative and subject to change.

Fri 12 Sep

Displayed time zone: Auckland, Wellington change

15:30 - 16:30
Session 18 - Quality Assurance 3Registered Reports / Industry Track / Research Papers Track at Case Room 260-057
Chair(s): Chaiyong Ragkhitwetsagul Mahidol University, Thailand
15:30
15m
Full-paper
Evaluation of the Language Server Protocol for Static Dependency Analysis
Research Papers Track
Falko Galperin Axivion GmbH, Michel Krause Universtität Bremen, Rainer Koschke University of Bremen
Pre-print
15:45
10m
Are We Over-Testing? Studying Redundant Test Executions in Multi-Job CI Workflows
Registered Reports
Chuka Obi Department of Computer Science, Trent University, Peterborough, Canada, Taher A. Ghaleb Trent University
15:55
15m
Monitoring Continuous Integration Practices in Industry: A Case Study
Industry Track
Jadson Santos Universidade Federal do Rio Grande do Norte, Daniel Alencar Da Costa University of Otago, Uirá Kulesza Federal University of Rio Grande do Norte
16:10
15m
Constraint Discovery for Structured Generation via LLM-Guided SMT Inference
Industry Track
Hrishikesh Karmarkar TCS Research, Siddhesh Pagar TCS Research, Supriya Agrawal Tata Consultancy Services Ltd. (TCS), Vaibhavi Joshi TCS Research, Naman Paul TCS Research, Sagar Verma Tata Consultancy Services Ltd. (TCS)