SCAM 2026
Mon 14 - Tue 15 September 2026 Benevento, Italy
co-located with ICSME 2026

This program is tentative and subject to change.

Mon 14 Sep 2026 11:18 - 11:36 at A51S - Testing Chair(s): Saba Alimadadi

Large language models (LLMs) have shown promise in automated unit test generation, yet achieving high branch coverage for complex methods remains challenging. We attribute this limitation not to model capability alone, but to a fundamental cognitive misalignment between what LLM prompts expose and what branch reachability requires: implicit cross-method preconditions are hidden, while branch-irrelevant code is over-provided. To address this misalignment, we propose CogPath, a unit test generation framework that externalizes hidden path conditions as Constraint-Hints and isolates branch-relevant dependencies via coverage-driven Backward Slicing, within an iterative generate-execute-repair loop. Our empirical evaluation, conducted on 2971 methods with cyclomatic complexity greater than 10 from 14 open-source projects, demonstrates that CogPath achieves 12.8% higher line coverage and 10.2% higher branch coverage compared to the state-of-the-art, with improvements reaching 32.3% and 20% respectively on the most complex project.

This program is tentative and subject to change.

Mon 14 Sep

Displayed time zone: Amsterdam, Berlin, Bern, Rome, Stockholm, Vienna change

11:00 - 12:30
TestingResearch Track / Engineering Track at A51S
Chair(s): Saba Alimadadi Simon Fraser University
11:00
18m
Talk
Text Probes: Unit Testing of Source Code Analysis
Research Track
Anton Risberg Alaküla Lund University, Görel Hedin Lund University, Niklas Fors Lund University
Pre-print
11:18
18m
Talk
CogPath: A Constraint-Guided Context-Reduction Framework for LLM-Based Test Generation
Research Track
Jianhan Liu Nanjing University, Tangzhi Xu Nanjing Unversity, Yuan Yao Nanjing University, Feng Xu Nanjing University, Xiaoxing Ma Nanjing University
11:36
18m
Talk
Uncovering AGENTS.md: What Testing Guidance Open Source Projects Provide to Coding Agents
Research Track
Baris Ardic Delft University of Technology, Mitchell Olsthoorn Delft University of Technology, Andy Zaidman TU Delft
11:54
18m
Talk
Retrieval First: An Empirical Study of LLM-Based Type Inference in JavaScript
Research Track
Lucian Negru Delft University of Technology, Mitchell Olsthoorn Delft University of Technology
12:12
9m
Talk
Code Health in LLM-Based Test Generation: Effectiveness and Token Efficiency
Engineering Track
Freya Wirdemann Heidelberg University, Markus Borg CodeScene, Nadim Hagatulah Lund University, Adam Tornhill Codescene AB
12:21
9m
Talk
Fuzzing with Aliasing Structures
Engineering Track
Ruben Backx Delft University of Technology, Andy Zaidman TU Delft, Andreea Costea Delft University of Technology