An LLM-orchestrated multi-agent UI testing system discovers over 100 features and achieves 70% repair convergence across 300 reports but fails to produce executable tests in 38% of cases and relies on assertion weakening or deletion.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.SE 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Practical Limits of Autonomous Test Repair: A Multi-Agent Case Study with LLM-Driven Discovery and Self-Correction
An LLM-orchestrated multi-agent UI testing system discovers over 100 features and achieves 70% repair convergence across 300 reports but fails to produce executable tests in 38% of cases and relies on assertion weakening or deletion.