TestHumanizer uses LLMs as controlled refactoring layers on EvoSuite suites to boost readability and maintainability, achieving 88-98% compilation rates and developer preference gains on 350 classes from Defects4J and SF110.
arXiv preprint arXiv:2510.16579 , year =
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.SE 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
Experiments across code LLMs show no-review collapses fastest, human-gated filters slow collapse, and AI self-gates lose effect over time, degenerating to ungated self-training under self-confirming acceptance as proven via gated distributional reweighting and spectral analysis.
citing papers explorer
-
Humanizing Automatically Generated Unit Test Suites with LLM-Based Refactoring
TestHumanizer uses LLMs as controlled refactoring layers on EvoSuite suites to boost readability and maintainability, achieving 88-98% compilation rates and developer preference gains on 350 classes from Defects4J and SF110.
-
When AI Reviews Its Own Code: Recursive Self-Training Collapse in Code LLMs
Experiments across code LLMs show no-review collapses fastest, human-gated filters slow collapse, and AI self-gates lose effect over time, degenerating to ungated self-training under self-confirming acceptance as proven via gated distributional reweighting and spectral analysis.