Agent-generated tests mainly act as observational feedback channels and do not meaningfully improve issue resolution success in current LLM software engineering agents.
In2023 IEEE 23rd International Conference on Software Quality, Reliability, and Security (QRS)
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
-
Rethinking the Value of Agent-Generated Tests for LLM-Based Software Engineering Agents
Agent-generated tests mainly act as observational feedback channels and do not meaningfully improve issue resolution success in current LLM software engineering agents.