LLM method guarantees valid imports in generated tests without errors
No More Import Errors: Constrained Decoding for Guaranteed Valid Imports in LLM-Generated Tests
Software Engineering
Summary
Generating tests using large language models often fails because the code imports are invalid, causing errors. The authors propose a new way called Constrained Beam Decoding with Lookup (CBDL) that prevents these import errors while avoiding getting stuck in repetitive import patterns. This method can be added to existing test generation workflows and works for languages like Python and Java. Their approach fixes all import errors and improves the correctness of generated tests by reallocating effort to other issues. It also enables cheaper hybrid setups that split import and test generation between models.
What this means in practice
- •For software testing teams: Create fully import-valid automated tests in Python and Java, improving overall test reliability while reducing error-related generation failures.
- •For software integration engineers: Incorporate guaranteed-valid import generation into existing LLM-based pipelines to cut down on debugging and increase efficiency.
Authors
Konstantinos Kitsios, Matteo Biagiola, Guido Salvaneschi, Alberto Bacchelli
Abstract
One of the most common failure reasons of LLM-based test generation pipelines is import errors. Grammar constrained decoding can modify the probabilities of the LLM's final layer during token generation to guarantee valid imports. However, we find that it can lead to degenerate solutions, i.e., repetitive valid imports (import streaks) that exhaust the token budget before a test is generated. We propose Constrained Beam Decoding with Lookup (CBDL), a novel method that guarantees import validity while addressing import streaks by avoiding premature commitment to tokens that lead to repetitive imports. Because it operates at the decoding stage, CBDL can be integrated into existing LLM-based test generation pipelines. We implement CBDL for Python and Java and evaluate it on two test generation tasks, showing that generated tests have 100% valid imports, outperforming both static (RAG and LIBRO) and dynamic (Agentic Repair) import-validity approaches that leave 45%, 10%, and 43% of import errors unresolved. CBDL converts on average 14% of initial import errors into functionally correct tests, increasing overall test correctness by 1-4 percentage points over unconstrained generation. Even when import validity does not directly lead to a correct test, it allows agentic repair pipelines to allocate the repair budget to remaining errors, increasing correct tests by 2-7 percentage points for Python and 1-6 for Java. Finally, we show that a hybrid pipeline, where a small local model equipped with CBDL generates imports while the frontier GPT-5.2 model generates the test body, outperforms generating both with GPT-5.2 alone by 1-3 percentage points. Overall, CBDL removes a major bottleneck in LLM-based test generation: by guaranteeing import validity with a small overhead (16-21 seconds per test), it improves test correctness and enables cost-effective hybrid generation.