Co-Located Tests, Better AI Code: How Test Syntax Structure Affects Foundation Model Code Generation

• URL: http://arxiv.org/abs/2604.19826v1
• Date: Mon, 20 Apr 2026 14:47:46 GMT
• Title: Co-Located Tests, Better AI Code: How Test Syntax Structure Affects Foundation Model Code Generation
• Abstract summary: How developers structure test code, whether inline with the implementation or in separate blocks, has traditionally been a matter of testing philosophy.<n>We investigate whether this choice affects AI code generation quality using SEGA, a three-dimensional evaluation framework measuring Determinism, Preservation, and Correctness.
• Score: 0.7310043452300737
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: AI coding assistants increasingly generate code alongside tests. How developers structure test code, whether inline with the implementation or in separate blocks, has traditionally been a matter of testing philosophy. We investigate whether this choice affects AI code generation quality. We conduct a large-scale empirical study (830+ generated files, 12 models, 3 providers) using SEGA, a three-dimensional evaluation framework measuring Determinism, Preservation, and Correctness. Comparing inline test syntax (Python doctests) against separated test syntax (Rust #[test] blocks) on a d-ary heap implementation, we find that: (1) inline tests yield near-perfect preservation (100%) and correctness (92-100%) across all models; (2) separated tests expose stark model-tier gaps (0-100% correctness) and independence between preservation and correctness; (3) model behavior evolves across generations, and notably one model breaks the test suppression pattern of its three predecessors; (4) mechanistic analysis on 7 open-source architectures (6 transformers and a gated-linear Recurrent Neural Network (RNN)) reveals inline test markers receive 2.8-4.4$\times$ stronger attention in 5/7 models, with causal validation via knockout and steering experiments on the 4 code-specialized transformers and RWKV-6; the co-location mechanism extends to a non-transformer architecture, suggesting the design recommendation is robust to future architectural shifts. In the Foundation Model era, test syntax structure is a software design concern: co-locating tests with implementation code produces measurably better AI-generated code. This arxiv long version includes appendices that further qualify the effect as bounded by both model capability and programming language.

Related papers

• ZeroCoder: Can LLMs Improve Code Generation Without Ground-Truth Supervision? [13.984583399745157]
  Reinforcement Learning with Verifiable Rewards (RLVR) is a powerful paradigm to improve it through execution-based feedback.<n>Existing work tried to use self-generated tests to ground rewards, but the lack of discriminative tests constrains the effect due to the sub-optimal performance of the model on test generation.<n>We present ZeroCoder, a fully label-free co-evolutionary framework that jointly trains a Coder and a Tester using execution feedback from self-generated code-test interactions.
  arXiv  Detail & Related papers  (2026-04-09T06:24:54Z)
• Consistency Meets Verification: Enhancing Test Generation Quality in Large Language Models Without Ground-Truth Solutions [1.9196411948992402]
  We present ConVerTest, a novel two-stage pipeline for synthesizing reliable tests without requiring prior code implementations.<n>Experiments on BIGCODEBENCH and LESS BASIC PYTHON PROBLEMS benchmarks demonstrate that ConVerTest improves test validity, line coverage, and mutation scores by up to 39%, 28%, and 18% respectively.
  arXiv  Detail & Related papers  (2026-02-11T04:40:38Z)
• AlgoVeri: An Aligned Benchmark for Verified Code Generation on Classical Algorithms [54.99368693313797]
  Existing benchmarks test only individual languages/tools, so the performance numbers are not directly comparable.<n>We address this gap with AlgoVeri, a benchmark that evaluates vericoding of $77$ classical algorithms in Dafny, Verus, and Lean.
  arXiv  Detail & Related papers  (2026-02-10T06:58:26Z)
• BRIDGE: Building Representations In Domain Guided Program Verification [67.36686119518441]
  BRIDGE decomposes verification into three interconnected domains: Code, Specifications, and Proofs.<n>We show that this approach substantially improves both accuracy and efficiency beyond standard error feedback methods.
  arXiv  Detail & Related papers  (2025-11-26T06:39:19Z)
• ATGen: Adversarial Reinforcement Learning for Test Case Generation [78.48498301767079]
  Large Language Models (LLMs) excel at code generation, yet their outputs often contain subtle bugs.<n>Existing test generation methods rely on static datasets.<n>We introduce ATGen, a framework that trains a test case generator via adversarial reinforcement learning.
  arXiv  Detail & Related papers  (2025-10-16T12:49:25Z)
• Impact of Code Context and Prompting Strategies on Automated Unit Test Generation with Modern General-Purpose Large Language Models [0.0]
  Generative AI is gaining increasing attention in software engineering.<n>Unit tests constitute the majority of test cases and are often schematic.<n>This paper investigates the impact of code context and prompting strategies on the quality and adequacy of unit tests.
  arXiv  Detail & Related papers  (2025-07-18T11:23:17Z)
• Test3R: Learning to Reconstruct 3D at Test Time [58.0912500917036]
  Test3R is a surprisingly simple test-time learning technique that significantly boosts geometric accuracy.<n>Our technique significantly outperforms previous state-of-the-art methods on the 3D reconstruction and multi-view depth estimation tasks.
  arXiv  Detail & Related papers  (2025-06-16T17:56:22Z)
• Learning to Solve and Verify: A Self-Play Framework for Code and Test Generation [69.62857948698436]
  Recent advances in large language models (LLMs) have improved their performance on coding benchmarks.<n>However, improvement is plateauing due to the exhaustion of readily available high-quality data.<n>We propose Sol-Ver, a self-play solver-verifier framework that jointly improves a single model's code and test generation capacity.
  arXiv  Detail & Related papers  (2025-02-20T18:32:19Z)
• SWT-Bench: Testing and Validating Real-World Bug-Fixes with Code Agents [10.730852617039451]
  We investigate the capability of LLM-based Code Agents to formalize user issues into test cases.<n>We propose a novel benchmark based on popular GitHub repositories, containing real-world issues, ground-truth bug-fixes, and golden tests.<n>We find that LLMs generally perform surprisingly well at generating relevant test cases, with Code Agents designed for code repair exceeding the performance of systems designed for test generation.
  arXiv  Detail & Related papers  (2024-06-18T14:54:37Z)
• CAT-LM: Training Language Models on Aligned Code And Tests [19.526181671936243]
  Testing is an integral part of the software development process. Yet, writing tests is time-consuming and therefore often neglected. We propose the Aligned Code And Tests Language Model (CAT-LM), a GPT-style language model with 2.7 Billion parameters, trained on a corpus of Python and Java projects.
  arXiv  Detail & Related papers  (2023-10-02T19:52:22Z)
• CodeT: Code Generation with Generated Tests [49.622590050797236]
  We explore the use of pre-trained language models to automatically generate test cases. CodeT executes the code solutions using the generated test cases, and then chooses the best solution. We evaluate CodeT on five different pre-trained models with both HumanEval and MBPP benchmarks.
  arXiv  Detail & Related papers  (2022-07-21T10:18:37Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.