CodeAssay: A Multi-Metric Benchmark with Audited Ground Truth for LLM Code Generation

• URL: http://arxiv.org/abs/2608.03535v1
• Date: Tue, 04 Aug 2026 12:15:17 GMT
• Title: CodeAssay: A Multi-Metric Benchmark with Audited Ground Truth for LLM Code Generation
• Abstract summary: We present CodeAssay, a taxonomy-first benchmark of 185 Python tasks across ten software-engineering categories.<n>It combines audited ground truth, public tests for generation and repair, hidden tests for grading, mutation-based test-suite validation, and selected code-property measures.
• Score: 1.8589311604031529
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Large Language Models are increasingly evaluated for code generation using test-based benchmarks. The validity of such evaluations depends on the reliability of their references and tests, while test-based correctness captures only part of the observable properties of generated code. We present CodeAssay, a taxonomy-first benchmark of 185 Python tasks across ten software-engineering categories. It combines audited ground truth, public tests for generation and repair, hidden tests for grading, mutation-based test-suite validation, and selected code-property measures. Regrading fixed model outputs after the audit changed 170 of 1,890 correctness labels (9.0%) and increased the measured best-to-worst model spread from 11.9 to 23.7 percentage points, although aggregate correctness remained nearly unchanged. The complete and hidden test suites achieved mutation scores of 82.6% and 74.8%, respectively. Across seven proprietary LLMs, standard-prompt correctness ranged from 77.3% to 98.9%, with significant differences in 12 of 21 model pairs. On the 120 tasks solved by all 14 model-prompt configurations, no model performed best across all selected code properties. A security-focused prompt produced no significant change in correctness or consistent reduction in the selected static-analysis findings, while increasing program length and cyclomatic complexity across all models. These findings show that reliable evaluation of LLM-generated code requires validated ground truth, protected tests, and multiple explicitly interpreted measures. CodeAssay provides a reproducible basis for evidence-based model evaluation in AI-augmented software development.

Related papers

• Are Benchmark Tests Strong Enough? Mutation-Guided Diagnosis and Augmentation of Regression Suites [49.16055123488827]
  Insufficiently strong test suites can admit plausible yet semantically incorrect patches, inflating reported success rates.<n>We introduce STING, a framework for targeted test augmentation that uses semantically altered program variants as diagnostic stressors.<n>STING produces 1,014 validated tests spanning 211 instances and increases patch-region line and branch coverage by 10.8% and 9.5%.
  arXiv  Detail & Related papers  (2026-04-02T01:13:40Z)
• HLE-Verified: A Systematic Verification and Structured Revision of Humanity's Last Exam [63.84155758655084]
  Humanity's Last Exam (HLE) has become a widely used benchmark for evaluating frontier large language models.<n>We introduce HLE-Verified, a verified and revised version of HLE with a transparent verification protocol and fine-grained error taxonomy.<n>We evaluate seven state-of-the-art language models on HLE and HLE-Verified, observing an average absolute accuracy gain of 7--10 percentage points.
  arXiv  Detail & Related papers  (2026-02-15T02:50:15Z)
• Unit Test Update through LLM-Driven Context Collection and Error-Type-Aware Refinement [5.8748750353007635]
  Test maintenance methods primarily focus on repairing broken tests, neglecting the scenario of enhancing existing tests to verify new functionality.<n>We propose TESTUPDATER, a novel approach that enables automated just-in-time test updates in response to production code changes.<n>TestUPDATER achieves a compilation pass rate of 94.4% and a test pass rate of 86.7%, outperforming the state-of-the-art method SYNTER by 15.9% and 20.0%, respectively.
  arXiv  Detail & Related papers  (2025-09-29T08:08:22Z)
• 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)
• Toward Automated Validation of Language Model Synthesized Test Cases using Semantic Entropy [0.5057850174013127]
  Modern Large Language Model (LLM)-based programming agents often rely on test execution feedback to refine their generated code.<n>This paper introduces VALTEST, a novel framework that leverages semantic entropy to automatically validate test cases generated by LLMs.<n>Experiments show that VALTEST boosts test validity by up to 29% and improves code generation performance, as evidenced by significant increases in pass@1 scores.
  arXiv  Detail & Related papers  (2024-11-13T00:07:32Z)
• TestGenEval: A Real World Unit Test Generation and Test Completion Benchmark [24.14654309612826]
  TestGenEval comprises 68,647 tests from 1,210 code and test file pairs across 11 well-maintained Python repositories.<n>It covers initial tests authoring, test suite completion, and code coverage improvements.<n>We evaluate several popular models, with sizes ranging from 7B to 405B parameters.
  arXiv  Detail & Related papers  (2024-10-01T14:47:05Z)
• Measuring the Influence of Incorrect Code on Test Generation [22.168699378889148]
  We show that tests generated for incorrect code experience a 47% worse bug detection rate.<n>Improvements of +18% in accuracy, +4% coverage, and +34% in bug detection can be achieved by providing natural language code descriptions.
  arXiv  Detail & Related papers  (2024-09-14T15:17:34Z)
• Self-Evaluation Improves Selective Generation in Large Language Models [54.003992911447696]
  We reformulate open-ended generation tasks into token-level prediction tasks. We instruct an LLM to self-evaluate its answers. We benchmark a range of scoring methods based on self-evaluation.
  arXiv  Detail & Related papers  (2023-12-14T19:09:22Z)
• LLMs as Factual Reasoners: Insights from Existing Benchmarks and Beyond [135.8013388183257]
  We propose a new protocol for inconsistency detection benchmark creation and implement it in a 10-domain benchmark called SummEdits. Most LLMs struggle on SummEdits, with performance close to random chance. The best-performing model, GPT-4, is still 8% below estimated human performance.
  arXiv  Detail & Related papers  (2023-05-23T21:50:06Z)
• An ensemble meta-estimator to predict source code testability [1.4213973379473652]
  The size of a test suite determines the test effort and cost, while the coverage measure indicates the test effectiveness. This paper offers a new equation to estimate testability regarding the size and coverage of a given test suite.
  arXiv  Detail & Related papers  (2022-08-20T06:18:16Z)
• 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.