Do Large Language Models Respect Contracts? Evaluating and Enforcing Contract-Adherence in Code Generation

2025-10-13

AI Summary

Existing code generation benchmarks (e.g., HumanEval+, MBPP+) evaluate only functional correctness, neglecting adherence to formal contracts—such as preconditions and input constraints—leading to brittle, non-robust generated code. Method: We propose PACT, the first systematic evaluation framework for contract violation detection. It extends standard benchmarks into contract-aware test suites; introduces a hybrid static-dynamic methodology for assessing contract compliance; and employs contrastive prompting with explicit violation examples. We further define a quantifiable Contract Adherence Rate (CAR) metric. Contribution/Results: Experiments reveal that state-of-the-art LLMs achieve high functional correctness on traditional benchmarks but exhibit CARs below 40%—indicating widespread contract violations. Incorporating violation examples in prompting boosts CAR by up to 2.3× (e.g., for GPT-4). PACT establishes the first rigorous, measurable benchmark for contract robustness in code generation, addressing a critical gap in current evaluation paradigms.

Technology Category

• Natural Language Processing: Code Generation / Program Synthesis from Natural Language
• Constraint Satisfaction and Optimization: Satisfiability Modulo Theories
• Machine Learning: Large Multimodal Models (LMMs)

Application Category

• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Prevailing code generation benchmarks, such as HumanEval+ and MBPP+, primarily evaluate large language models (LLMs) with pass@k on functional correctness using well-formed inputs. However, they ignore a crucial aspect of real-world software: adherence to contracts-the preconditions and validity constraints that dictate how ill-formed inputs must be rejected. This critical oversight means that existing benchmarks fail to measure, and models consequently fail to generate, truly robust and reliable code snippets. We introduce PACT, a program assessment and contract-adherence evaluation framework, to bridge this gap. PACT is the first framework designed to systematically evaluate and enhance contract-adherence in LLM-generated code snippets alongside functional correctness. PACT's contributions are threefold: First, it provides a comprehensive test-suite corpus focused on contract violations, extending HumanEval+ and MBPP+. Second, it enables a systematic analysis of code generation under varied prompting conditions. This analysis demonstrates that augmenting prompts with contract-violating test cases significantly enhance a model's ability to respect contracts compared to using contract description alone. Finally, it introduces novel metrics to rigorously quantify contract adherence in both test generation and code generation. By revealing critical errors that conventional benchmarks overlook, PACT provides the rigorous and interpretable metrics to evaluate the robustness of LLM-generated code snippets in both functionality and contract-adherence.Our code and data are available at https://github.com/suhanmen/PACT.

Problem

• Evaluating LLMs' adherence to contracts in code generation
• Assessing how models handle ill-formed inputs and constraints
• Measuring robustness beyond functional correctness benchmarks

Innovation

• Introduces PACT framework for contract-adherence evaluation
• Augments prompts with contract-violating test cases
• Provides novel metrics to quantify contract adherence