Search papers, labs, and topics across Lattice.
This paper introduces SpecTune, a novel automated program repair (APR) framework that leverages LLM-generated specifications as intermediate behavioral signals to guide the repair process. SpecTune decomposes the repair task into suspicious regions and derives localized postconditions representing expected program behaviors, using these to generate micro-level debugging signals. By incorporating specification validation and discriminative signals to address the unreliability of LLM-generated postconditions, SpecTune achieves improved fault localization and APR effectiveness compared to baseline methods.
LLMs can debug code more effectively by reasoning about intermediate program states, not just relying on coarse pass/fail test signals.
Automated Program Repair (APR) has recently benefited from large language models (LLMs). However, most LLM-based APR approaches still rely primarily on coarse end-to-end signals from test-suite outcomes to guide repair, providing limited insight into where a program's internal logic deviates from its intended behavior. In contrast, human debugging often relies on intermediate reasoning about program states through localized correctness conditions or assertions. Inspired by this observation, we propose SpecTune, a specification-guided debugging framework that incorporates intermediate behavioral reasoning into APR. SpecTune decomposes the repair task into suspicious regions connected by execution checkpoints and derives localized postconditions representing expected program behaviors at those points. By executing the buggy program and evaluating these postconditions, SpecTune produces micro-level debugging signals that indicate mismatches between observed and intended behaviors, enabling more precise fault localization and targeted patch generation. To address the potential unreliability of LLM-generated postconditions, we introduce two complementary signals: a specification validation signal alpha, which estimates the consistency of generated postconditions using partially passing test cases, and a discriminative signal beta, which detects violations of validated postconditions during execution. With these signals, SpecTune safely leverages automatically generated specifications for APR. Experimental results show that SpecTune improves fault localization and APR effectiveness than the baselines.
