# SemGuard: Real-Time Semantic Evaluator for Correcting LLM-Generated Code

arXiv: https://arxiv.org/abs/2509.24507 · HTML: https://mlsanity.com/paper/2509.24507/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 29 Sep 2025.

## Built on by (most cited first, 2 of 2)

- [Hydra: Efficient, Correct Code Generation via Checkpoint-and-Rollback Support](https://mlsanity.com/paper/2605.15238/) (2026)
- [SemOPT: Fixing Semantic Errors in LLM-based Optimization Modeling via Reward-Guided Search](https://mlsanity.com/paper/2609.37361/) (2026)

## Abstract (arXiv)

Large Language Models (LLMs) can translate natural language requirements into code, yet empirical analyses of representative models reveal that semantic errors-programs that compile but behave incorrectly-constitute the majority of observed faults (e.g., >60% on DeepSeek-Coder-6.7B and QwenCoder-7B). Post-hoc repair pipelines detect such faults only after execution, incurring latency, relying on incomplete test suites, and often mis-localizing the defect. Since semantic drift originates in the autoregressive decoding process, intervening while the code is being generated is a direct way to stop error propagation. Constrained-decoding approaches such as ROCODE attempt this, but still wait until the entire program runs to obtain feedback and use entropy heuristics that do not truly capture semantics. A more effective solution must inject semantic signals-early and precisely-into the decoding process.We present SemGuard, a semantic-evaluator-driven framework that performs real-time, line-level semantic supervision. To train the evaluator, we build SemDiff, the first dataset with fine-grained annotations that mark the exact line where a correct and an incorrect implementation diverge. The evaluator, once embedded in the LLM's decoder, flags deviations on partial code, rolls back to the faulty line, and guides regeneration-without executing the program or requiring test cases. Across four benchmarks, SemGuard consistently outperforms state-of-the-art baselines. It lowers the semantic error rate by 19.86% on SemDiff relative to ROCODE, and lifts Pass@1 by 48.92% on the real-world LiveCodeBench with CodeLlama-7B. Similar gains hold for StarCoder2-7B on MBPP and for DeepSeekCoder-6.7B on the Java benchmark SemDiff-Java, demonstrating model- and language-agnostic effectiveness.