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

• URL: http://arxiv.org/abs/2509.24507v1
• Date: Mon, 29 Sep 2025 09:21:32 GMT
• Title: SemGuard: Real-Time Semantic Evaluator for Correcting LLM-Generated Code
• Abstract summary: Post-hoc repair pipelines detect such faults only after execution.<n>We present SemGuard, a semantic-evaluator-driven framework that performs real-time, line-level semantic supervision.
• Score: 46.20378145112059
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: 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.

Related papers

• Robustness of LLM-Generated SystemVerilog Assertions to Semantics-Preserving RTL Transformations [0.0]
  Large language models (LLMs) are increasingly being explored for automated SystemVerilog Assertion (SVA) generation.<n>Most evaluations report correctness on a single syntactic representation of an input.<n>Such point accuracy does not reveal whether a model's correct output is stable when the same RTL behavior is written differently.
  arXiv  Detail & Related papers  (2026-09-04T18:41:58Z)
• Fantastic Adaptive Taxonomies and How to Use Them [100.20614173157708]
  AdaMAST converts execution traces into a compact, evidence-grounded failure taxonomy.<n>No code is hand-authored, no trace human-annotated.<n>In agent-system search, taxonomy-coded diagnoses of failed candidates outperform free-form.
  arXiv  Detail & Related papers  (2026-07-17T17:41:16Z)
• Constrained Decoding for Diffusion Language Models via Efficient Inference over Finite Automata [57.27430779838529]
  We present an exact and tractable algorithm for sampling from the constrained mean-field posterior under any constraint expressible as a finite automaton.<n>The approach guarantees constraint satisfaction by construction, supports both greedy and sampling-based decoding, and is compatible with parallel and block-wise decoding.<n> Empirical evaluations on Dream-7B and LLaDA-8 show substantial accuracy gains across various tasks.
  arXiv  Detail & Related papers  (2026-07-08T05:48:57Z)
• An Execution-Verified Multi-Language Benchmark for Code Semantic Reasoning [10.815335026032377]
  TraceEval is the first execution-verified, multi-language benchmark for code semantic reasoning.<n>TraceEval consists of 10,583 real-world programs extracted from 1,600+ open-source repositories across Python, JavaScript, and Java.<n>The strongest model, Claude-Opus-4.6, reaches an average F1 of 72.9% across the three languages.
  arXiv  Detail & Related papers  (2026-05-10T07:17:00Z)
• Beyond Translation Accuracy: Addressing False Failures in LLM-Based Code Translation [2.3411689215528346]
  Large Language Models (LLMs) have achieved remarkable success in automated code translation.<n>In this paper, we demonstrate that a significant number of reported failures in code translation are not due to incorrect logic, but rather evaluation-induced errors.
  arXiv  Detail & Related papers  (2026-05-04T03:49:58Z)
• SPARC: Scenario Planning and Reasoning for Automated C Unit Test Generation [1.0010193170880752]
  We introduce a neuro-symbolic, scenario-based framework that bridges the gap between high-level program intent and the rigid syntactic constraints of pointer arithmetic and manual memory management.<n>We evaluate it on 59 real-world and algorithmic subjects, where it outperforms the vanilla prompt generation baseline by 31.36% in line coverage, 26.01% in branch coverage, and 20.78% in mutation score, matching or exceeding the symbolic execution tool KLEE.
  arXiv  Detail & Related papers  (2026-02-18T18:09:03Z)
• 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)
• Can LLMs Compress (and Decompress)? Evaluating Code Understanding and Execution via Invertibility [36.41073880422337]
  We present RoundTripCodeEval (RTCE), a comprehensive benchmark consisting of four distinct code execution reasoning tasks.<n>We systematically evaluate state-of-the-art Code-LLMs using zero-shot prompting, supervised fine-tuning on execution traces, and self-reflection mechanisms.<n>RTCE surfaces several new and previously unmeasured insights that are not captured by existing I/O-prediction, execution-reasoning, or round-trip natural-language benchmarks.
  arXiv  Detail & Related papers  (2026-01-19T21:09:48Z)
• CARE What Fails: Contrastive Anchored-REflection for Verifiable Multimodal [84.71254539482369]
  Group-relative reinforcement learning with verifiable rewards (RLVR) often wastes the most informative data it already has the failures.<n>We present CARE, a failure-centric post-training framework for multimodal reasoning that turns errors into supervision.<n> CARE improves accuracy and training smoothness while explicitly increasing the share of learning signal that comes from failures.
  arXiv  Detail & Related papers  (2025-12-22T16:34:21Z)
• Time Travel: LLM-Assisted Semantic Behavior Localization with Git Bisect [8.55768450285885]
  We present a novel framework that integrates Large Language Models (LLMs) into the Git bisect process for semantic fault localization.<n>Our system augments bisect traversal with structured chain of thought reasoning, enabling commit by commit analysis under noisy conditions.
  arXiv  Detail & Related papers  (2025-11-24T07:49:59Z)
• Towards Automated Error Discovery: A Study in Conversational AI [48.735443116662026]
  We introduce Automated Error Discovery, a framework for detecting and defining errors in conversational AI.<n>We also propose SEEED (Soft Clustering Extended-Based Error Detection), as an encoder-based approach to its implementation.
  arXiv  Detail & Related papers  (2025-09-13T14:53:22Z)
• Probing Pre-trained Language Models on Code Changes: Insights from ReDef, a High-Confidence Just-in-Time Defect Prediction Dataset [0.0]
  We present ReDef, a high-confidence benchmark of function-level modifications curated from 22 large-scale C/C++ projects.<n>Defective cases are anchored by revert commits, while clean cases are validated through post-hoc history checks.<n>This pipeline yields 3,164 defective and 10,268 clean modifications, offering substantially more reliable labels than prior existing resources.
  arXiv  Detail & Related papers  (2025-09-11T07:07:11Z)
• Diffusion Language Models Know the Answer Before Decoding [56.96815863705218]
  Diffusion language models (DLMs) have emerged as an alternative to autoregressive approaches.<n>Our work highlights and leverage an overlooked property of DLMs early answer convergence.<n>We introduce Prophet, a training-free fast decoding paradigm that enables early commit decoding.
  arXiv  Detail & Related papers  (2025-08-27T15:40:25Z)
• Repairing vulnerabilities without invisible hands. A differentiated replication study on LLMs [5.10123605644148]
  Automated Vulnerability Repair (AVR) is a fast-growing branch of program repair.<n>Recent studies show that large language models (LLMs) outperform traditional techniques.
  arXiv  Detail & Related papers  (2025-07-28T16:39:16Z)
• EquiBench: Benchmarking Large Language Models' Reasoning about Program Semantics via Equivalence Checking [58.15568681219339]
  We introduce EquiBench, a new benchmark for evaluating large language models (LLMs)<n>This task directly tests a model's ability to reason about program semantics.<n>We evaluate 19 state-of-the-art LLMs and find that in the most challenging categories, the best accuracies are 63.8% and 76.2%, only modestly above the 50% random baseline.
  arXiv  Detail & Related papers  (2025-02-18T02:54:25Z)
• A test-free semantic mistakes localization framework in Neural Code Translation [32.5036379897325]
  We present EISP, a static analysis framework based on the Large Language Model (LLM) The framework generates a semantic mapping between source code and translated code. EISP connects each pair of sub-code fragments with fine-grained knowledge hints through an AI chain.
  arXiv  Detail & Related papers  (2024-10-30T08:53:33Z)
• Impact of Large Language Models of Code on Fault Localization [2.936007114555107]
  We propose a simple but effective sequence generation approach for fine-tuning large language models of code for FL tasks. Specifically, we fine-tune representative encoder, encoder-decoder, and decoder-based 13 LLMCs for FL tasks. Experimental results show that LLMCs fine-tuned with our approach successfully pinpoint error positions in 50.6%, 64.2%, and 72.3% of 1,291 methods in Defects4J for Top-2/3/5 prediction.
  arXiv  Detail & Related papers  (2024-08-19T02:36:07Z)
• Let the Code LLM Edit Itself When You Edit the Code [50.46536185784169]
  underlinetextbfPositional textbfIntegrity textbfEncoding (PIE)<n>PIE reduces computational overhead by over 85% compared to the standard full recomputation approach.<n>Results demonstrate that PIE reduces computational overhead by over 85% compared to the standard full recomputation approach.
  arXiv  Detail & Related papers  (2024-07-03T14:34:03Z)

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.