Image de couverture de livre

How Robustly do LLMs Understand Execution Semantics?

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Spiess, Claudio, Devanbu, Prem, Barr, Earl T.
Format:	Preprint
Publié:	2026
Sujets:	Software Engineering Artificial Intelligence Machine Learning
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

On LLMs' Internal Representation of Code Correctness
par: Ribeiro, Francisco, et autres
Publié: (2025)

Model See, Model Do? Exposure-Aware Evaluation of Bug-vs-Fix Preference in Code LLMs
par: Al-Kaswan, Ali, et autres
Publié: (2026)

Automatic Semantic Augmentation of Language Model Prompts (for Code Summarization)
par: Ahmed, Toufique, et autres
Publié: (2023)

Localized Calibrated Uncertainty in Code Language Models
par: Gros, David, et autres
Publié: (2025)

Does In-IDE Calibration of Large Language Models work at Scale?
par: Koohestani, Roham, et autres
Publié: (2025)

Investigating Autonomous Agent Contributions in the Wild: Activity Patterns and Code Change over Time
par: Popescu, Razvan Mihai, et autres
Publié: (2026)

CRUXEval: A Benchmark for Code Reasoning, Understanding and Execution
par: Gu, Alex, et autres
Publié: (2024)

Semantic Voting: Execution-Grounded Consensus for LLM Code Generation
par: Jiang, Shan, et autres
Publié: (2026)

Operational Robustness of LLMs on Code Generation
par: Paul, Debalina Ghosh, et autres
Publié: (2026)

Calibration and Correctness of Language Models for Code
par: Spiess, Claudio, et autres
Publié: (2024)

Towards Understanding What Code Language Models Learned
par: Ahmed, Toufique, et autres
Publié: (2023)

CAPE: Capability Achievement via Policy Execution
par: Ball, David
Publié: (2025)

How Robust are LLM-Generated Library Imports? An Empirical Study using Stack Overflow
par: Latendresse, Jasmine, et autres
Publié: (2025)

Automatic Generation of Executable BPMN Models from Medical Guidelines
par: Sekar, Praveen Kumar Menaka, et autres
Publié: (2026)

Mage: Multi-Axis Evaluation of LLM-Generated Executable Game Scenes Beyond Compile-Pass Rate
par: Liu, Hugh Xuechen, et autres
Publié: (2026)

Feedback Over Form: Why Execution Feedback Matters More Than Pipeline Topology in 1-3B Code Generation
par: McAndrews, Charles Junichi
Publié: (2026)

RepairAgent: An Autonomous, LLM-Based Agent for Program Repair
par: Bouzenia, Islem, et autres
Publié: (2024)

Understanding LLM-Driven Test Oracle Generation
par: Bodicoat, Adam, et autres
Publié: (2026)

SPELL: Synthesis of Programmatic Edits using LLMs
par: Ramos, Daniel, et autres
Publié: (2026)

Evaluating the Use of LLMs for Documentation to Code Traceability
par: Alor, Ebube, et autres
Publié: (2025)

Methodological Framework for Quantifying Semantic Test Coverage in RAG Systems
par: Broestl, Noah, et autres
Publié: (2025)

DRAFT-ing Architectural Design Decisions using LLMs
par: Dhar, Rudra, et autres
Publié: (2025)

LLMs in Coding and their Impact on the Commercial Software Engineering Landscape
par: Belozerov, Vladislav, et autres
Publié: (2025)

Protocode: Prototype-Driven Interpretability for Code Generation in LLMs
par: Bodla, Krishna Vamshi, et autres
Publié: (2025)

The Struggles of LLMs in Cross-lingual Code Clone Detection
par: Moumoula, Micheline Bénédicte, et autres
Publié: (2024)

Machine Learning Robustness: A Primer
par: Braiek, Houssem Ben, et autres
Publié: (2024)

On Wasted Contributions: Understanding the Dynamics of Contributor-Abandoned Pull Requests
par: Khatoonabadi, SayedHassan, et autres
Publié: (2021)

Automating Code Adaptation for MLOps -- A Benchmarking Study on LLMs
par: Patel, Harsh, et autres
Publié: (2024)

Drawing Pandas: A Benchmark for LLMs in Generating Plotting Code
par: Galimzyanov, Timur, et autres
Publié: (2024)

A Regression Framework for Understanding Prompt Component Impact on LLM Performance
par: Lauziere, Andrew, et autres
Publié: (2026)

An Empirical Evaluation of Locally Deployed LLMs for Bug Detection in Python Code
par: Vulićević, Jelena Ilić
Publié: (2026)

LGMT: Logic-Grounded Metamorphic Testing for Evaluating the Reasoning Reliability of LLMs
par: Zhou, Zenghui, et autres
Publié: (2026)

CodeTaste: Can LLMs Generate Human-Level Code Refactorings?
par: Thillen, Alex, et autres
Publié: (2026)

SnipGen: A Mining Repository Framework for Evaluating LLMs for Code
par: Rodriguez-Cardenas, Daniel, et autres
Publié: (2025)

LiCoEval: Evaluating LLMs on License Compliance in Code Generation
par: Xu, Weiwei, et autres
Publié: (2024)

Free and Customizable Code Documentation with LLMs: A Fine-Tuning Approach
par: Chakrabarty, Sayak, et autres
Publié: (2024)

Can LLMs Generate Architectural Design Decisions? -An Exploratory Empirical study
par: Dhar, Rudra, et autres
Publié: (2024)

Real Faults in Deep Learning Fault Benchmarks: How Real Are They?
par: Jahangirova, Gunel, et autres
Publié: (2024)

Cross-Architecture Model Diffing with Crosscoders: Unsupervised Discovery of Differences Between LLMs
par: Jiralerspong, Thomas, et autres
Publié: (2026)

Towards More Trustworthy and Interpretable LLMs for Code through Syntax-Grounded Explanations
par: Palacio, David N., et autres
Publié: (2024)