Copertina

Mechanistic Interpretability of Code Correctness in LLMs via Sparse Autoencoders

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Tahimic, Kriz, Cheng, Charibeth
Natura:	Preprint
Pubblicazione:	2025
Soggetti:	Software Engineering Machine Learning
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

Unsupervised Evaluation of Code LLMs with Round-Trip Correctness
di: Allamanis, Miltiadis, et al.
Pubblicazione: (2024)

On LLMs' Internal Representation of Code Correctness
di: Ribeiro, Francisco, et al.
Pubblicazione: (2025)

Enabling Global, Human-Centered Explanations for LLMs:From Tokens to Interpretable Code and Test Generation
di: Khati, Dipin, et al.
Pubblicazione: (2025)

Protocode: Prototype-Driven Interpretability for Code Generation in LLMs
di: Bodla, Krishna Vamshi, et al.
Pubblicazione: (2025)

Calibration and Correctness of Language Models for Code
di: Spiess, Claudio, et al.
Pubblicazione: (2024)

Are Sparse Autoencoders Useful for Java Function Bug Detection?
di: Melo, Rui, et al.
Pubblicazione: (2025)

Teaching Code Refactoring Using LLMs
di: Khairnar, Anshul, et al.
Pubblicazione: (2025)

Towards Verified Code Reasoning by LLMs
di: Sistla, Meghana, et al.
Pubblicazione: (2025)

Towards More Trustworthy and Interpretable LLMs for Code through Syntax-Grounded Explanations
di: Palacio, David N., et al.
Pubblicazione: (2024)

OSS-Bench: Benchmark Generator for Coding LLMs
di: Jiang, Yuancheng, et al.
Pubblicazione: (2025)

Understanding Robustness of Model Editing in Code LLMs
di: Chhetri, Vinaik, et al.
Pubblicazione: (2025)

Ensuring Functional Correctness of Large Code Models with Selective Generation
di: Jeong, Jaewoo, et al.
Pubblicazione: (2025)

Correctness-Guaranteed Code Generation via Constrained Decoding
di: Li, Lingxiao, et al.
Pubblicazione: (2025)

TritonRL: Training LLMs to Think and Code Triton Without Cheating
di: Woo, Jiin, et al.
Pubblicazione: (2025)

Towards Effectively Leveraging Execution Traces for Program Repair with Code LLMs
di: Haque, Mirazul, et al.
Pubblicazione: (2025)

Correctness Assessment of Code Generated by Large Language Models Using Internal Representations
di: Bui, Tuan-Dung, et al.
Pubblicazione: (2025)

K-ASTRO: Structure-Aware Adaptation of LLMs for Code Vulnerability Detection
di: Zhang, Yifan, et al.
Pubblicazione: (2022)

ReVeal: Self-Evolving Code Agents via Reliable Self-Verification
di: Jin, Yiyang, et al.
Pubblicazione: (2025)

Where Do LLMs Still Struggle? An In-Depth Analysis of Code Generation Benchmarks
di: Sharifloo, Amir Molzam, et al.
Pubblicazione: (2025)

Leveraging LLMs for Legacy Code Modernization: Challenges and Opportunities for LLM-Generated Documentation
di: Diggs, Colin, et al.
Pubblicazione: (2024)

FunPRM: Function-as-Step Process Reward Model with Meta Reward Correction for Code Generation
di: Zhang, Ruiyi, et al.
Pubblicazione: (2026)

Prompt-Driven Code Summarization: A Systematic Literature Review
di: Farjana, Afia, et al.
Pubblicazione: (2026)

Can LLMs Find Bugs in Code? An Evaluation from Beginner Errors to Security Vulnerabilities in Python and C++
di: Mhatre, Akshay, et al.
Pubblicazione: (2025)

Renaissance of Literate Programming in the Era of LLMs: Enhancing LLM-Based Code Generation in Large-Scale Projects
di: Zhang, Wuyang, et al.
Pubblicazione: (2024)

PromSec: Prompt Optimization for Secure Generation of Functional Source Code with Large Language Models (LLMs)
di: Nazzal, Mahmoud, et al.
Pubblicazione: (2024)

Operational Robustness of LLMs on Code Generation
di: Paul, Debalina Ghosh, et al.
Pubblicazione: (2026)

Verifier-Guided Code Translation via Meta-Step Decoding
di: Zhou, Tianyang, et al.
Pubblicazione: (2026)

Understanding Formal Reasoning Failures in LLMs as Abstract Interpreters
di: Mitchell, Jacqueline L., et al.
Pubblicazione: (2025)

CodeTaste: Can LLMs Generate Human-Level Code Refactorings?
di: Thillen, Alex, et al.
Pubblicazione: (2026)

Evaluating the Use of LLMs for Documentation to Code Traceability
di: Alor, Ebube, et al.
Pubblicazione: (2025)

CONCUR: Benchmarking LLMs for Concurrent Code Generation
di: Huang, Jue, et al.
Pubblicazione: (2026)

StackEval: Benchmarking LLMs in Coding Assistance
di: Shah, Nidhish, et al.
Pubblicazione: (2024)

QiMeng-PRepair: Precise Code Repair via Edit-Aware Reward Optimization
di: Ke, Changxin, et al.
Pubblicazione: (2026)

LLMs in Coding and their Impact on the Commercial Software Engineering Landscape
di: Belozerov, Vladislav, et al.
Pubblicazione: (2025)

The Struggles of LLMs in Cross-lingual Code Clone Detection
di: Moumoula, Micheline Bénédicte, et al.
Pubblicazione: (2024)

CodeTool: Enhancing Programmatic Tool Invocation of LLMs via Process Supervision
di: Lu, Yifei, et al.
Pubblicazione: (2025)

CodeSAM: Source Code Representation Learning by Infusing Self-Attention with Multi-Code-View Graphs
di: Mathai, Alex, et al.
Pubblicazione: (2024)

SemRep: Generative Code Representation Learning with Code Transformations
di: Li, Weichen, et al.
Pubblicazione: (2026)

LeetCodeDataset: A Temporal Dataset for Robust Evaluation and Efficient Training of Code LLMs
di: Xia, Yunhui, et al.
Pubblicazione: (2025)

RocketPPA: Code-Level Power, Performance, and Area Prediction via LLM and Mixture of Experts
di: Abdollahi, Armin, et al.
Pubblicazione: (2025)