Image de couverture de livre

Not All Code Is Equal: A Data-Centric Study of Code Complexity and LLM Reasoning

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Twist, Lukas, Yang, Shu, Yan, Hanqi, Gong, Jingzhi, Wang, Di, Yannakoudakis, Helen, Zhang, Jie M.
Format:	Preprint
Publié:	2026
Sujets:	Machine Learning
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Reasoning-Trace Collapse: Evaluating the Loss of Explicit Reasoning During Fine-Tuning
par: Twist, Lukas, et autres
Publié: (2026)

Not All LLM Reasoners Are Created Equal
par: Hosseini, Arian, et autres
Publié: (2024)

A Function-Centric Perspective on Flat and Sharp Minima
par: Mason-Williams, Israel, et autres
Publié: (2025)

Not All Negative Samples Are Equal: LLMs Learn Better from Plausible Reasoning
par: Di, Zixiang, et autres
Publié: (2026)

Not All LLM-Generated Data Are Equal: Rethinking Data Weighting in Text Classification
par: Kuo, Hsun-Yu, et autres
Publié: (2024)

Learning New Tasks from a Few Examples with Soft-Label Prototypes
par: Singh, Avyav Kumar, et autres
Publié: (2022)

MedAgentGym: A Scalable Agentic Training Environment for Code-Centric Reasoning in Biomedical Data Science
par: Xu, Ran, et autres
Publié: (2025)

Ensemble Learning for Large Language Models in Text and Code Generation: A Survey
par: Ashiga, Mari, et autres
Publié: (2025)

A Functional Perspective on Knowledge Distillation in Neural Networks
par: Mason-Williams, Israel, et autres
Publié: (2025)

Not All Bits Are Equal: Scale-Dependent Memory Optimization Strategies for Reasoning Models
par: Kim, Junhyuck, et autres
Publié: (2025)

All Nodes are created Not Equal: Node-Specific Layer Aggregation and Filtration for GNN
par: Wang, Shilong, et autres
Publié: (2024)

From Data-Centric to Sample-Centric: Enhancing LLM Reasoning via Progressive Optimization
par: Chen, Xinjie, et autres
Publié: (2025)

The Impact of Off-Policy Training Data on Probe Generalisation
par: Kirch, Nathalie, et autres
Publié: (2025)

A Study of LLMs' Preferences for Libraries and Programming Languages
par: Twist, Lukas, et autres
Publié: (2025)

QiMeng-CodeV-R1: Reasoning-Enhanced Verilog Generation
par: Zhu, Yaoyu, et autres
Publié: (2025)

Code to Think, Think to Code: A Survey on Code-Enhanced Reasoning and Reasoning-Driven Code Intelligence in LLMs
par: Yang, Dayu, et autres
Publié: (2025)

Let the Code LLM Edit Itself When You Edit the Code
par: He, Zhenyu, et autres
Publié: (2024)

Selective Matching Losses -- Not All Scores Are Created Equal
par: Shamir, Gil I., et autres
Publié: (2025)

DAJ: Data-Reweighted LLM Judge for Test-Time Scaling in Code Generation
par: Qin, Peijia, et autres
Publié: (2026)

Deep Configuration Performance Learning: A Systematic Survey and Taxonomy
par: Gong, Jingzhi, et autres
Publié: (2024)

PurpCode: Reasoning for Safer Code Generation
par: Liu, Jiawei, et autres
Publié: (2025)

Code Less, Align More: Efficient LLM Fine-tuning for Code Generation with Data Pruning
par: Tsai, Yun-Da, et autres
Publié: (2024)

Not All Invariants Are Equal: Curating Training Data to Accelerate Program Verification with SLMs
par: Pinto, Ido, et autres
Publié: (2026)

Not All Errors Are Equal: Investigation of Speech Recognition Errors in Alzheimer's Disease Detection
par: Kang, Jiawen, et autres
Publié: (2024)

Predicting Configuration Performance in Multiple Environments with Sequential Meta-learning
par: Gong, Jingzhi, et autres
Publié: (2024)

ReST-RL: Achieving Accurate Code Reasoning of LLMs with Optimized Self-Training and Decoding
par: Zhoubian, Sining, et autres
Publié: (2025)

All Code, No Thought: Current Language Models Struggle to Reason in Ciphered Language
par: Guo, Shiyuan, et autres
Publié: (2025)

CodeIF: Benchmarking the Instruction-Following Capabilities of Large Language Models for Code Generation
par: Yan, Kaiwen, et autres
Publié: (2025)

A (More) Realistic Evaluation Setup for Generalisation of Community Models on Malicious Content Detection
par: Verhoeven, Ivo, et autres
Publié: (2024)

CodeARC: Benchmarking Reasoning Capabilities of LLM Agents for Inductive Program Synthesis
par: Wei, Anjiang, et autres
Publié: (2025)

Not All Frequencies Are Created Equal:Towards a Dynamic Fusion of Frequencies in Time-Series Forecasting
par: Zhang, Xingyu, et autres
Publié: (2024)

Unified Error Correction Code Transformer with Low Complexity
par: Yan, Yongli, et autres
Publié: (2024)

Code-Centric Detection of Vulnerability-Fixing Commits: A Unified Benchmark and Empirical Study
par: Loose, Nils, et autres
Publié: (2026)

CodeScaler: Scaling Code LLM Training and Test-Time Inference via Reward Models
par: Zhu, Xiao, et autres
Publié: (2026)

Teaching LLM to Reason: Reinforcement Learning from Algorithmic Problems without Code
par: Bao, Keqin, et autres
Publié: (2025)

A Systematic Study of Code Obfuscation Against LLM-based Vulnerability Detection
par: Li, Xiao, et autres
Publié: (2025)

Not All Federated Learning Algorithms Are Created Equal: A Performance Evaluation Study
par: Baumgart, Gustav A., et autres
Publié: (2024)

The Quest for Efficient Reasoning: A Data-Centric Benchmark to CoT Distillation
par: Zhang, Ruichen, et autres
Publié: (2025)

Which Data Attributes Stimulate Math and Code Reasoning? An Investigation via Influence Functions
par: Kou, Siqi, et autres
Publié: (2025)

Towards Effective Code-Integrated Reasoning
par: Bai, Fei, et autres
Publié: (2025)