SOTAVerified

Code Generation

Code Generation is an important field to predict explicit code or program structure from multimodal data sources such as incomplete code, programs in another programming language, natural language descriptions or execution examples. Code Generation tools can assist the development of automatic programming tools to improve programming productivity.

Source: Deep Learning for Source Code Modeling and Generation

Image source: Measuring Coding Challenge Competence With APPS

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 1–50 of 1697 papers

TitleStatusHype
CUDA-L1: Improving CUDA Optimization via Contrastive Reinforcement Learning—0
Towards Formal Verification of LLM-Generated Code from Natural Language Prompts—0
MERA Code: A Unified Framework for Evaluating Code Generation Across Tasks—0
Scaling Up RL: Unlocking Diverse Reasoning in LLMs via Prolonged Training—0
The Devil behind the mask: An emergent safety vulnerability of Diffusion LLMsCode2
CodeJudgeBench: Benchmarking LLM-as-a-Judge for Coding Tasks—0
Turning the Tide: Repository-based Code Reflection—0
CodeAssistBench (CAB): Dataset & Benchmarking for Multi-turn Chat-Based Code Assistance—0
Kodezi Chronos: A Debugging-First Language Model for Repository-Scale, Memory-Driven Code UnderstandingCode9
Multilingual Multimodal Software Developer for Code Generation—0
OpenCodeReasoning-II: A Simple Test Time Scaling Approach via Self-Critique—0
Automating MD simulations for Proteins using Large language Models: NAMD-Agent—0
Rethinking Verification for LLM Code Generation: From Generation to TestingCode1
ArtifactsBench: Bridging the Visual-Interactive Gap in LLM Code Generation Evaluation—0
Learn Globally, Speak Locally: Bridging the Gaps in Multilingual Reasoning—0
CoreCodeBench: A Configurable Multi-Scenario Repository-Level BenchmarkCode1
EvoAgentX: An Automated Framework for Evolving Agentic WorkflowsCode7
CORE: Benchmarking LLMs Code Reasoning Capabilities through Static Analysis Tasks—0
LLM-based Realistic Safety-Critical Driving Video Generation—0
A Large Language Model-Empowered Agent for Reliable and Robust Structural Analysis—0
Estimating Correctness Without Oracles in LLM-Based Code GenerationCode0
Where to find Grokking in LLM Pretraining? Monitor Memorization-to-Generalization without Test—0
SACL: Understanding and Combating Textual Bias in Code Retrieval with Semantic-Augmented Reranking and Localization—0
DiffuCoder: Understanding and Improving Masked Diffusion Models for Code GenerationCode4
SV-LLM: An Agentic Approach for SoC Security Verification using Large Language Models—0
Language Modeling by Language ModelsCode2
ReCode: Updating Code API Knowledge with Reinforcement LearningCode1
Why Do Open-Source LLMs Struggle with Data Analysis? A Systematic Empirical Study—0
From Reproduction to Replication: Evaluating Research Agents with Progressive Code MaskingCode0
QHackBench: Benchmarking Large Language Models for Quantum Code Generation Using PennyLane Hackathon Challenges—0
Steering Conceptual Bias via Transformer Latent-Subspace Activation—0
The Debugging Decay Index: Rethinking Debugging Strategies for Code LLMs—0
Use Property-Based Testing to Bridge LLM Code Generation and Validation—0
RoboTwin 2.0: A Scalable Data Generator and Benchmark with Strong Domain Randomization for Robust Bimanual Robotic Manipulation—0
TeXpert: A Multi-Level Benchmark for Evaluating LaTeX Code Generation by LLMsCode1
cAST: Enhancing Code Retrieval-Augmented Generation with Structural Chunking via Abstract Syntax TreeCode2
Xolver: Multi-Agent Reasoning with Holistic Experience Learning Just Like an Olympiad TeamCode1
Massive Supervised Fine-tuning Experiments Reveal How Data, Layer, and Training Factors Shape LLM Alignment Quality—0
Sampling from Your Language Model One Byte at a TimeCode1
Comprehensive Verilog Design Problems: A Next-Generation Benchmark Dataset for Evaluating Large Language Models and Agents on RTL Design and VerificationCode2
How Does LLM Reasoning Work for Code? A Survey and a Call to Action—0
LocationReasoner: Evaluating LLMs on Real-World Site Selection ReasoningCode0
A Technical Study into Small Reasoning Language Models—0
FrontendBench: A Benchmark for Evaluating LLMs on Front-End Development via Automatic Evaluation—0
The Safety Reminder: A Soft Prompt to Reactivate Delayed Safety Awareness in Vision-Language Models—0
Humanity's Last Code Exam: Can Advanced LLMs Conquer Human's Hardest Code Competition?Code2
Structured Program Synthesis using LLMs: Results and Insights from the IPARC Challenge—0
SWE-Bench-CL: Continual Learning for Coding AgentsCode0
code_transformed: The Influence of Large Language Models on Code—0
ReVeal: Self-Evolving Code Agents via Iterative Generation-Verification—0
Show:102550
← PrevPage 1 of 34Next →
All datasetsMBPPAPPSCoNaLaDjangoWikiSQLRES-QCodeContestsHumanEvalPECCWebApp1K-ReactCoNaLa-ExtWebApp1k-Duo-React

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1EG-CFG (DeepSeek-V3-0324)Accuracy96.6—Unverified
2QualityFlow (Sonnet-3.5)Accuracy94.2—Unverified
3o1-mini + MapCoder (Hamming.ai)Accuracy93.2—Unverified
4MGDebugger (DeepSeek-V3-0324)Accuracy92.4—Unverified
5GPT-4 + AgentCoderAccuracy91.8—Unverified
6CodeSim (GPT4o)Accuracy90.7—Unverified
7Jiutian-大模型Accuracy90—Unverified
8GPT-3.5 Turbo (ChatGPT) + AgentCoderAccuracy89.9—Unverified
9MapCoder (GPT-4o)Accuracy89.7—Unverified
10GPT-4 (ChatGPT Plus)Accuracy87.5—Unverified
#ModelMetricClaimedVerifiedStatus
1LPW (GPT-4o)Introductory Pass@187.2—Unverified
2MoTCoder-32B-V1.5Introductory Pass@168.44—Unverified
3MoTCoder-7B-V1.5Introductory Pass@154.26—Unverified
4code-davinci-002 175B (CodeT)Introductory Pass@147.3—Unverified
5deepseek-ai/deepseek-coder-6.7b-instructIntroductory Pass@133.8—Unverified
6code-davinci-002 175BIntroductory Pass@131.92—Unverified
7CodeChain+WizardCoder-15bIntroductory Pass@129.3—Unverified
8WizardCoder-15bIntroductory Pass@126.29—Unverified
9CodeSim (GPT4)Introductory Pass@126.04—Unverified
10AlphaCode 1B Filtered from 50000Competition Pass@any22—Unverified
#ModelMetricClaimedVerifiedStatus
1PanGu-Coder-FT-IBLEU44.32—Unverified
2RoBERTaMarianBLEU35.74—Unverified
3MarianCGBLEU34.43—Unverified
4TranX + BERT w/minedBLEU34.2—Unverified
5BERT + TAEBLEU33.41—Unverified
6BERTMarianBLEU32.46—Unverified
7External Knowledge With API + RerankingBLEU32.26—Unverified
8External Knowledge With APIBLEU30.69—Unverified
9BART W/ MinedBLEU30.55—Unverified
10ELECTRAMarianBLEU30.18—Unverified
#ModelMetricClaimedVerifiedStatus
1MarianCGAccuracy81.83—Unverified
2BERT + TAEAccuracy81.03—Unverified
3TranX + BERT w/minedAccuracy81.03—Unverified
4RerankerAccuracy80.2—Unverified
5LUKEMarianAccuracy78.5—Unverified
6RoBERTaMarianAccuracy77.95—Unverified
7BERTMarianAccuracy76.68—Unverified
8TranxAccuracy73.7—Unverified
9ELECTRAMarianAccuracy65.32—Unverified
10lpn (Ling et al., 2016)Accuracy62.3—Unverified
#ModelMetricClaimedVerifiedStatus
1NL2SQL-RULEExecution Accuracy89.2—Unverified
2TypeSQL+TC (Yu et al., 2018)+Execution Accuracy82.6—Unverified
3TranxExecution Accuracy78.6—Unverified
4STAMP+RL (Sun et al., 2018)+Execution Accuracy74.6—Unverified
5STAMP (Sun et al., 2018)+Execution Accuracy74.4—Unverified
6TypeSQL (Yu et al., 2018)Execution Accuracy73.5—Unverified
7PT-MAML (Huang et al., 2018)Execution Accuracy68—Unverified
8Bidirectional Attention for SQL GenerationExecution Accuracy62.5—Unverified
9Seq2SQL (Zhong et al., 2017)Execution Accuracy59.4—Unverified
10Seq2Seq (Zhong et al., 2017)Execution Accuracy35.9—Unverified
#ModelMetricClaimedVerifiedStatus
1QurrentOS-coder + Claude 3.5 Sonnetpass@158—Unverified
2QurrentOS-coder + GPT-4opass@146—Unverified
3QurrentOS-coder + GPT-4 Turbopass@137—Unverified
4QurrentOS-coder + Claude 3 Opuspass@136—Unverified
5QurrentOS-coder + Gemini 1.5 Propass@130—Unverified
6QurrentOS-coder + GPT-4pass@130—Unverified
7QurrentOS-coder + DeepSeek-Coder-V2pass@129—Unverified
8QurrentOS-coder + Llama 3 70bpass@120—Unverified
9QurrentOS-coder + Qwen-72B-Instructpass@118—Unverified
#ModelMetricClaimedVerifiedStatus
1EG-CFG (DeepSeek-V3-0324)Test Set pass@158.18—Unverified
2LPW (GPT-4o)Test Set pass@134.7—Unverified
3MapCoder (GPT-4)Test Set pass@128.5—Unverified
4CodeSim (GPT4)Test Set pass@128.4—Unverified
5MoTCoder-15BTest Set pass@126.34—Unverified
6MoTCoder-7B-v1.5Test Set pass@120.77—Unverified
7CodeChain + WizardCoder-15BTest Set pass@12.35—Unverified
8WizardCoder-15BTest Set pass@11.11—Unverified
#ModelMetricClaimedVerifiedStatus
1DeepSeek-R1 (MGDebugger)Pass@1100—Unverified
2LLaMA 3Pass@199.4—Unverified
3QualityFlow (Sonnet-3.5)Pass@198.8—Unverified
4Phi-2Pass@198.2—Unverified
5EG-CFG (DeepSeek-V3-0324)Pass@196.95—Unverified
6Mistral 7BPass@193.9—Unverified
7Claude Sonnet 3.5Pass@190.85—Unverified
8L2MAC (GPT-4)Pass@190.2—Unverified
#ModelMetricClaimedVerifiedStatus
1Claude 3 HaikuPass@327.67—Unverified
2GPT-3.5 TurboPass@323.75—Unverified
3codechat-bisonPass@311.39—Unverified
4chat-bisonPass@38.48—Unverified
5Mixtral-8x7B-InstructPass@38.35—Unverified
6Phi-3-mini-128k-instructPass@37.18—Unverified
7WizardLM-2-7BPass@33.72—Unverified
8Llama-3-8B-InstructPass@33.1—Unverified
#ModelMetricClaimedVerifiedStatus
1o1-previewpass@10.95—Unverified
2o1-minipass@10.94—Unverified
3gpt-4o-2024-08-06pass@10.89—Unverified
4claude-3.5-sonnetpass@10.88—Unverified
5deepseek-v2.5pass@10.83—Unverified
6mistral-large-2pass@10.78—Unverified
7deepseek-coder-v2-instructpass@10.7—Unverified
8llama-v3p1-405b-instructpass@10.3—Unverified
#ModelMetricClaimedVerifiedStatus
1BART W/ MinedBLEU35.32—Unverified
2BART BaseBLEU34.35—Unverified
3External Knowledge With API + RerankingBLEU20.54—Unverified
4External Knowledge With APIBLEU20.37—Unverified
5RerankerBLEU19.85—Unverified
6TranXBLEU18.85—Unverified
#ModelMetricClaimedVerifiedStatus
1claude-3-5-sonnetpass@10.68—Unverified
2o1-minipass@10.67—Unverified