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 1–10 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
Turning the Tide: Repository-based Code Reflection—0
CodeAssistBench (CAB): Dataset & Benchmarking for Multi-turn Chat-Based Code Assistance—0
CodeJudgeBench: Benchmarking LLM-as-a-Judge for Coding Tasks—0
Kodezi Chronos: A Debugging-First Language Model for Repository-Scale, Memory-Driven Code UnderstandingCode9
Multilingual Multimodal Software Developer for Code Generation—0
Show:102550
← PrevPage 1 of 170Next →
All datasetsMBPPAPPSCoNaLaDjangoWikiSQLRES-QCodeContestsHumanEvalPECCWebApp1K-ReactCoNaLa-ExtWebApp1k-Duo-React

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1MarianCGAccuracy81.83—Unverified
2TranX + BERT w/minedAccuracy81.03—Unverified
3BERT + TAEAccuracy81.03—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