15 datasets found

Wizardcoder: Empowering code large language models with evol-instruct

Magicoder: Source code is all you need

Instruction-finetuned code language models have shown promise in various programming tasks. They are trained, using a language modeling objective, on natural language...

The dataset used in this paper is the MBPP dataset, which contains code snippets and their corresponding test cases.

The APPS dataset is a collection of programming problems used to evaluate the performance of code generation models.

The paper presents the results of the OpenAI Codex evaluation on generating Python code.

A set of 50 prompts to evaluate execution-based evaluation for NL2Bash task

Code completion is considered as an essential feature towards efficient software development in modern Integrated Development Environments (IDEs).

CodeUltraFeedback is a preference dataset of 10,000 complex instructions to tune and align LLMs to coding preferences through AI feedback.

The dataset used in the paper for training and testing the DPO and PPO models.

The dataset used in the paper is a code generation benchmark, consisting of 164 function declarations alongside their documentation, 500 test examples, each one is an...

Code generation models fail at completing code with potential bugs.

SLTrans is a parallel dataset consisting of nearly 4M pairs of self-contained source code and corresponding LLVM IR.

Evol-Instruct-Code-80k is a dataset for evaluating the performance of code generation models.

The dataset used in this paper is the CodeT5 dataset, which is a code generation large language model.