Quantifying Contamination in Evaluating Code Generation Capabilities of Language Models.
Martin RiddellAnsong NiArman CohanPublished in: CoRR (2024)
Keyphrases
- language model
- code generation
- language modeling
- application development
- n gram
- code generator
- retrieval model
- software development
- probabilistic model
- document retrieval
- query expansion
- information retrieval
- smoothing methods
- speech recognition
- modeling language
- rapid prototyping
- formal specification
- statistical language models
- test collection
- language modelling
- context sensitive
- translation model
- model driven
- software reuse
- design patterns
- language models for information retrieval
- document ranking
- pseudo relevance feedback
- design tools
- software components
- process model
- relational databases