From Crowdsourced Data to High-Quality Benchmarks: Arena-Hard and BenchBuilder Pipeline.
Tianle LiWei-Lin ChiangEvan FrickLisa DunlapTianhao WuBanghua ZhuJoseph E. GonzalezIon StoicaPublished in: CoRR (2024)
Keyphrases
- high quality
- data sets
- database
- low quality
- data analysis
- prior knowledge
- data processing
- ground truth
- data structure
- data objects
- raw data
- sensor data
- data collection
- input data
- knowledge discovery
- training data
- image quality
- data sources
- computer systems
- statistical analysis
- synthetic data
- relational databases
- data mining algorithms
- high dimensional
- original data
- complex data
- machine learning