Brick Complexity Extractor: 76,831 User Queries Labeled by Complexity

76,831 user queries annotated with complexity labels (easy, medium, or hard) to indicate the cognitive effort required for an answer. The dataset was created by regolo.ai to train a LoRA adapter for the Brick Semantic Router. It was last updated on Hugging Face in April 2026.

Use Cases

• Training classifiers for query complexity based on the provided three-tier labeling scheme
• Benchmarking LLM routing systems based on the labeled difficulty of user queries
• Analyzing the distribution of query complexity in a large corpus of user interactions

Strengths

• 76,831 labeled queries provide a substantial corpus for model training
• Complexity labels (easy, medium, hard) offer a structured target for classification tasks

Limitations

• Column-level documentation is absent; field semantics must be inferred after download
• Row count is unknown, which may limit suitability assessment

Provenance

Source

regolo.ai

Collection Method

Created to train the Brick Complexity Extractor, a LoRA adapter for the Brick Semantic Router

Time Range

null

Freshness

Last updated 2026-04-12 13:28:18; freshness should be verified

Geography

null