25,000 items generated on 2026-01-04 by WithinUsAI. The dataset likely contains 8,000 verification tasks, 9,000 quantitative reasoning problems, and 8,000 definition refreshers.
Use Cases
- Train models for fact verification based on TRUE/FALSE + correction tasks.
- Develop models for structured reasoning based on quantitative/structured reasoning problems.
- Create educational refresher tools based on definitions / micro-refreshers.
- Benchmark AI performance on psychology-related knowledge tasks.
Strengths
- 25,000 total items.
- 8,000 verification tasks with corrections.
- 9,000 self-contained quantitative reasoning problems.
Limitations
- Description metadata is limited; actual data quality requires manual inspection after download.
- Column-level documentation is absent; field semantics must be inferred after download.
- Row count is unknown, which may limit suitability assessment.
Provenance
- Source
- WithinUsAI
- Collection Method
- Generated.
- Freshness
- Last updated 2026-01-04 10:29:46; freshness should be verified.