Menu
O SkillsMP coletou 77 skills de yogsoth-ai/knowledge-acquisition. Abra uma skill para revisar a origem e os detalhes.
Mostrando 40 de 77 skills coletadas.
SOTA Performance Baseline Campaign — 5 strategies for systematically collecting, standardizing, and analyzing performance data across methods. Produces standardized comparison tables, progress curves, and headroom analysis.
Idioma do texto original: inglês
Assess systematic biases in the evidence body — publication bias, reporting bias, and selective outcome reporting. Budget: 40 studies, 40 effect sizes, 40 web searches.
Idioma do texto original: inglês
Track evidence accumulation over time — cumulative meta-analysis protocol design. Budget: 40 studies, 40 effect sizes, 30 web searches.
Idioma do texto original: inglês
Design structured data extraction form for systematic meta-analysis data collection
Idioma do texto original: inglês
Systematically extract effect sizes and conditions from papers for meta-analytic synthesis
Idioma do texto original: inglês
Determine effect size types and calculation methods for meta-analytic synthesis
Idioma do texto original: inglês
Build evidence network graph for network meta-analysis — nodes, edges, geometry assessment
Idioma do texto original: inglês
Plan the statistical synthesis approach — model selection, heterogeneity strategy, and reporting
Idioma do texto original: inglês
Explain why different studies reach different conclusions — heterogeneity investigation protocol. Budget: 30 studies, 30 effect sizes, 50 web searches.
Idioma do texto original: inglês
Identify and classify sources of between-study heterogeneity (clinical, methodological, statistical)
Idioma do texto original: inglês
Define inclusion/exclusion criteria for systematic study selection in meta-analysis
Idioma do texto original: inglês
Cross-Study Statistical Synthesis Campaign — 5 strategies for systematic collection and methodological planning of multi-study evidence synthesis. Covers pairwise, network, cumulative meta-analysis, heterogeneity investigation, and bias detection. Stops at…
Idioma do texto original: inglês
Produce final meta-analysis protocol document assembling all planning outputs into PRISMA-compliant protocol
Idioma do texto original: inglês
Compare N methods simultaneously including indirect evidence — network meta-analysis protocol design. Budget: 50 studies, 80 effect sizes, 60 web searches.
Idioma do texto original: inglês
Compare two methods across multiple studies — paired meta-analysis protocol design. Budget: 30 studies, 30 effect sizes, 40 web searches.
Idioma do texto original: inglês
Construct PICO/PECO framework for the meta-analysis research question
Idioma do texto original: inglês
Plan funnel plots, Egger's test, trim-and-fill, p-curve, and selection model analyses for publication bias
Idioma do texto original: inglês
Methodological quality and bias risk assessment of included studies using validated tools
Idioma do texto original: inglês
Assess methodological bias using RoB2, PROBAST, or QUADAS-2 validated tools
Idioma do texto original: inglês
Design leave-one-out, influence diagnostics, subgroup analyses, and robustness checks
Idioma do texto original: inglês
Detect annotation artifacts and shortcuts in benchmarks
Idioma do texto original: inglês
Evaluation Methodology Archaeology Campaign — 5 strategies for systematic analysis of AI/ML benchmarks, metrics, and leaderboards. Reveals construct validity issues, saturation, data contamination, and evaluation protocol inconsistencies.
Idioma do texto original: inglês
Systematic quality assessment using BetterBench 46-criterion framework — 5 benchmarks, 30 papers, 40 web searches
Idioma do texto original: inglês
Identify and catalog all relevant benchmarks in target domain
Idioma do texto original: inglês
Produce final structured audit report
Idioma do texto original: inglês
Build capability taxonomy, map existing benchmark coverage
Idioma do texto original: inglês
Evaluate whether benchmark measures its claimed capability
Idioma do texto original: inglês
Detect train-test data leakage and memorization artifacts
Idioma do texto original: inglês
Map evaluation coverage, identify untested capability dimensions — 20 benchmarks, 30 papers, 50 web searches
Idioma do texto original: inglês
Assess documentation completeness against BetterBench/Datasheets standards
Idioma do texto original: inglês
Compare implementation differences of same benchmark across papers
Idioma do texto original: inglês
Analyze leaderboard score distributions, compression, selective reporting
Idioma do texto original: inglês
Decompose composite metrics into constituent signals, analyze polarity and ceiling effects
Idioma do texto original: inglês
Extract evaluation protocol parameters from papers
Idioma do texto original: inglês
Analyze evaluation protocol differences across papers for same benchmark — 5 benchmarks, 60 papers, 30 web searches
Idioma do texto original: inglês
Track score trajectories, detect saturation/failure points — 15 benchmarks, 50 papers, 60 web searches
Idioma do texto original: inglês
Collect historical scores, fit saturation curves, detect inflection points
Idioma do texto original: inglês
Challenge construct validity — does benchmark measure claimed capability? — 3 benchmarks, 40 papers, 30 web searches
Idioma do texto original: inglês
Patent claim syntax parsing — independent/dependent relationships and element extraction
Idioma do texto original: inglês
Determine patent legal status — active, expired, pending, lapsed, or revoked
Idioma do texto original: inglês