cxcscmu/SkillLearnBench - Page 9

image-object-detection

Techniques for image preprocessing and template matching to count objects in images.

Langue du texte source : anglais

video-processing

Tools and techniques for extracting keyframes and processing video files using FFmpeg.

Langue du texte source : anglais

excel-lookup-functions

Covers advanced Excel lookup functions including VLOOKUP, INDEX/MATCH, and XLOOKUP for multi-dimensional data retrieval.

Langue du texte source : anglais

excel-statistical-analysis

Covers descriptive statistics (mean, median, percentiles) and weighted averages using Excel formulas.

Langue du texte source : anglais

run2-advanced-pil-techniques

Advanced PIL/Pillow techniques for depth effects, sophisticated shapes, and professional visual design

Langue du texte source : anglais

run2-brand-color-strategy

Strategic color application following Anthropic brand standards for professional technical materials

Langue du texte source : anglais

run2-professional-technical-layout

Professional technical diagram layout with grid systems, visual hierarchy, and spatial organization

Langue du texte source : anglais

run2-composition-technique

Advanced techniques for composing seven-character regulated verse with thematic coherence

Langue du texte source : anglais

run2-tonal-patterns

Detailed tonal pattern requirements for seven-character regulated verse in modern Mandarin

Langue du texte source : anglais

run2-verification-checklist

Comprehensive checklist to verify a completed seven-character regulated verse meets all requirements

Langue du texte source : anglais

run2-pdf-skill-form-filling

Fill PDF forms using the PDF skill's form-filling scripts and workflow

Langue du texte source : anglais

run2-sc100-form-mapping

SC-100 California Small Claims form field ID mappings and data entry guidelines

Langue du texte source : anglais

run2-clustering-metrics

Compute and aggregate F1 score and delta from clustering evaluation across images

Langue du texte source : anglais

run2-custom-distance-metrics

Implement weighted Euclidean distance metric for DBSCAN with shape parameter

Langue du texte source : anglais

run2-greedy-matching

Greedy one-to-one matching of cluster centroids to expert points with maximum distance constraint

Langue du texte source : anglais

run2-parallel-grid-search

Efficient parallel hyperparameter grid search using joblib with 847 combinations

Langue du texte source : anglais

run2-pareto-frontier

Find Pareto-optimal solutions balancing multiple conflicting objectives with rigorous verification

Langue du texte source : anglais

run2-csv-report-generation-advanced

Advanced CSV security report generation with validation, normalization, and comprehensive error handling

Langue du texte source : anglais

run2-cvss-scoring-strategy

Strategic CVSS score extraction with multi-source lookup, fallback handling, and NVD/GHSA integration

Langue du texte source : anglais

run2-npm-vulnerability-scanning-enhanced

Enhanced npm vulnerability scanning with Trivy supporting multiple severity levels and complete metadata extraction

Langue du texte source : anglais

multipart-geometry-handling

Work with MultiPolygon and MultiLineString geometries from real plate boundary datasets.

Langue du texte source : anglais

optimal-projection-selection

Select appropriate projected CRS based on geographic region for accurate distance calculations.

Langue du texte source : anglais

spatial-analysis-validation

Validate spatial analysis results to ensure correctness of point-in-polygon and distance calculations.

Langue du texte source : anglais

advanced-slack-mining

Mine Slack data for embedded URLs, mentioned products, competitive intelligence, and cross-referencing with external data sources.

Langue du texte source : anglais

document-metadata-extraction

Extract author, reviewer, and feedback information from structured documents and their associated discussions.

Langue du texte source : anglais

multi-source-data-correlation

Correlate and merge data from multiple sources (documents, Slack, meetings, URLs) to answer enterprise questions.

Langue du texte source : anglais

run2-cross-fund-analysis

Find which funds hold a specific stock and rank them by holding value

Langue du texte source : anglais

run2-fund-analysis

Analyze individual fund holdings, AUM, and portfolio composition from 13-F data

Langue du texte source : anglais

run2-quarter-comparison

Compare holdings across quarters to identify increased/decreased investments

Langue du texte source : anglais

run2-sec13f-data-loader

Load and parse SEC 13-F TSV files with proper data type handling and validation

Langue du texte source : anglais

run2-druid-javascript-deserialization-security

Securing Apache Druid JavaScript filters against unknown property injection attacks

Langue du texte source : anglais

run2-jackson-security-patching

Creating and applying security patches for Jackson deserialization vulnerabilities in Java

Langue du texte source : anglais

run2-maven-security-build

Building Apache Druid with Maven while maintaining security patch integrity and skipping unnecessary checks

Langue du texte source : anglais

run2-data-aggregation

Aggregate and validate GitHub API data for metrics calculation with comprehensive error handling.

Langue du texte source : anglais

run2-github-rest-api

Query GitHub REST API with pagination and date filtering for repository data (PRs, issues).

Langue du texte source : anglais

run2-json-validation

Validate, format, and write JSON output with schema enforcement and error handling.

Langue du texte source : anglais

run2-simpo-loss-detailed-implementation

Complete SimPO loss implementation with length-normalized rewards and target margin

Langue du texte source : anglais

run2-simpo-testing-and-verification

Testing SimPO loss function with fixed tensors and output verification

Langue du texte source : anglais

run2-simpo-trainer-initialization

Proper initialization of SimPOTrainer with model loading and args setup

Langue du texte source : anglais

run2-placeholder-replacement

Robust pattern-based template processing with conditionals, data validation, and comprehensive placeholder replacement.

Langue du texte source : anglais