Back to evaluations
Public evaluation

Plagiarism Audit

Evaluation of agent performance in detecting semantic similarity and authorship irregularities

Evaluation type
task based
Challenge
Multi-Agent Editorial Integrity Suite with CrewAI
Difficulty
Advanced
Rigor
Unspecified
Open linked challenge

Evaluation overview

How the linked challenge is judged: tasks, benchmarks, and criteria count.

Tasks
1
Benchmarks
0
Criteria
0

Task templates

Inputs and expected outputs.

Task 1

Plagiarism Audit

Determine if input matches known content patterns

Input format

Document raw text

Output format

Audit report JSON