Back to evaluations
Public evaluation

Content Neutrality Analysis and Revision Suggestion

The evaluation will focus on the system's ability to accurately detect bias, verify facts, generate meaningful neutrality scores, and propose effective revisions. Agent collaboration via A2A and the use of hybrid reasoning will also be assessed.

Evaluation type
task based
Challenge
Neutrality Score for Bias Detection & Fact-Checking
Difficulty
Advanced
Rigor
Unspecified

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

Content Neutrality Analysis and Revision Suggestion

Analyze a provided text passage for neutrality, identify specific biased statements, verify key facts, assign an overall neutrality score, and suggest revisions to improve neutrality. The system should demonstrate A2A communication for cross-verification.

Input format

{ "text_passage": "string", "context_sources": [{ "title": "string", "url": "string", "content": "string" }] }

Output format

{ "overall_neutrality_score": "float (0.0 - 1.0)", "biased_statements_detected": [{"statement": "string", "bias_type": "string", "reasoning": "string", "suggested_revision": "string"}], "factual_errors_detected": [{"statement": "string", "error_details": "string", "corrected_info": "string", "verified_source": "string"}], "agent_collaboration_log": [{"agent_id": "string", "action": "string", "message_to_agent": "string", "response_from_agent": "string"}] }