Back to evaluations
Public evaluation
nuclear_verdict_prediction
Evaluate verdict size prediction accuracy and nuclear risk probability calibration.
Evaluation type
task based
Challenge
Nuclear Verdict Litigation Risk Forecaster with Google ADK and Daytona
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
nuclear_verdict_prediction
Analyze trial case brief and estimate probability of verdict exceeding $10M.
Input format
JSON containing case details, venue, plaintiff claims, and injury severity
Output format
JSON with nuclear risk probability and risk factor breakdown