Test fixture
Grounded answers under adversarial or missing-context prompts.
The model receives the prompt (and optional system message). The run uses scorer rubric_json_metrics with the JSON configuration below. Pass/fail and partial credit are determined entirely by that scorer against the model output; no human grading.
Return JSON only with keys answer, supported_facts, uncertainty. A recursion computes fib(n) with memoization. State expected time complexity and avoid claiming exponential after memoization.
{
"metrics": {
"accuracy": {
"checks": [
{
"contains": [
"O(n)"
]
},
{
"contains": [
"memoization"
]
},
{
"contains": [
"fib"
]
}
]
},
"fabricated": {
"checks": [
{
"contains": [
"not exponential"
]
},
{
"contains": [
"recursive"
]
},
{
"contains": [
"complexity"
]
}
]
}
}
}temperature
0
max_tokens
460
timeout (s)
120
type
scored
file
halluc-complexity-recursive-memo.json