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. For BFS over adjacency lists with V vertices and E edges, state time complexity and avoid claiming O(V squared) for all graphs.
{
"metrics": {
"accuracy": {
"checks": [
{
"contains": [
"O(V+E)"
]
},
{
"contains": [
"adjacency lists"
]
},
{
"contains": [
"BFS"
]
}
]
},
"fabricated": {
"checks": [
{
"contains": [
"not O(V squared)"
]
},
{
"contains": [
"vertices"
]
},
{
"contains": [
"edges"
]
}
]
}
}
}temperature
0
max_tokens
460
timeout (s)
120
type
scored
file
halluc-complexity-graph-bfs.json