RaindropA practical guide to choosing decision modelsGitHub ↗

Similar distractors · Add fewer similar but irrelevant items variant

Original question · Other versions of this question · All variants

This version keeps the original reference answer. Its context, question instructions or answer choices were changed as described below.

Changed fields: context.

How the edits were constructed

Context

[{"input": "[6, 59]", "output": "[98, 6, 59, 37]"}, {"input": "[1, 35, 4, 3, 9, 51, 81, 48, 7, 33]", "output": "[98, 1, 35, 4, 3, 9, 51, 81, 48, 7, 33, 37]"}, {"input": "[9, 73, 17, 19, 0, 26, 90, 85, 2, 5]", "output": "[98, 9, 73, 17, 19, 0, 26, 90, 85, 2, 5, 37]"}, {"input": "[24, 0, 45, 7, 43, 71, 8, 33, 50]", "output": "[98, 24, 0, 45, 7, 43, 71, 8, 33, 50, 37]"}, {"input": "[]", "output": "[98, 37]"}, {"input": "[7, 78, 1, 38, 80, 66]", "output": "[98, 7, 78, 1, 38, 80, 66, 37]"}, {"input": "[7, 86, 28, 75, 21, 0, 50, 98, 8, 22]", "output": "[98, 7, 86, 28, 75, 21, 0, 50, 98, 8, 22, 37]"}, {"input": "[30, 77, 67, 2, 0, 3, 94, 6]", "output": "[98, 30, 77, 67, 2, 0, 3, 94, 6, 37]"}]

[{"input": "[59, 6]"}]

Question

Apply the same inferred rule to this new input: [73, 3, 58, 93, 6, 44]

Answer choices

  • A. [98, 73, 3, 58, 93, 6, 44, 37]
  • B. [3, 58, 93, 6, 44, 3, 58, 93]
  • C. [73, 3, 58, 93, 6, 44, 73, 3]
  • D. [3, 58, 93, 6, 44, 73, 3, 58]

Reference answer: A.

Exact evaluated input as JSON
{
  "state": "[{\"input\": \"[6, 59]\", \"output\": \"[98, 6, 59, 37]\"}, {\"input\": \"[1, 35, 4, 3, 9, 51, 81, 48, 7, 33]\", \"output\": \"[98, 1, 35, 4, 3, 9, 51, 81, 48, 7, 33, 37]\"}, {\"input\": \"[9, 73, 17, 19, 0, 26, 90, 85, 2, 5]\", \"output\": \"[98, 9, 73, 17, 19, 0, 26, 90, 85, 2, 5, 37]\"}, {\"input\": \"[24, 0, 45, 7, 43, 71, 8, 33, 50]\", \"output\": \"[98, 24, 0, 45, 7, 43, 71, 8, 33, 50, 37]\"}, {\"input\": \"[]\", \"output\": \"[98, 37]\"}, {\"input\": \"[7, 78, 1, 38, 80, 66]\", \"output\": \"[98, 7, 78, 1, 38, 80, 66, 37]\"}, {\"input\": \"[7, 86, 28, 75, 21, 0, 50, 98, 8, 22]\", \"output\": \"[98, 7, 86, 28, 75, 21, 0, 50, 98, 8, 22, 37]\"}, {\"input\": \"[30, 77, 67, 2, 0, 3, 94, 6]\", \"output\": \"[98, 30, 77, 67, 2, 0, 3, 94, 6, 37]\"}]\n\n[{\"input\": \"[59, 6]\"}]",
  "question": "Apply the same inferred rule to this new input: [73, 3, 58, 93, 6, 44]",
  "choices": {
    "A": "[98, 73, 3, 58, 93, 6, 44, 37]",
    "B": "[3, 58, 93, 6, 44, 3, 58, 93]",
    "C": "[73, 3, 58, 93, 6, 44, 73, 3]",
    "D": "[3, 58, 93, 6, 44, 73, 3, 58]"
  }
}

Recorded model answers

The original column uses the control response paired with this variant. Missing responses and inputs outside a model’s limit remain visible.

Model Original answer Variant answer Change
GLiDE v1.2 A — [98, 73, 3, 58, 93, 6, 44, 37] A — [98, 73, 3, 58, 93, 6, 44, 37] Same answer
Decider v1.1 A — [98, 73, 3, 58, 93, 6, 44, 37] A — [98, 73, 3, 58, 93, 6, 44, 37] Same answer
Jev A — [98, 73, 3, 58, 93, 6, 44, 37] A — [98, 73, 3, 58, 93, 6, 44, 37] Same answer
Microsoft Decision-1 A — [98, 73, 3, 58, 93, 6, 44, 37] A — [98, 73, 3, 58, 93, 6, 44, 37] Same answer
Clef A — [98, 73, 3, 58, 93, 6, 44, 37] A — [98, 73, 3, 58, 93, 6, 44, 37] Same answer
Luna Decisions A — [98, 73, 3, 58, 93, 6, 44, 37] A — [98, 73, 3, 58, 93, 6, 44, 37] Same answer
Clef Flash A — [98, 73, 3, 58, 93, 6, 44, 37] A — [98, 73, 3, 58, 93, 6, 44, 37] Same answer
Strands Decider A — [98, 73, 3, 58, 93, 6, 44, 37] A — [98, 73, 3, 58, 93, 6, 44, 37] Same answer
GLiNER2.5 Decide A — [98, 73, 3, 58, 93, 6, 44, 37] A — [98, 73, 3, 58, 93, 6, 44, 37] Same answer
Laya A — [98, 73, 3, 58, 93, 6, 44, 37] A — [98, 73, 3, 58, 93, 6, 44, 37] Same answer
CalDec v1 A — [98, 73, 3, 58, 93, 6, 44, 37] A — [98, 73, 3, 58, 93, 6, 44, 37] Same answer
Identifiers and recorded conditions

Variant ID: list_functions/f00533702f645272e534/similar_distractors/low/clean-source_first_induction_new_inputs_full_0184

Input fingerprint: e4fec0cea3bff63adf23bae600d1b5f71b89e56baf6d2a406e9eeb8ea8239ddf

Change Recorded version
Similar distractors Add fewer similar but irrelevant items

Source attribution and dataset terms are linked from the original question.

Type to search questions, variants and documentation.