RaindropA practical guide to choosing decision modelsGitHub ↗

Learning and induction · question 33

Back to Learning and induction · All capabilities

Question ID: list_functions/33c14eeb3a128d1ed4df
Source: List Functions
Input format: adapted choices

Explore this question’s altered versions

Context

[{"input": "[3, 89, 79, 3, 61, 7, 61]", "output": "[89, 79, 61, 7, 61]"}, {"input": "[2, 66, 3, 23, 87, 0, 3]", "output": "[2, 66, 23, 87, 0]"}, {"input": "[2, 10, 1, 3, 7, 6, 3]", "output": "[2, 10, 1, 7, 6]"}, {"input": "[61, 23, 65, 1, 6, 8, 10, 15, 4, 3]", "output": "[61, 23, 65, 1, 6, 8, 10, 15, 4]"}, {"input": "[2, 67, 21, 3, 7, 83, 79, 3]", "output": "[2, 67, 21, 7, 83, 79]"}, {"input": "[5, 71, 98, 55, 3, 71, 3, 3]", "output": "[5, 71, 98, 55, 71]"}, {"input": "[9, 3, 5, 54, 9, 3, 76, 76, 94, 3]", "output": "[9, 5, 54, 9, 76, 76, 94]"}, {"input": "[3, 3, 3, 3, 3, 3]", "output": "[]"}]

Question

Apply the same inferred rule to this new input: [30, 9, 3, 3, 8]

Answer choices

  • A: [30, 9, 8]
  • B: [1, 1, 1]
  • C: [9, 3, 9]
  • D: [30, 3, 8]

Reference answer: [30, 9, 8] (A)

Exact evaluated input as JSON
{
  "state": "[{\"input\": \"[3, 89, 79, 3, 61, 7, 61]\", \"output\": \"[89, 79, 61, 7, 61]\"}, {\"input\": \"[2, 66, 3, 23, 87, 0, 3]\", \"output\": \"[2, 66, 23, 87, 0]\"}, {\"input\": \"[2, 10, 1, 3, 7, 6, 3]\", \"output\": \"[2, 10, 1, 7, 6]\"}, {\"input\": \"[61, 23, 65, 1, 6, 8, 10, 15, 4, 3]\", \"output\": \"[61, 23, 65, 1, 6, 8, 10, 15, 4]\"}, {\"input\": \"[2, 67, 21, 3, 7, 83, 79, 3]\", \"output\": \"[2, 67, 21, 7, 83, 79]\"}, {\"input\": \"[5, 71, 98, 55, 3, 71, 3, 3]\", \"output\": \"[5, 71, 98, 55, 71]\"}, {\"input\": \"[9, 3, 5, 54, 9, 3, 76, 76, 94, 3]\", \"output\": \"[9, 5, 54, 9, 76, 76, 94]\"}, {\"input\": \"[3, 3, 3, 3, 3, 3]\", \"output\": \"[]\"}]",
  "question": "Apply the same inferred rule to this new input: [30, 9, 3, 3, 8]",
  "choices": {
    "A": "[30, 9, 8]",
    "B": "[1, 1, 1]",
    "C": "[9, 3, 9]",
    "D": "[30, 3, 8]"
  }
}

Recorded model answers

Model Answer Outcome
Astra [30, 9, 8] Correct
Opus 5.5 [30, 9, 8] Correct
GLiDE v1.2 [30, 9, 8] Correct
Decider v1.1 [30, 9, 8] Correct
Jev [30, 9, 8] Correct
Microsoft Decision-1 [30, 9, 8] Correct
Clef [30, 9, 8] Correct
Luna Decisions [30, 9, 8] Correct
Clef Flash [30, 9, 8] Correct
Strands Decider [30, 9, 8] Correct
GLiNER2.5 Decide [30, 9, 8] Correct
Laya [30, 9, 8] Correct
CalDec v1 [30, 9, 8] Correct

Source and sharing

Licensed upstream task: c137. All eight supplied examples and the evaluated test input/output match the original BIG-bench data.

Attribution: Joshua S. Rule, Steven T. Piantadosi, Joshua B. Tenenbaum; BIG-bench contributors; Linlu Qiu et al. (selection).
Terms: Apache-2.0. Source sharing decision.
The capability assignment and choice-format normalization are part of this experiment. Source data retain their own terms.

Exact source locator and input fingerprint
{
  "source_locator": {
    "path": "data/list_function.jsonl",
    "record": "136/5",
    "kind": "source_file_record"
  },
  "source_revision": "0eae21f3d3014b92be34c80b23c9120a7b15a51f",
  "input_hash": "ac5ecc17d6fcf8a01d9a4a33304087da5d8c29d438eb8f8d85f579f888e11045"
}

Type to search questions, variants and documentation.