RaindropA practical guide to choosing decision modelsGitHub ↗

Spatial cognition · question 18

Back to Spatial cognition · All capabilities

Question ID: spartqa/a986344d6b4bea903ff9
Source: SPARTQA
Input format: native choices

Explore this question’s altered versions

Context

We have two blocks. Lets call them A and B. Block A is to the left of B. Block A contains a small black triangle. Block B has a medium blue square, a medium black square and one small yellow triangle. The medium blue square is above and to the right of the medium black square which is near to a small black square and the small yellow triangle.

Question

Is the small black triangle , to the left of all black squares that are near to a black square?

Answer choices

  • Yes: Yes
  • No: No
  • DK: DK

Reference answer: Yes (Yes)

Exact evaluated input as JSON
{
  "state": "We have two blocks. Lets call them A and B. Block A is to the left of B. Block A contains a small black triangle. Block B has a medium blue square, a medium black square and one small yellow triangle. The medium blue square is above and to the right of the medium black square which is near to a small black square and the small yellow triangle.",
  "question": "Is the small black triangle , to the left of all black squares that are near to a black square?",
  "choices": {
    "Yes": "Yes",
    "No": "No",
    "DK": "DK"
  }
}

Recorded model answers

Model Answer Outcome
Astra Yes Correct
Opus 5.5 Yes Correct
GLiDE v1.2 Yes Correct
Decider v1.1 Yes Correct
Jev Yes Correct
Microsoft Decision-1 Yes Correct
Clef Yes Correct
Luna Decisions Yes Correct
Clef Flash Yes Correct
Strands Decider Yes Correct
GLiNER2.5 Decide Yes Correct
Laya Yes Correct
CalDec v1 Yes Correct

Source and sharing

Attribution: HLR / Roshanak Mirzaee et al..
Terms: MIT. 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": "extracted/SpartQA_Auto.zip/SpartQA_Auto/train.json",
    "record": "12841/0",
    "kind": "source_file_record"
  },
  "source_revision": "9892ed1bff9103f5a9103c68bd271b7630ab661e",
  "input_hash": "77677490595fce474b48711ac4350397c83fbbcc5bf732d6b92ab93e40aa3ffb"
}

Type to search questions, variants and documentation.