RaindropA practical guide to choosing decision modelsGitHub ↗

Unrelated examples · No worked example (prose-control recipe) 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

Merovan returns to its melodic refrain with a resolving tonic cadence.

Conversation: Speaker_1: did you see the melons Speaker_2: Lucas said there is a red treasure chest in the office Speaker_1: did Lily leave them there Speaker_2: no Speaker_1: where were you Speaker_2: I travelled to the master_bedroom and the crawlspace Speaker_1: did Aria leave the tomatoes there Speaker_2: she left them in the master_bedroom if not the crawlspace Speaker_1: did she put the asparagus in the office Speaker_2: I don't know Speaker_1: are all the grapefruits in the playroom Speaker_2: yes Speaker_1: are some of the persimmons in the crawlspace Speaker_2: I left all of them in the crawlspace Speaker_1: did you see the plum Speaker_2: the plum is in the master_bedroom Speaker_1: did you place the plum there Speaker_2: no, I didn't Speaker_1: did Aria put the plum there Speaker_2: she put it there and went to the playroom Question: did Aria put the plum in the master_bedroom?

Question

Choose the interpretation or answer requested by the text.

Answer choices

  • A. yes
  • B. no

Reference answer: A.

Exact evaluated input as JSON
{
  "state": "Merovan returns to its melodic refrain with a resolving tonic cadence.\n\nConversation:\nSpeaker_1: did you see the melons\nSpeaker_2: Lucas said there is a red treasure chest in the office\nSpeaker_1: did Lily leave them there\nSpeaker_2: no\nSpeaker_1: where were you\nSpeaker_2: I travelled to the master_bedroom and the crawlspace\nSpeaker_1: did Aria leave the tomatoes there\nSpeaker_2: she left them in the master_bedroom if not the crawlspace\nSpeaker_1: did she put the asparagus in the office\nSpeaker_2: I don't know\nSpeaker_1: are all the grapefruits in the playroom\nSpeaker_2: yes\nSpeaker_1: are some of the persimmons in the crawlspace\nSpeaker_2: I left all of them in the crawlspace\nSpeaker_1: did you see the plum\nSpeaker_2: the plum is in the master_bedroom\nSpeaker_1: did you place the plum there\nSpeaker_2: no, I didn't\nSpeaker_1: did Aria put the plum there\nSpeaker_2: she put it there and went to the playroom\nQuestion: did Aria put the plum in the master_bedroom?\n",
  "question": "Choose the interpretation or answer requested by the text.",
  "choices": {
    "A": "yes",
    "B": "no"
  }
}

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 — yes A — yes Same answer
Decider v1.1 A — yes A — yes Same answer
Jev A — yes A — yes Same answer
Microsoft Decision-1 A — yes A — yes Same answer
Clef A — yes A — yes Same answer
Luna Decisions A — yes A — yes Same answer
Clef Flash A — yes A — yes Same answer
Strands Decider A — yes B — no Different answer
GLiNER2.5 Decide B — no B — no Same answer
Laya A — yes A — yes Same answer
CalDec v1 A — yes A — yes Same answer
Identifiers and recorded conditions

Variant ID: pub/1a4d3e308dd38192e07f/examples/prose_control/clean-source_first_pub_prose_controls_full_0168

Input fingerprint: ad6e40d935ebe629ac0d13178f0e027c92325b3b43221f0a02b528574334fb0d

Change Recorded version
Unrelated examples No worked example (prose-control recipe)

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

Type to search questions, variants and documentation.