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: was Mila in the garden Speaker_2: she was in the garden or the hallway Speaker_1: where are the pineapples Speaker_2: Aria said there is a green treasure chest in the garden Speaker_1: did Mila place the pineapples there Speaker_2: she placed them there and travelled to the hallway Speaker_1: are some of the pineapples there Speaker_2: two pineapples are in the garden Speaker_1: were you there Speaker_2: I was not there Speaker_1: then where Speaker_2: I journeyed to the hallway and the back_yard Speaker_1: where can I find the celery Speaker_2: there is a red crate in the garden Speaker_1: did Aria place the celery there Speaker_2: yes Speaker_1: where are the oranges Speaker_2: the oranges are in the back_yard or the hallway Speaker_1: how about the beans Speaker_2: they are in the back_yard or the workshop Question: did Aria put the celery in the garden?

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: was Mila in the garden\nSpeaker_2: she was in the garden or the hallway\nSpeaker_1: where are the pineapples\nSpeaker_2: Aria said there is a green treasure chest in the garden\nSpeaker_1: did Mila place the pineapples there\nSpeaker_2: she placed them there and travelled to the hallway\nSpeaker_1: are some of the pineapples there\nSpeaker_2: two pineapples are in the garden\nSpeaker_1: were you there\nSpeaker_2: I was not there\nSpeaker_1: then where\nSpeaker_2: I journeyed to the hallway and the back_yard\nSpeaker_1: where can I find the celery\nSpeaker_2: there is a red crate in the garden\nSpeaker_1: did Aria place the celery there\nSpeaker_2: yes\nSpeaker_1: where are the oranges\nSpeaker_2: the oranges are in the back_yard or the hallway\nSpeaker_1: how about the beans\nSpeaker_2: they are in the back_yard or the workshop\nQuestion: did Aria put the celery in the garden?\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 A — yes Same answer
GLiNER2.5 Decide A — yes A — yes Same answer
Laya A — yes A — yes Same answer
CalDec v1 A — yes A — yes Same answer
Identifiers and recorded conditions

Variant ID: pub/107251d5eb85b300c265/examples/prose_control/clean-source_first_pub_prose_controls_full_0168

Input fingerprint: d7a5d459c82daa7bd8d1fdc0ef8ca6d9c83b23ff5bb1749294c2f4a1e9f316bf

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.