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: question instructions.

How the edits were constructed

Context

It’s good to hear once again how the government are setting a great example when it comes to following their own COVID restrictions.....

Question

Is this message intended to be sarcastic?

The notebook has a linen cover and stitched binding.

Answer choices

  • no. Not sarcastic
  • yes. Sarcastic

Reference answer: yes.

Exact evaluated input as JSON
{
  "state": "It’s good to hear once again how the government are setting a great example when it comes to following their own COVID restrictions.....",
  "question": "Is this message intended to be sarcastic?\n\nThe notebook has a linen cover and stitched binding.",
  "choices": {
    "no": "Not sarcastic",
    "yes": "Sarcastic"
  }
}

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 yes — Sarcastic yes — Sarcastic Same answer
Decider v1.1 yes — Sarcastic yes — Sarcastic Same answer
Jev yes — Sarcastic yes — Sarcastic Same answer
Microsoft Decision-1 yes — Sarcastic yes — Sarcastic Same answer
Clef yes — Sarcastic no — Not sarcastic Different answer
Luna Decisions yes — Sarcastic no — Not sarcastic Different answer
Clef Flash yes — Sarcastic yes — Sarcastic Same answer
Strands Decider no — Not sarcastic no — Not sarcastic Same answer
GLiNER2.5 Decide no — Not sarcastic no — Not sarcastic Same answer
Laya no — Not sarcastic no — Not sarcastic Same answer
CalDec v1 no — Not sarcastic no — Not sarcastic Same answer
Identifiers and recorded conditions

Variant ID: isarcasmeval/045b4295d9bb572acadc/examples/prose_control/clean-source_first_sarcasm_question_contexts_full_0182

Input fingerprint: 46cada66269663f1ae9bd1a90befb2f7ed4e77b1e70a6019967096c0480cee5a

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.