RaindropA practical guide to choosing decision modelsGitHub ↗

Communication · question 63

Back to Communication · All capabilities

Question ID: isarcasmeval/015bc09c82beb2679504
Source: iSarcasmEval
Input format: native choices

Explore this question’s altered versions

Context

Second hand embarrassment for anyone that actually queued to get into a nightclub last night🧐

Question

Is this message intended to be sarcastic?

Answer choices

  • no: Not sarcastic
  • yes: Sarcastic

Reference answer: Not sarcastic (no)

Exact evaluated input as JSON
{
  "state": "Second hand embarrassment for anyone that actually queued to get into a nightclub last night🧐",
  "question": "Is this message intended to be sarcastic?",
  "choices": {
    "no": "Not sarcastic",
    "yes": "Sarcastic"
  }
}

Recorded model answers

Model Answer Outcome
Astra Not sarcastic Correct
Opus 5.5 Not sarcastic Correct
GLiDE v1.2 Not sarcastic Correct
Decider v1.1 Not sarcastic Correct
Jev Sarcastic Incorrect
Microsoft Decision-1 Not sarcastic Correct
Clef Not sarcastic Correct
Luna Decisions Sarcastic Incorrect
Clef Flash Not sarcastic Correct
Strands Decider Sarcastic Incorrect
GLiNER2.5 Decide Not sarcastic Correct
Laya Not sarcastic Correct
CalDec v1 Not sarcastic Correct

Source and sharing

Attribution: Ibrahim Abu Farha et al.; iSarcasmEval authors.
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": "data/prepared/isarcasmeval.jsonl",
    "record": "2583",
    "kind": "prepared_normalizer_record",
    "native_id": "isarcasmeval/train.En/2583/baseline"
  },
  "source_revision": "dfc708b53bde1bb571abfb5692f63231c2232195",
  "input_hash": "8474c7d18bd9b6cb59453459d0dc77f588eb089c7de1cdff73a58a687a07428c"
}

Type to search questions, variants and documentation.