Communication · question 38
Back to Communication · All capabilities
Question ID: pub/d56e739cd4f847b2da94
Source: PUB
Input format: native choices
Explore this question’s altered versions
Context
Speaker_1: He was very sad. Speaker_2: Yeah, He was sad like a widow.
Question
Choose the interpretation or answer requested by the text.
Answer choices
- A: Agrees
- B: Sarcastic
Reference answer: Agrees (A)
Exact evaluated input as JSON
{
"state": "Speaker_1: He was very sad.\nSpeaker_2: Yeah, He was sad like a widow.",
"question": "Choose the interpretation or answer requested by the text.",
"choices": {
"A": "Agrees",
"B": "Sarcastic"
}
}
Recorded model answers
| Model | Answer | Outcome |
|---|---|---|
| Astra | Agrees | Correct |
| Opus 5.5 | Agrees | Correct |
| GLiDE v1.2 | Agrees | Correct |
| Decider v1.1 | Agrees | Correct |
| Jev | Agrees | Correct |
| Microsoft Decision-1 | Agrees | Correct |
| Clef | Agrees | Correct |
| Luna Decisions | Agrees | Correct |
| Clef Flash | Agrees | Correct |
| Strands Decider | Agrees | Correct |
| GLiNER2.5 Decide | Sarcastic | Incorrect |
| Laya | Sarcastic | Incorrect |
| CalDec v1 | Agrees | Correct |
Source and sharing
Attribution: CFILT, IIT Bombay; PUB 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": "extracted/task_6.zip/task_6.jsonl",
"record": "980",
"kind": "source_file_record"
},
"source_revision": "65a6a87359fe4aa5278952741d5ed7e0ecb0f0ff",
"input_hash": "5c4643e38d14a34447cf04e74392450a777df67957cb09a7850e82a2234140e1"
}