File size: 1,826 Bytes
3c332d2 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 | from researchbench.clients import MockClient
from researchbench.retrieve import parse_retrieve_response, run_retrieve_record, score_retrieve
def test_parse_retrieve_response_extracts_titles_and_reasons() -> None:
parsed = parse_retrieve_response(
"Title: First paper\n"
"Reason: useful bridge.\n"
"Title: Second paper\n"
"Reason: complementary mechanism."
)
assert parsed == [
{"title": "First paper", "reason": "useful bridge."},
{"title": "Second paper", "reason": "complementary mechanism."},
]
def test_parse_retrieve_response_extracts_moosechem_title_markers() -> None:
parsed = parse_retrieve_response(
"**Title 1 starts:** First paper **Title 1 ends**\n"
"**Title 2 starts:** Second paper **Title 2 ends**"
)
assert parsed == [
{"title": "First paper", "reason": ""},
{"title": "Second paper", "reason": ""},
]
def test_mock_retrieve_two_rounds_keep_fifteen_then_three() -> None:
candidates = [
{"index": idx, "title": f"Paper {idx}", "abstract": f"Abstract {idx}", "label": "gold" if idx < 2 else "negative_t1"}
for idx in range(75)
]
record = {
"sample_id": "sample/1",
"research_question": "question",
"background_survey": "survey",
"candidates": candidates,
}
pred = run_retrieve_record(record, MockClient())
assert len(pred["selected_round1_titles"]) == 15
assert len(pred["selected_round2_titles"]) == 3
assert pred["selected_round1_titles"][:3] == ["Paper 0", "Paper 1", "Paper 2"]
assert pred["selected_round2_titles"] == ["Paper 0", "Paper 1", "Paper 2"]
score = score_retrieve([pred], [record])
assert score["summary"]["round1"]["gold"] == 1.0
assert score["summary"]["round2"]["gold"] == 1.0
|