Datasets:
dataset card
Browse files
README.md
CHANGED
|
@@ -65,6 +65,11 @@ All tasks have qrels files `qrels/<task>_<split>.tsv` (splits `train`, `dev`, `t
|
|
| 65 |
- **Pre-submission variant** (`precedent`, `precedenthc`): same citation edges, eligible = answers published before
|
| 66 |
the question date.
|
| 67 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 68 |
## Files
|
| 69 |
|
| 70 |
All files of a release live under `<version>/` (this card: `v1.0/`); the git tag `v1.0` freezes them. Runs and
|
|
@@ -99,9 +104,11 @@ alone scores MRR@10 0.989) and a 0–180-day date window lifts BM25 from 0.547 t
|
|
| 99 |
as-of filters of the citation tasks.
|
| 100 |
|
| 101 |
**Baseline protocol caveat.** Fine-tuning on the training split with BM25 hard negatives mined across the whole
|
| 102 |
-
archive (the pre-registered protocol) hurts (MRR@10 0.
|
| 103 |
-
|
| 104 |
-
|
|
|
|
|
|
|
| 105 |
|
| 106 |
Splits are temporal (train 2014–2021, dev 2022–2023, test 2024–2026); 2015 questions are dated from their XML
|
| 107 |
header.
|
|
|
|
| 65 |
- **Pre-submission variant** (`precedent`, `precedenthc`): same citation edges, eligible = answers published before
|
| 66 |
the question date.
|
| 67 |
|
| 68 |
+
**Example (test).** Question `P-9-2024-000684` (4 Mar 2024, asbestos screening proposal): its official reply
|
| 69 |
+
`P-9-2024-000684-ASW` is a referral stub ("…would refer the Honourable Member to its answer to written question
|
| 70 |
+
[REF]"), which is the AnswerRet target. The answer it cites, `E-9-2023-003699-ASW` (12 Mar 2024), is the CiteRet
|
| 71 |
+
target; it is eligible at drafting time (before the reply) but not in the pre-submission variant (after the question).
|
| 72 |
+
|
| 73 |
## Files
|
| 74 |
|
| 75 |
All files of a release live under `<version>/` (this card: `v1.0/`); the git tag `v1.0` freezes them. Runs and
|
|
|
|
| 104 |
as-of filters of the citation tasks.
|
| 105 |
|
| 106 |
**Baseline protocol caveat.** Fine-tuning on the training split with BM25 hard negatives mined across the whole
|
| 107 |
+
archive (the pre-registered protocol) hurts (MRR@10 0.424 ± 0.008 over 3 seeds vs 0.490 zero-shot). Post-hoc
|
| 108 |
+
controls: random negatives 0.666, random negatives within ±365 days 0.708, date-matched BM25 negatives 0.637. The
|
| 109 |
+
mined negatives are mostly at least partially relevant answers the qrels miss (audited test queries: 18% "directly
|
| 110 |
+
addresses", 51% "partially"), plus a date signal. Avoid naive hard-negative mining here; these results are post hoc,
|
| 111 |
+
to be confirmed on the next test period. Best pre-registered baseline: RRF(BM25, BGE-M3), 0.588.
|
| 112 |
|
| 113 |
Splits are temporal (train 2014–2021, dev 2022–2023, test 2024–2026); 2015 questions are dated from their XML
|
| 114 |
header.
|