dataset card

README.md

@@ -28,25 +28,19 @@ configs:
 - config_name: qrels-answer
   data_files:
   - split: train
-    path: v1.0/qrels/answer_train.tsv
+    path: v1.0/qrels_jsonl/answer_train.jsonl
   - split: dev
-    path: v1.0/qrels/answer_dev.tsv
+    path: v1.0/qrels_jsonl/answer_dev.jsonl
   - split: test
-    path: v1.0/qrels/answer_test.tsv
-  sep: " "
-  header: null
-  column_names: [query_id, iteration, doc_id, relevance]
+    path: v1.0/qrels_jsonl/answer_test.jsonl
 - config_name: qrels-cite
   data_files:
   - split: train
-    path: v1.0/qrels/cite_train.tsv
+    path: v1.0/qrels_jsonl/cite_train.jsonl
   - split: dev
-    path: v1.0/qrels/cite_dev.tsv
+    path: v1.0/qrels_jsonl/cite_dev.jsonl
   - split: test
-    path: v1.0/qrels/cite_test.tsv
-  sep: " "
-  header: null
-  column_names: [query_id, iteration, doc_id, relevance]
+    path: v1.0/qrels_jsonl/cite_test.jsonl
 ---
 
 # ParlIR
@@ -84,7 +78,9 @@ audit records are under `v1.0/experiments/{runs,audit}/`.
 - `precedent_meta.jsonl`: per query, citation edges (`cited`, `answer`, provenance `via` ⊆ {`hyperlink`, `url`,
   `text`}, `referral` flag), with the eligible answers under the pre-submission and drafting cutoffs.
 - `qrels/<task>_<split>.tsv` for `answer`, `answerlenient`, `cite`, `citehc`, `precedent`, `precedenthc`: TREC
-  format `qid 0 docid rel` (whitespace-separated, no header).
+  format `qid 0 docid rel` (whitespace-separated, no header), for `trec_eval`/`ir_measures`.
+- `qrels_jsonl/<task>_<split>.jsonl`: the same qrels as JSONL (`query_id`, `iteration`, `doc_id`, `relevance`);
+  these back the `qrels-*` dataset configs.
 - `build_manifest.json`: counts and sha256 checksums.
 - LLM pooled-audit pools (sensitivity analysis only; reproducible from the sample seed and the runs).