hotpotqa: hard negatives and teacher scores (jina35)

README.md

@@ -57,12 +57,11 @@ in this collection.
 | Queries / documents / qrels (all splits) | 97,852 / 5,233,329 / 195,704 |
 | Qrels per query | min 2 · mean 2.0 · max 2 |
 | Score values | 1 ×14,810 |
-| Layout | `queries` · `corpus` · `qrels`, split `test`; `queries`/`qrels` also carry `train`, `dev` — one shared corpus; `hard-negatives` and `teacher-scores` (empty) for `train` |
+| Layout | `queries` · `corpus` · `qrels`, split `test`; `queries`/`qrels` also carry `train`, `dev` — one shared corpus; `hard-negatives` and `teacher-scores` for `train` |
 | Splits | `hard-negatives`: train · `qrels`: train, dev, test · `queries`: train, dev, test · `teacher-scores`: train |
-| Training extension | `hard-negatives` and `teacher-scores` configs exist for the `train` split with **0 rows** — placeholders for the owner's own mining and scoring; `queries`/`qrels` `train` are the benchmark's training data |
-| Hard negatives | **none yet** — config present with 0 rows |
-| Teacher scores | **none yet** — config present with 0 rows |
-| Reading empty configs | `datasets` cannot return a 0-example split (`load_dataset` raises "corresponds to no data"); until rows exist, read the Parquet directly with `pyarrow`/`polars`/`pandas`. The schema is declared in the file and in `configs:` above |
+| Training extension | `hard-negatives` and `teacher-scores` for the `train` split are the owner's own mining and scoring ([details](#hard-negatives-and-teacher-scores)); `queries`/`qrels` `train` are the benchmark's training data |
+| Hard negatives | sources: `dense` · 8,366,684 rows |
+| Teacher scores | `jinaai/jina-reranker-v3.5` · 8,451,684 rows (positives included) |
 | License | `cc-by-sa-4.0` |
 
 ## Schema
@@ -89,6 +88,29 @@ before publishing; `provenance.json` records the source file hashes, what change
 - renamed `_id` → `id`
 - renamed source splits (`queries`←`queries/queries`, `corpus`←`corpus/corpus`) to `test`
 
+## Hard negatives and teacher scores
+
+Filled by the owner's annotation pipeline (`annotation=jina35`) for the `train` split of the
+query set(s) below; queries without a labelled positive are left out.
+
+- **Candidates**: dense retrieval with `jinaai/jina-embeddings-v5-text-small` over the full corpus to depth 1,000;
+  100 candidates per query drawn from the rank windows 1–30 (30), 31–100 (30), 101–300 (20), 301–1000 (20), the query's labelled positives excluded.
+  `rank` is the dense rank; `source` is `dense` for a mined row and `dataset` for a negative the source labels itself
+  (those are kept for every query of the split, sampled or not).
+- **Teacher**: `jinaai/jina-reranker-v3.5`, listwise: a query's positive and all of its candidates are scored together in one
+  context of up to 32,768 tokens. `score` is the raw cosine score, one row per (query, positive) and per
+  (query, candidate); a labelled negative that was also mined is scored once. No filtering is applied to the tables.
+
+| configs | queries | hard negatives | teacher scores |
+|---|---:|---:|---:|
+| `hard-negatives` · `teacher-scores` | 85,000 (seeded sample, seed 1) | 8,366,684 (8,366,684 dense) | 8,451,684 |
+
+```python
+from datasets import load_dataset
+negatives = load_dataset("Hyukkyu/beir-hotpotqa", "hard-negatives", split="train")
+scores    = load_dataset("Hyukkyu/beir-hotpotqa", "teacher-scores", split="train")
+```
+
 ## Load it
 
 ```python

hard-negatives/train-00000-of-00001.parquet

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f62b1b56789db979f3b0d83e5e329c7d6533a72d3da081396053e4bb2a4543b0
-size 823
+oid sha256:ee6497861c4fc61d911c5073cb004cff07b049d4b4dda23f8b2e063f96f595bf
+size 48923489

teacher-scores/{train-00000-of-00001.parquet → train-00000-of-00002.parquet}

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8316266e7cbe6867227eefc104eb93ee2252b885a6127a24b8e5c1982716406a
-size 815
+oid sha256:2ec7e8ab0a20b69a1f68494a819a5807508bb277d6d42b6ef825dde79732ebe6
+size 37071090

teacher-scores/train-00001-of-00002.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:436bde3f08317a163659e8d8b3bb329a484100a6a83296daf5d4fe29c10ce95a
+size 37075653