Commit
·
349a655
1
Parent(s):
9c5e5a3
Update README.md
Browse files
README.md
CHANGED
|
@@ -11,23 +11,23 @@ This model is DPR trained on MS MARCO. The training details and evaluation resul
|
|
| 11 |
|BERI Dataset|NDCG@10|
|
| 12 |
|:----|:----|
|
| 13 |
|TREC-COVID|58.8|
|
| 14 |
-
|NFCorpus|
|
| 15 |
-
|FiQA|
|
| 16 |
-
|ArguAna|
|
| 17 |
-
|Touché-2020|
|
| 18 |
-
|Quora|0
|
| 19 |
-
|SCIDOCS|
|
| 20 |
-
|SciFact|
|
| 21 |
-
|NQ|
|
| 22 |
-
|HotpotQA|
|
| 23 |
-
|Signal-1M|
|
| 24 |
-
|TREC-NEWS|
|
| 25 |
-
|DBPedia-entity|
|
| 26 |
-
|Fever|0
|
| 27 |
-
|Climate-Fever|
|
| 28 |
-
|BioASQ|
|
| 29 |
-
|Robust04|
|
| 30 |
-
|CQADupStack|
|
| 31 |
|
| 32 |
|
| 33 |
The implementation is the same as our EMNLP 2022 paper ["Reduce Catastrophic Forgetting of Dense Retrieval Training with Teleportation Negatives"](https://arxiv.org/pdf/2210.17167.pdf). The associated GitHub repository is available at https://github.com/OpenMatch/ANCE-Tele.
|
|
|
|
| 11 |
|BERI Dataset|NDCG@10|
|
| 12 |
|:----|:----|
|
| 13 |
|TREC-COVID|58.8|
|
| 14 |
+
|NFCorpus|23.4|
|
| 15 |
+
|FiQA|20.6|
|
| 16 |
+
|ArguAna|39.4|
|
| 17 |
+
|Touché-2020|22.3|
|
| 18 |
+
|Quora|78.0|
|
| 19 |
+
|SCIDOCS|11.9|
|
| 20 |
+
|SciFact|49.4|
|
| 21 |
+
|NQ|43.9|
|
| 22 |
+
|HotpotQA|45.3|
|
| 23 |
+
|Signal-1M|20.2|
|
| 24 |
+
|TREC-NEWS|31.8|
|
| 25 |
+
|DBPedia-entity|28.7|
|
| 26 |
+
|Fever|65.0|
|
| 27 |
+
|Climate-Fever|14.9|
|
| 28 |
+
|BioASQ|24.1|
|
| 29 |
+
|Robust04|32.3|
|
| 30 |
+
|CQADupStack|28.3|
|
| 31 |
|
| 32 |
|
| 33 |
The implementation is the same as our EMNLP 2022 paper ["Reduce Catastrophic Forgetting of Dense Retrieval Training with Teleportation Negatives"](https://arxiv.org/pdf/2210.17167.pdf). The associated GitHub repository is available at https://github.com/OpenMatch/ANCE-Tele.
|