Hugging Face's logo

namdp-ptit
/
ViRanker

Text Classification
Transformers
Safetensors
Vietnamese
xlm-roberta
cross-encoder
rerank
Inference Endpoints
Model card Files Community
Dang Phuong Nam commited on
Commit
ce67b8a
•
1 Parent(s): 938bb7b

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +3 -1
README.md CHANGED
@@ -28,6 +28,8 @@ widget:
28
  * [Using Huggingface transformers](#using-huggingface-transformers)
29
  * [Fine tune](#fine-tune)
30
  * [Data format](#data-format)
 
 
31
 
32
  Different from embedding model, reranker uses question and document as input and directly output similarity instead of
33
  embedding.
@@ -125,7 +127,7 @@ the [MS MMarco Passage Reranking - Vi - Dev](https://huggingface.co/datasets/uni
125
  | Model-Name | NDCG@3 | MRR@3 | NDCG@5 | MRR@5 | NDCG@10 | MRR@10 | Docs / Sec |
126
  |-----------------------------------------------------------------------------------------------------------------------------------------|:-----------|:-----------|:-----------|:-----------|:-----------|:-----------|:-----------|
127
  | [namdp/ViRanker](https://huggingface.co/namdp/ViRanker) | **0.6685** | **0.6564** | 0.6842 | **0.6811** | 0.7278 | **0.6985** | 2.02
128
- | [itdainb/PhoRankere](https://huggingface.co/itdainb/PhoRanker) | 0.6625 | 0.6458 | **0.7147** | 0.6731 | **0.7422** | 0.6830 | **15**
129
  | [kien-vu-uet/finetuned-phobert-passage-rerank-best-eval](https://huggingface.co/kien-vu-uet/finetuned-phobert-passage-rerank-best-eval) | 0.0963 | 0.0883 | 0.1396 | 0.1131 | 0.1681 | 0.1246 | **15**
130
  | [BAAI/bge-reranker-v2-m3](https://huggingface.co/BAAI/bge-reranker-v2-m3) | 0.6087 | 0.5841 | 0.6513 | 0.6062 | 0.6872 | 0.62091 | 3.51
131
  | [BAAI/bge-reranker-v2-gemma](https://huggingface.co/BAAI/bge-reranker-v2-gemma) | 0.6088 | 0.5908 | 0.6446 | 0.6108 | 0.6785 | 0.6249 | 1.29
 
28
  * [Using Huggingface transformers](#using-huggingface-transformers)
29
  * [Fine tune](#fine-tune)
30
  * [Data format](#data-format)
31
+ * [Performance](#performance)
32
+ * [Citation](#citation)
33
 
34
  Different from embedding model, reranker uses question and document as input and directly output similarity instead of
35
  embedding.
 
127
  | Model-Name | NDCG@3 | MRR@3 | NDCG@5 | MRR@5 | NDCG@10 | MRR@10 | Docs / Sec |
128
  |-----------------------------------------------------------------------------------------------------------------------------------------|:-----------|:-----------|:-----------|:-----------|:-----------|:-----------|:-----------|
129
  | [namdp/ViRanker](https://huggingface.co/namdp/ViRanker) | **0.6685** | **0.6564** | 0.6842 | **0.6811** | 0.7278 | **0.6985** | 2.02
130
+ | [itdainb/PhoRanker](https://huggingface.co/itdainb/PhoRanker) | 0.6625 | 0.6458 | **0.7147** | 0.6731 | **0.7422** | 0.6830 | **15**
131
  | [kien-vu-uet/finetuned-phobert-passage-rerank-best-eval](https://huggingface.co/kien-vu-uet/finetuned-phobert-passage-rerank-best-eval) | 0.0963 | 0.0883 | 0.1396 | 0.1131 | 0.1681 | 0.1246 | **15**
132
  | [BAAI/bge-reranker-v2-m3](https://huggingface.co/BAAI/bge-reranker-v2-m3) | 0.6087 | 0.5841 | 0.6513 | 0.6062 | 0.6872 | 0.62091 | 3.51
133
  | [BAAI/bge-reranker-v2-gemma](https://huggingface.co/BAAI/bge-reranker-v2-gemma) | 0.6088 | 0.5908 | 0.6446 | 0.6108 | 0.6785 | 0.6249 | 1.29