README.md

---
license: apache-2.0
language:
- ko
base_model:
- Alibaba-NLP/gte-multilingual-reranker-base
datasets:
- sigridjineth/korean_nli_dataset_reranker_v0
tags:
- reranker
- korean
---
# Model Card: sigridjineth/ko-reranker-v1.1

- The model, which is finetuned from [Alibaba-NLP/gte-multilingual-reranker-base](https://huggingface.co/Alibaba-NLP/gte-multilingual-reranker-base) is currently under development and may undergo further changes as we refine and improve its performance. Underwent A100 x 8 with 12 hours for training.

## Training Data

This model is trained on 328K Korean Triplets, which aggregates several publicly available datasets, ensuring rich linguistic diversity:

- **kor_nli (train)**: [https://huggingface.co/datasets/kor_nli](https://huggingface.co/datasets/kor_nli)  
- **mnli_ko (train)**: [https://huggingface.co/datasets/kozistr/mnli_ko](https://huggingface.co/datasets/kozistr/mnli_ko)  
- **ko-wiki-reranking (train)**: [https://huggingface.co/datasets/upskyy/ko-wiki-reranking](https://huggingface.co/datasets/upskyy/ko-wiki-reranking)  
- **mr_tydi_korean (train)**: [https://huggingface.co/datasets/castorini/mr-tydi](https://huggingface.co/datasets/castorini/mr-tydi)  
- **klue_nli (train)**: [https://huggingface.co/datasets/klue/klue](https://huggingface.co/datasets/klue/klue)

These combined resources ensure coverage across a wide range of topics, styles, and complexities in Korean language data, enabling the model to capture nuanced semantic differences.

## Key Features

- **Hard Negative Mining**:  
  Integrated [BAAI/bge-m3](https://huggingface.co/BAAI/bge-m3) to mine challenging negatives. This approach sharpens the model’s ability to distinguish subtle contrasts, boosting robustness and improving ranking quality.

- **Teacher-Student Distillation**:  
  Leveraged [BAAI/bge-reranker-v2.5-gemma2-lightweight](https://huggingface.co/BAAI/bge-reranker-v2.5-gemma2-lightweight) as a teacher model. The student reranker learned from teacher-provided positive/negative scores, accelerating convergence and achieving better final performance.

## Intended Use

- **Search & Information Retrieval**: Improve document ranking for Korean-language search queries.  
- **Question Answering (QA)**: Enhance QA pipelines by reordering candidate answers for improved relevance.  
- **Content Recommendation**: Refine recommendation engines that rely on textual signals to deliver more accurate suggestions.

## Limitations & Future Work

- **Preview Release**:  
  The model is still in the refinement phase. Expect future updates to improve stability, generalization, and performance.
  
- **Need for Evaluation**:  
  Developing and standardizing benchmarks for generalized Korean retrieval tasks (especially for rerankers) will be an ongoing effort.

## Evaluation
The [AutoRAG Benchmark](https://github.com/Marker-Inc-Korea/AutoRAG-example-korean-embedding-benchmark) serves as both the evaluation dataset and the toolkit for reporting these metrics.

### Model: `sigridjineth/ko-reranker-v1.1-preview`

| top_k | Execution Time | F1     | Recall | Precision | MAP    | MRR    | NDCG   | Is Best |
|-------|----------------|--------|--------|-----------|--------|--------|--------|---------|
| 1     | 0.0438         | 0.6754 | 0.6754 | 0.6754    | 0.6754 | 0.6754 | 0.6754 | True    |
| 3     | 0.0486         | 0.3684 | 0.7368 | 0.2456    | 0.7032 | 0.7032 | 0.7119 | False   |
| 5     | 0.0446         | 0.3684 | 0.7368 | 0.2456    | 0.7032 | 0.7032 | 0.7119 | False   |

---

### Model: `Alibaba-NLP/gte-multilingual-reranker-base`

| top_k | Execution Time | F1     | Recall | Precision | MAP    | MRR    | NDCG   | Is Best |
|-------|----------------|--------|--------|-----------|--------|--------|--------|---------|
| 1     | 0.0481         | 0.6316 | 0.6316 | 0.6316    | 0.6316 | 0.6316 | 0.6316 | True    |
| 3     | 0.0427         | 0.3596 | 0.7193 | 0.2398    | 0.6725 | 0.6725 | 0.6846 | False   |
| 5     | 0.0442         | 0.3596 | 0.7193 | 0.2398    | 0.6725 | 0.6725 | 0.6846 | False   |

---

### Model: `dragonkue/bge-reranker-v2-m3-ko`

| top_k | Execution Time | F1     | Recall | Precision | MAP    | MRR    | NDCG   | Is Best |
|-------|----------------|--------|--------|-----------|--------|--------|--------|---------|
| 1     | 0.0814         | 0.6930 | 0.6930 | 0.6930    | 0.6930 | 0.6930 | 0.6930 | True    |
| 3     | 0.0813         | 0.3596 | 0.7193 | 0.2398    | 0.7061 | 0.7061 | 0.7096 | False   |
| 5     | 0.0824         | 0.3596 | 0.7193 | 0.2398    | 0.7061 | 0.7061 | 0.7096 | False   |

```
Evaluation Results (k=1,3,5,10):
  Accuracy@1:  0.8070
  F1@1:        0.8070
  Recall@1:    0.8070
  Precision@1: 0.8070
  Accuracy@3:  0.9211
  F1@3:        0.4605
  Recall@3:    0.9211
  Precision@3: 0.3070
  Accuracy@5:  0.9474
  F1@5:        0.3158
  Recall@5:    0.9474
  Precision@5: 0.1895
  Accuracy@10:  0.9737
  F1@10:        0.1770
  Recall@10:    0.9737
  Precision@10: 0.0974

Total inference time (all queries): 142.64 sec
Average inference time (per query): 1.2512 sec
```

## Usage (transformers>=4.36.0)

```python
import torch
from transformers import AutoModelForSequenceClassification, AutoTokenizer

model_name_or_path = "sigridjineth/ko-reranker-v1.1-preview"

tokenizer = AutoTokenizer.from_pretrained(model_name_or_path)
model = AutoModelForSequenceClassification.from_pretrained(
    model_name_or_path, 
    trust_remote_code=True,
    torch_dtype=torch.float16
)
model.eval()

pairs = [
    ["중국의 수도는","베이징"], 
    ["2024년 대한민국 대통령은?", "대한민국 대통령은 윤석열이다"], 
    ["파이썬에서 퀵 소트를 구현하기","quick sort로 코테 1등 먹어보자"]
]

with torch.no_grad():
    inputs = tokenizer(pairs, padding=True, truncation=True, return_tensors='pt', max_length=512)
    scores = model(**inputs, return_dict=True).logits.view(-1, ).float()
    print(scores)
# Example output:
# tensor([1.2315, 0.5923, 0.3041])
```

## Usage with Infinity

[Infinity](https://github.com/michaelfeil/infinity) is an MIT-licensed inference REST API server that can easily host and serve models. For instance:

```bash
docker run --gpus all -v $PWD/data:/app/.cache -p "7997":"7997" \
michaelf34/infinity:0.0.68 \
v2 --model-id Alibaba-NLP/gte-multilingual-reranker-base --revision "main" \
--dtype bfloat16 --batch-size 32 --device cuda --engine torch --port 7997
```

## References

```
@misc{zhang2024mgtegeneralizedlongcontexttext,
  title={mGTE: Generalized Long-Context Text Representation and Reranking Models for Multilingual Text Retrieval}, 
  author={Xin Zhang and Yanzhao Zhang and Dingkun Long and Wen Xie and Ziqi Dai and Jialong Tang and Huan Lin and Baosong Yang and Pengjun Xie and Fei Huang and Meishan Zhang and Wenjie Li and Min Zhang},
  year={2024},
  eprint={2407.19669},
  archivePrefix={arXiv},
  primaryClass={cs.CL},
  url={https://arxiv.org/abs/2407.19669}, 
}

@misc{li2023making,
  title={Making Large Language Models A Better Foundation For Dense Retrieval}, 
  author={Chaofan Li and Zheng Liu and Shitao Xiao and Yingxia Shao},
  year={2023},
  eprint={2312.15503},
  archivePrefix={arXiv},
  primaryClass={cs.CL}
}

@misc{chen2024bge,
  title={BGE M3-Embedding: Multi-Lingual, Multi-Functionality, Multi-Granularity Text Embeddings Through Self-Knowledge Distillation}, 
  author={Jianlv Chen and Shitao Xiao and Peitian Zhang and Kun Luo and Defu Lian and Zheng Liu},
  year={2024},
  eprint={2402.03216},
  archivePrefix={arXiv},
  primaryClass={cs.CL}
}
```