|
|
---
|
|
|
license: apache-2.0
|
|
|
language:
|
|
|
- en
|
|
|
base_model:
|
|
|
- Qwen/Qwen3-Embedding-8B
|
|
|
---
|
|
|
## LEXA-8B
|
|
|
👉 **LEXA-8B**: LEXA: Legal Case Retrieval via Graph Contrastive Learning with Contextualised LLM Embeddings. More information is available in [**arXiv**](https://arxiv.org/abs/2405.11791) & [**GitHub**](https://github.com/yanran-tang/CaseGNN).
|
|
|
|
|
|
## Example Usage
|
|
|
|
|
|
```python
|
|
|
from transformers import AutoModel, AutoTokenizer
|
|
|
model = AutoModel.from_pretrained("AnnaStudy/LEXA-8B", torch_dtype="auto", device_map="auto")
|
|
|
tokenizer = AutoTokenizer.from_pretrained("AnnaStudy/LEXA-8B")
|
|
|
case_txt = "The following contains key components of a legal case. Legal facts..."
|
|
|
tokenized = tokenizer(case_txt, return_tensors='pt', padding=True, truncation=True, max_length=2048)
|
|
|
outputs = model(**tokenized)
|
|
|
case_embedding = outputs.last_hidden_state[:, -1]
|
|
|
```
|
|
|
## Base Model
|
|
|
|
|
|
ReaKase-8B is finetuned from **Qwen3-Embedding-8B**, which provides the underlying semantic representation capability.
|
|
|
|
|
|
Reference: [Qwen/Qwen3-Embedding-8B](https://huggingface.co/Qwen/Qwen3-Embedding-8B)
|
|
|
|
|
|
## Cite
|
|
|
If you find this repo useful, please cite
|
|
|
```
|
|
|
@article
|
|
|
{LEXA-8B,
|
|
|
author = {Yanran Tang, Ruihong Qiu, Xue Li, Zi Huang},
|
|
|
title = {LEXA: Legal Case Retrieval via Graph Contrastive Learning with Contextualised LLM Embeddings},
|
|
|
journal = {CoRR},
|
|
|
volume = {abs/2405.11791},
|
|
|
year = {2025}
|
|
|
}
|
|
|
``` |
