PEFT
Safetensors
English
Chinese
long-context
context-extension
hierarchical-attention
segmented-attention
qwen3
lora
hici
Instructions to use ZengXiangyu/Qwen3-8b-HiCI-48k-500steps with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use ZengXiangyu/Qwen3-8b-HiCI-48k-500steps with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("./models/Qwen3-8B") model = PeftModel.from_pretrained(base_model, "ZengXiangyu/Qwen3-8b-HiCI-48k-500steps") - Notebooks
- Google Colab
- Kaggle
Update README.md
Browse files
README.md
CHANGED
|
@@ -119,11 +119,11 @@ tokenizer = transformers.AutoTokenizer.from_pretrained("ZengXiangyu/Qwen3-8b-HiC
|
|
| 119 |
## Citation
|
| 120 |
|
| 121 |
```bibtex
|
| 122 |
-
@article{
|
| 123 |
-
title={HiCI: Hierarchical Construction-Integration for Long-Context
|
| 124 |
-
author={Zeng, Xiangyu
|
| 125 |
-
journal={arXiv:2603.20843},
|
| 126 |
-
year={
|
| 127 |
}
|
| 128 |
```
|
| 129 |
|
|
|
|
| 119 |
## Citation
|
| 120 |
|
| 121 |
```bibtex
|
| 122 |
+
@article{zeng2026hici,
|
| 123 |
+
title={HiCI: Hierarchical Construction-Integration for Long-Context Attention},
|
| 124 |
+
author={Zeng, Xiangyu and Xu, Qi and Wang, Yunke and Xu, Chang},
|
| 125 |
+
journal={arXiv preprint arXiv:2603.20843},
|
| 126 |
+
year={2026}
|
| 127 |
}
|
| 128 |
```
|
| 129 |
|