"Ziya-Reader-13B-v1.0" is a knowledge question-answering model. It can accurately answer questions given questions and knowledge documents, and is suitable for both multi-document and single-document question-answering. The model has an 8k context window, and compared to models with longer windows, we have achieved victory in evaluations across multiple long-text tasks.
The tasks include multi-document question-answering, synthetic tasks (document retrieval), and long-text summarization.
Additionally, the model also demonstrates excellent generalization capabilities, enabling it to be used for general question-answering. Its performance on our general ability evaluation set surpassed that of Ziya-Llama-13B.
"Multi-doc QA" is a multi-document question-answering task, where given a question and multiple documents, the model answers the question based on the documents that contain relevant information. This task measures the model's ability in relevance judgment, memory, and question-answering skills.
"Synthetic task" is a synthetic document retrieval task, where given a summary, the goal is to find the corresponding document from a large number of documents. This task evaluates the model's semantic matching ability.
"Summarization" is a long-text summarization task, where given meeting records containing multiple speakers, the model generates a meeting summary with an extremely long context.
We found that the documents in Multi-doc QA were arranged in descending order of relevance, with the correct answer often in the first or early positions, which did not truly reflect the model's ability in relevance judgment. Therefore, we shuffled the document order in this test set and evaluated the performance of various models. The results showed a significant decrease in performance for most models, ranging from 5% to 17%. In contrast, our model demonstrated high robustness with a decrease of less than 2%.
如果您在您的工作中使用了我们的模型,可以引用我们的
论文
:
If you are using the resource for your work, please cite our
paper
:
@article{ziya-reader,
title={Never Lost in the Middle: Improving Large Language Models via Attention Strengthening Question Answering},
author={Junqing He and Kunhao Pan and Xiaoqun Dong and Zhuoyang Song and Yibo Liu and Yuxin Liang and Hao Wang and Qianguo Sun and Songxin Zhang and Zejian Xie and Jiaxing Zhang},
year={2023},
eprint={2311.09198},
archivePrefix={arXiv},
primaryClass={cs.CL}
}
@misc{zhang2023fengshenbang,
title={Fengshenbang 1.0: Being the Foundation of Chinese Cognitive Intelligence},
author={Jiaxing Zhang and Ruyi Gan and Junjie Wang and Yuxiang Zhang and Lin Zhang and Ping Yang and Xinyu Gao and Ziwei Wu and Xiaoqun Dong and Junqing He and Jianheng Zhuo and Qi Yang and Yongfeng Huang and Xiayu Li and Yanghan Wu and Junyu Lu and Xinyu Zhu and Weifeng Chen and Ting Han and Kunhao Pan and Rui Wang and Hao Wang and Xiaojun Wu and Zhongshen Zeng and Chongpei Chen},
year={2023},
eprint={2209.02970},
archivePrefix={arXiv},
primaryClass={cs.CL}
}
Ziya-Reader-13B-v1.0 huggingface.co is an AI model on huggingface.co that provides Ziya-Reader-13B-v1.0's model effect (), which can be used instantly with this IDEA-CCNL Ziya-Reader-13B-v1.0 model. huggingface.co supports a free trial of the Ziya-Reader-13B-v1.0 model, and also provides paid use of the Ziya-Reader-13B-v1.0. Support call Ziya-Reader-13B-v1.0 model through api, including Node.js, Python, http.
Ziya-Reader-13B-v1.0 huggingface.co is an online trial and call api platform, which integrates Ziya-Reader-13B-v1.0's modeling effects, including api services, and provides a free online trial of Ziya-Reader-13B-v1.0, you can try Ziya-Reader-13B-v1.0 online for free by clicking the link below.
IDEA-CCNL Ziya-Reader-13B-v1.0 online free url in huggingface.co:
Ziya-Reader-13B-v1.0 is an open source model from GitHub that offers a free installation service, and any user can find Ziya-Reader-13B-v1.0 on GitHub to install. At the same time, huggingface.co provides the effect of Ziya-Reader-13B-v1.0 install, users can directly use Ziya-Reader-13B-v1.0 installed effect in huggingface.co for debugging and trial. It also supports api for free installation.
Ziya-Reader-13B-v1.0 install url in huggingface.co: