封神榜-IDEA研究院

这个项目的核心优势、应用场景和相关项目如下：
– **核心优势**：封神榜是中文最大的开源预训练模型体系目前已有超过98个开源预训练模型。目前已经开源了第一个中文Stable Diffusion开源了中文第一个CLIP模型二郎神UniMC等模型多次获得FewCLUE/ZeroCLUE等榜单冠军。
– **应用场景**：多模态、知识检索、代码生成。
– **相关项目**：
– [BioBART: Pretraining and Evaluation of A Biomedical Generative Language Model.](https://arxiv.org/abs/2106.06162)
– [MAP: Multimodal Uncertainty-Aware Vision-Language Pre-training Model.](https://arxiv.org/abs/2110.04916)
– [Unified BERT for Few-shot Natural Language Understanding.](https://arxiv.org/abs/2110.04916)
– [Towards No.1 in CLUE Semantic Matching Challenge: Pre-trained Language Model Erlangshen with Propensity-Corrected Loss.](https://arxiv.org/abs/2110.04916)
– [Fengshenbang 1.0: Being the Foundation of Chinese Cognitive Intelligence.](https://arxiv.org/abs/2110.04916)
– [Flat Multi-modal Interaction Transformer for Named Entity Recognition.](https://arxiv.org/abs/2110.04916)
– [TCBERT: A Technical Report for Chinese Topic Classification BERT.](https://arxiv.org/abs/2110.04916)
– [Ziya-Visual: Bilingual Large Vision-Language Model via Multi-Task Instruction Tuning.](https://arxiv.org/abs/2110.04916)
– [Ziya2: Data-centric Learning is All LLMs Need.](https://arxiv.org/abs/2110.04916)
– [Never Lost in the Middle: Improving Large Language Models via Attention Strengthening Question Answering.](https://arxiv.org/abs/2110.04916)