DeepSpeed-VisualChat：通过多模态因果注意力实现多轮多图像交错对话

摘要

大多数现有的多模态模型受限于无法熟练处理多图像、多轮对话中交错的图像和文本输入，存在着在训练资源分配和数据获取方面的重大约束，影响了它们在不同交互领域中的适应性和可扩展性。为了解决这一问题，我们提出了DeepSpeed-VisualChat框架，旨在通过融合多模态能力来优化大型语言模型（LLMs），重点是提高大型视觉和语言模型处理交错输入的效率。我们的框架值得注意的地方在于：（1）支持多轮和多图像对话的开源支持；（2）引入创新的多模态因果注意机制；以及（3）利用现有数据集的数据融合技术，以确保在多轮、多图像对话中的无缝交互。与现有框架相比，DeepSpeed-VisualChat展现出卓越的可扩展性，可扩展至70B参数的语言模型规模，代表了多模态语言模型的重大进步，并为未来的探索奠定了坚实基础。

English

Most of the existing multi-modal models, hindered by their incapacity to adeptly manage interleaved image-and-text inputs in multi-image, multi-round dialogues, face substantial constraints in resource allocation for training and data accessibility, impacting their adaptability and scalability across varied interaction realms. To address this, we present the DeepSpeed-VisualChat framework, designed to optimize Large Language Models (LLMs) by incorporating multi-modal capabilities, with a focus on enhancing the proficiency of Large Vision and Language Models in handling interleaved inputs. Our framework is notable for (1) its open-source support for multi-round and multi-image dialogues, (2) introducing an innovative multi-modal causal attention mechanism, and (3) utilizing data blending techniques on existing datasets to assure seamless interactions in multi-round, multi-image conversations. Compared to existing frameworks, DeepSpeed-VisualChat shows superior scalability up to 70B parameter language model size, representing a significant advancement in multi-modal language models and setting a solid foundation for future explorations.

DeepSpeed-VisualChat：通过多模态因果注意力实现多轮多图像交错对话

DeepSpeed-VisualChat: Multi-Round Multi-Image Interleave Chat via Multi-Modal Causal Attention

摘要

Support