章魚 v4：語言模型圖

Octopus v4: Graph of language models

April 30, 2024

作者: Wei Chen, Zhiyuan Li

cs.AI

摘要

語言模型在各種應用中表現出色，然而最複雜的模型通常是專有的。例如，OpenAI 的 GPT-4 和 Anthropic 的各種模型價格昂貴且消耗大量能源。相比之下，開源社區已經生產出具競爭力的模型，例如 Llama3。此外，針對特定領域的較小語言模型，如針對法律、醫療或金融任務定制的模型，已經超越了它們的專有對手。本文介紹了一種新方法，該方法利用功能標記來整合多個開源模型，每個模型都經過優化以適應特定任務。我們新開發的 Octopus v4 模型利用功能標記智能地將用戶查詢引導到最適合的垂直模型，並重新格式化查詢以達到最佳性能。Octopus v4 是 Octopus v1、v2 和 v3 模型的進化，擅長於選擇和參數理解以及重新格式化。此外，我們探索了將圖形作為一種多功能數據結構的使用，通過利用 Octopus 模型和功能標記的能力有效協調多個開源模型。請使用我們的開源 GitHub（https://www.nexa4ai.com/）來嘗試 Octopus v4 模型（https://huggingface.co/NexaAIDev/Octopus-v4），並貢獻給更大的語言模型圖形。通過啟用小於 10B 參數的模型，我們在相同級別模型中實現了 74.8 的 SOTA MMLU 得分。

English

Language models have been effective in a wide range of applications, yet the most sophisticated models are often proprietary. For example, GPT-4 by OpenAI and various models by Anthropic are expensive and consume substantial energy. In contrast, the open-source community has produced competitive models, like Llama3. Furthermore, niche-specific smaller language models, such as those tailored for legal, medical or financial tasks, have outperformed their proprietary counterparts. This paper introduces a novel approach that employs functional tokens to integrate multiple open-source models, each optimized for particular tasks. Our newly developed Octopus v4 model leverages functional tokens to intelligently direct user queries to the most appropriate vertical model and reformat the query to achieve the best performance. Octopus v4, an evolution of the Octopus v1, v2, and v3 models, excels in selection and parameter understanding and reformatting. Additionally, we explore the use of graph as a versatile data structure that effectively coordinates multiple open-source models by harnessing the capabilities of the Octopus model and functional tokens. Use our open-sourced GitHub (https://www.nexa4ai.com/) to try Octopus v4 models (https://huggingface.co/NexaAIDev/Octopus-v4), and contrite to a larger graph of language models. By activating models less than 10B parameters, we achieved SOTA MMLU score of 74.8 among the same level models.