Lemur:为语言智能体协调自然语言与代码
Lemur: Harmonizing Natural Language and Code for Language Agents
AI总结:
提出开源语言模型Lemur与Lemur-Chat,通过融合自然语言与代码能力,在多项基准和智能体任务上表现优异,显著缩小了与专有模型的差距。
摘要:
我们介绍了Lemur和Lemur-Chat,它们是经过优化、同时具备自然语言和编程能力的开源语言模型,旨在作为通用语言智能体的骨干。从语言聊天模型向功能性语言智能体的演进,要求模型不仅要掌握人类交互、推理和规划,还必须确保在相关环境中的基础支撑。这就要求模型中自然语言和编程能力的和谐融合。提出Lemur和Lemur-Chat正是为了应对这一需求,它们在这两个领域展现出平衡的熟练度,而不同于往往只专精其一的现有开源模型。通过使用代码密集型语料库进行细致的预训练,并在文本和代码数据上进行指令微调,我们的模型在开源模型中的各种文本和编程基准测试上取得了最先进的平均性能。综合实验表明,Lemur优于现有的开源模型,并在涉及人类通信、工具使用以及完全可观测和部分可观测环境下交互的各种智能体任务中表现出色。自然语言和编程语言之间的协调使Lemur-Chat能够显著缩小与专有模型在智能体能力上的差距,为开发在推理、规划和跨环境无缝操作方面熟练的先进开源智能体提供了关键见解。
英文摘要:
We introduce Lemur and Lemur-Chat, openly accessible language models optimized for both natural language and coding capabilities to serve as the backbone of versatile language agents. The evolution from language chat models to functional language agents demands that models not only master human interaction, reasoning, and planning but also ensure grounding in the relevant environments. This calls for a harmonious blend of language and coding capabilities in the models. Lemur and Lemur-Chat are proposed to address this necessity, demonstrating balanced proficiencies in both domains, unlike existing open-source models that tend to specialize in either. Through meticulous pre-training using a code-intensive corpus and instruction fine-tuning on text and code data, our models achieve state-of-the-art averaged performance across diverse text and coding benchmarks among open-source models. Comprehensive experiments demonstrate Lemur's superiority over existing open-source models and its proficiency across various agent tasks involving human communication, tool usage, and interaction under fully- and partially- observable environments. The harmonization between natural and programming languages enables Lemur-Chat to significantly narrow the gap with proprietary models on agent abilities, providing key insights into developing advanced open-source agents adept at reasoning, planning, and operating seamlessly across environments. https://github.com/OpenLemur/Lemur
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。