[论文解读] Interactive Natural Language Processing

Zekun Wang, Ge Zhang+20 more|arXiv (Cornell University)|May 22, 2023

Topic Modeling被引用 22

一句话总结

本论文综述了 Interactive NLP (iNLP)，定义一个统一框架，在该框架中，语言模型作为代理观察、行动并从人类、知识库、工具和环境中获得反馈，以提升性能与对齐。

ABSTRACT

Interactive Natural Language Processing (iNLP) has emerged as a novel paradigm within the field of NLP, aimed at addressing limitations in existing frameworks while aligning with the ultimate goals of artificial intelligence. This paradigm considers language models as agents capable of observing, acting, and receiving feedback iteratively from external entities. Specifically, language models in this context can: (1) interact with humans for better understanding and addressing user needs, personalizing responses, aligning with human values, and improving the overall user experience; (2) interact with knowledge bases for enriching language representations with factual knowledge, enhancing the contextual relevance of responses, and dynamically leveraging external information to generate more accurate and informed responses; (3) interact with models and tools for effectively decomposing and addressing complex tasks, leveraging specialized expertise for specific subtasks, and fostering the simulation of social behaviors; and (4) interact with environments for learning grounded representations of language, and effectively tackling embodied tasks such as reasoning, planning, and decision-making in response to environmental observations. This paper offers a comprehensive survey of iNLP, starting by proposing a unified definition and framework of the concept. We then provide a systematic classification of iNLP, dissecting its various components, including interactive objects, interaction interfaces, and interaction methods. We proceed to delve into the evaluation methodologies used in the field, explore its diverse applications, scrutinize its ethical and safety issues, and discuss prospective research directions. This survey serves as an entry point for researchers who are interested in this rapidly evolving area and offers a broad view of the current landscape and future trajectory of iNLP.

研究动机与目标

• 提供 iNLP 的统一定义与形成作为一种新的 NLP 范式。
• 就交互对象、接口与方法的全面分类。
• 综述在 iNLP 中的评估方法、应用及伦理/安全考量。
• 讨论未来方向与挑战，以引导这一快速发展领域的研究者。

提出的方法

• 提出一个统一的 iNLP 框架，在该框架中 LMs 作为观察、行动并从外部对象获取反馈的代理。
• 将交互对象分类为人、知识库、模型/工具和环境，并描述它们的作用。
• 系统性回顾交互接口（自然语言/形式语言、编辑、机语言、共享内存）和交互方法（提示、微调、强化学习、主动学习/模仿学习等）。
• 分析每种交互类型的评估策略，并总结当前的应用和安全关切。
• 识别未来的研究方向，以推进 iNLP 的具身化、社会互动和 grounding 的推进。

实验结果

研究问题

• RQ1 什么构成 Interactive NLP 的统一、正式定义与框架？
• RQ2 iNLP 如何通过交互对象、接口和方法进行系统分类？
• RQ3 在 iNLP 的各类交互中，哪些评估方法和伦理考虑是相关的？
• RQ4 iNLP 的主要应用领域与未来方向是什么？

主要发现

• iNLP 通过将语言模型视为与人类、知识库、工具/模型和环境互动的代理，来解决幻觉、对齐和 grounding 问题，从而扩展了传统 NLP。
• 交互接口的分类包括自然语言、形式语言、编辑、机语言和共享内存。
• 交互方法的分类包括提示、微调、RL、主动学习和模仿学习，以及消息融合策略。
• 知识库交互涵盖语料库、互联网以及生成/隐式知识检索，并考虑检索质量与噪声。
• 模型/工具与环境促进任务分解、模块化子任务，以及代理之间的协作以解决复杂问题。