本文介绍了一种新型系统NarrativePlay,它能让用户扮演虚构角色,在沉浸式环境中与小说等叙事作品里的其他角色互动。我们利用大语言模型(Large Language Models,LLMs),以从叙事文本中提取的性格特征为引导,生成类人回应。该系统整合了自动生成的叙事场景视觉展示、角色肖像和角色语音,大幅提升了用户体验。我们的方法摒弃了预定义沙盒,转而聚焦于从用户所选角色视角,从叙事文本中提取的主线剧情事件。NarrativePlay已在侦探和冒险两类叙事作品上完成评估,用户既可以探索世界,也能通过对话提升自己与叙事角色的好感度。
英文摘要:
In this paper, we introduce NarrativePlay, a novel system that allows users to role-play a fictional character and interact with other characters in narratives such as novels in an immersive environment. We leverage Large Language Models (LLMs) to generate human-like responses, guided by personality traits extracted from narratives. The system incorporates auto-generated visual display of narrative settings, character portraits, and character speech, greatly enhancing user experience. Our approach eschews predefined sandboxes, focusing instead on main storyline events extracted from narratives from the perspective of a user-selected character. NarrativePlay has been evaluated on two types of narratives, detective and adventure stories, where users can either explore the world or improve their favorability with the narrative characters through conversations.
Can LLM Agents Stick to the Script? A Benchmark for Long-Horizon Consistency in Interactive Narratives
University of Macau(澳门大学);Westlake University(西湖大学);Harbin Institute of Technology(哈尔滨工业大学);University of Cambridge(剑桥大学);University of Aberdeen(阿伯丁大学)
NARRA-Gym for Evaluating Interactive Narrative Agents
University of Notre Dame(诺丁汉大学);LMU Munich(慕尼黑大学);Munich Center for Machine Learning(慕尼黑机器学习中心);University of Pennsylvania(宾夕法尼亚大学);Lehigh University(莱恩大学);Massachusetts Institute of Technology(麻省理工学院);Bake AI UC Santa Barbara(加州大学圣芭芭拉分校);University of Washington(华盛顿大学)
Reframe Your Life Story: Interactive Narrative Therapist and Innovative Moment Assessment with Large Language Models
Beijing Key Laboratory of Traffic Data Mining and Embodied Intelligence(北京交通大数据挖掘与具身智能重点实验室);Beijing Jiaotong University(北京交通大学);CoAI Group(CoAI集团);DCST(国防科技大学);IAI(人工智能院);BNRIST(北航机器人与智能技术研究院);Tsinghua University(清华大学);Tencent Jarvis Lab(腾讯Jarvis实验室);Singapore University of Technology and Design(新加坡科技与设计大学);Central South University(中南大学);Lingxin AI(灵犀AI)
BIASEDTALES-ML: A Multilingual Dataset for Analyzing Narrative Attribute Distributions in LLM-Generated Stories
School of Computer Science and Engineering, Northeastern University, Shenyang, China(东北大学计算机科学与工程学院,中国沈阳);NiuTrans Research, Shenyang, China(NiuTrans研究院,中国沈阳)
AI Stories: An Interactive Narrative System for Children
The University of Antwerp(安特卫普大学)
AI Stories是一个建议的互动对话系统,让儿童通过对话共同创造叙事世界。在接下来的三年里,该系统将在儿科病房内开发和测试,在那里它提供了教育和游戏之间的差距有用的资源。讲故事和编故事是语言游戏的一个基本组成部分,它的健谈和无意义的品质是很重要的;因此,自动化系统提供的开场白对儿童是有益的。在本文中,我将介绍这个项目的现状,在其更多的实验和一般的幌子。从概念上讲,通过对话讲故事涉及到故事的预印本解释,超越了静态和线性的媒介,在那里故事是表演的,时间的和社会的。
OPEN-THEATRE: An Open-Source Toolkit for LLM-based Interactive Drama
UM–SJTU Joint Institute, Shanghai Jiao Tong University(上海交通大学与UM联合研究所);AGI Institute, School of Computer Science, Shanghai Jiao Tong University(上海交通大学人工智能研究所);Key Laboratory of Shanghai Education Commission for Intelligent Interaction and Cognitive Engineering, Shanghai Jiao Tong University(上海教育委员会智能交互与认知工程重点实验室);Shanghai Key Laboratory of Trusted Data Circulation and Governance in Web3(上海Web3可信数据流通与治理重点实验室)
Aligning to Social Norms and Values in Interactive Narratives
Allen Institute for Artificial Intelligence(艾伦人工智能研究所);University of Washington(华盛顿大学)
我们专注于创建代理,在互动叙事或基于文本的游戏中与社会有益的规范和价值观保持一致-环境中,代理通过自然语言感知并与世界互动。这种交互式代理通常通过强化学习进行训练,以优化任务性能,即使这种奖励可能导致违反社会规范的代理行为-对代理本身或环境中的其他实体造成伤害。社会价值对齐是指创造行为符合特定背景和人群预期道德和社会规范的代理人-在我们的情况下,它意味着代理人的行为方式对自己和他人的危害较小,更有益。我们建立在Jiminy Cricket基准(Hendrycks et al. 2021)的基础上,这是一组25个注释的交互式叙述,包含数千个道德上突出的场景,涵盖了从盗窃和身体伤害到利他主义的一切。我们引入GALAD(游戏-价值对齐通过行动蒸馏)代理,它使用专门训练的语言模型中存在的社会常识知识,根据上下文将其动作空间限制为仅与社会有益价值对齐的那些动作。一项实验研究表明,GALAD代理有效地做出决策,足以将最先进的任务性能提高4%,同时与强大的当代价值对齐方法相比,将社会有害行为的频率降低25%。
Towards Enhanced Immersion and Agency for LLM-based Interactive Drama
School of Computer Science, Shanghai Jiao Tong University(上海交通大学计算机科学学院);University of Zurich(苏黎世大学);Key Laboratory of Shanghai Education Commission for Intelligent Interaction and Cognitive Engineering, Shanghai Jiao Tong University(上海教育委员会智能交互与认知工程重点实验室,上海交通大学);Shanghai Key Laboratory of Trusted Data Circulation and Governance in Web3(上海Web3可信数据流通与治理重点实验室)
MBZUAI, UAE(阿联酋马斯喀特大学);Nazarbayev University, Kazakhstan(哈萨克斯坦纳扎尔巴耶夫大学);University of Maryland, USA(美国马里兰大学);University of Arizona, USA(美国亚利桑那大学);Athens University of Economics and Business, Greece(希腊雅典经济与商业大学);University of Porto, Portugal(葡萄牙波尔图大学);Sofia University "St. Kliment Ohridski", Bulgaria(保加利亚索菲亚大学);University of Helsinki, Finland(芬兰赫尔辛基大学);Beira Interior and INESC TEC, Portugal(葡萄牙贝拉内尔和INESC TEC);Porto and INESC TEC, Portugal(葡萄牙波尔图和INESC TEC);University of Padova, Italy(意大利帕多瓦大学);European Commission Joint Research Center, Italy(意大利欧洲委员会联合研究中心);Polish Academy of Sciences, Poland(波兰科学院)