arXivDaily arXiv每日学术速递 周一至周五更新
​通知:联邦节假日调休:哥伦布日 / 原住民日(Columbus Day / Indigenous Peoples' Day),今日(13日)无更新。
arXiv 2304.06634cs.CLcs.AIcs.LGcs.SI

PGTask:提出对话生成用户画像任务

PGTask: Introducing the Task of Profile Generation from Dialogues

INESC-ID(里斯本系统与计算机工程研究所);Instituto Superior Técnico, Universidade de Lisboa(里斯本大学高等技术学院)

Rui Ribeiro, Joao P. Carvalho, Luísa Coheur

AI总结:

针对对话系统中用户画像信息稀缺难获取的问题,提出PGTask画像生成任务,构建含对齐画像与对话的新数据集并基于SOTA方法建立基准,揭示任务挑战以推动新研究方向。

摘要:

近期研究尝试将用户画像信息融入模型以实现对话系统个性化,但这类知识稀缺且难以获取,因此从对话中提取/生成画像信息成为一项核心能力。为突破这一限制,我们提出用户画像生成任务(Profile Generation Task,PGTask)。我们为此任务构建了一个新数据集,包含从对话语料中提取的、与相关话语对齐的画像句子。此外,我们采用当前最优(state-of-the-art)方法在该新数据集上建立了画像生成基准。实验揭示了画像生成任务的挑战性,我们希望此举能开启一个新的研究方向。

英文摘要:

Recent approaches have attempted to personalize dialogue systems by leveraging profile information into models. However, this knowledge is scarce and difficult to obtain, which makes the extraction/generation of profile information from dialogues a fundamental asset. To surpass this limitation, we introduce the Profile Generation Task (PGTask). We contribute with a new dataset for this problem, comprising profile sentences aligned with related utterances, extracted from a corpus of dialogues. Furthermore, using state-of-the-art methods, we provide a benchmark for profile generation on this novel dataset. Our experiments disclose the challenges of profile generation, and we hope that this introduces a new research direction.

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。

↑