ActionPiece:面向生成式推荐的动作序列上下文分词方法
ActionPiece: Contextually Tokenizing Action Sequences for Generative Recommendation
AI总结:
本文提出 ActionPiece 方法,通过基于物品特征共现频率的上下文分词和集合置换正则化,解决生成式推荐中动作分词缺乏上下文感知的问题,提升模型性能。
摘要:
生成式推荐(GR)是一种新兴范式,其将用户动作分词为离散的 token 模式,并以自回归方式生成预测。然而,现有的 GR 模型对每个动作独立进行分词,在所有序列中为相同动作分配相同的固定 token,而不考虑上下文关系。这种上下文感知的缺失可能导致性能欠佳,因为同一动作根据其周围上下文可能具有不同含义。为解决该问题,我们提出 ActionPiece,在动作序列分词时显式引入上下文。在 ActionPiece 中,每个动作被表示为一组物品特征。给定动作序列语料库,我们根据特征模式在单个集合内部以及相邻集合之间的共现频率,将特征模式合并为新 token 来构建词表。考虑到特征集合的无序性,我们进一步引入集合置换正则化,从而生成具有相同语义的多种动作序列切分。我们的代码可在 https://github.com/google-deepmind/action_piece 获取。
英文摘要:
Generative recommendation (GR) is an emerging paradigm where user actions are tokenized into discrete token patterns and autoregressively generated as predictions. However, existing GR models tokenize each action independently, assigning the same fixed tokens to identical actions across all sequences without considering contextual relationships. This lack of context-awareness can lead to suboptimal performance, as the same action may hold different meanings depending on its surrounding context. To address this issue, we propose ActionPiece to explicitly incorporate context when tokenizing action sequences. In ActionPiece, each action is represented as a set of item features. Given the action sequence corpora, we construct the vocabulary by merging feature patterns as new tokens, based on their co-occurrence frequency both within individual sets and across adjacent sets. Considering the unordered nature of feature sets, we further introduce set permutation regularization, which produces multiple segmentations of action sequences with the same semantics. Our code is available at: https://github.com/google-deepmind/action_piece.
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。