From Past To Path: Masked History Learning for Next-Item Prediction in Generative Recommendation

ACL ARR 2026 January Submission3661 Authors

Keywords: Generative Recommendation

Abstract:

Generative recommendation, which directly generates item identifiers, has emerged as a promising paradigm for recommendation systems. However, this left-to-right paradigm inherently biases the model towards local contexts, failing to capture the bidirectional dependencies necessary for understanding complex user intents. To address this limitation, we propose Masked History Learning (MHL), a novel training framework that shifts the objective from simple next-step prediction to deep comprehension of history. MHL augments the standard autoregressive objective with an auxiliary task of reconstructing masked historical items, compelling the model to understand ‘’why'' an item path is formed from the user's past behaviors, rather than just ‘’what'' item comes next.
We introduce two key contributions to enhance this framework: (1) an entropy-guided masking policy that intelligently targets the most informative historical items for reconstruction, and (2) a curriculum learning scheduler that progressively transitions from history reconstruction to future prediction.
Experiments on three public datasets show that our method significantly outperforms state-of-the-art generative models, highlighting that a comprehensive understanding of the past is crucial for accurately predicting a user's future path. The code will be released. Anonymous code is available at https://anonymous.4open.science/r/MHL_anonymous-57C7

Paper Type: Long

Research Area: Information Extraction and Retrieval

Research Area Keywords: self-supervised learning, generative models

Contribution Types: NLP engineering experiment, Publicly available software and/or pre-trained models

Languages Studied: English

Submission Number: 3661