Variational Learning for Insertion-based Generation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhang, Yangtian, Wang, Zhe, Gretton, Arthur, Ying, Rex, van Dijk, David, Titsias, Michalis K., Shi, Jiaxin
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910281649618944
author Zhang, Yangtian
Wang, Zhe
Gretton, Arthur
Ying, Rex
van Dijk, David
Titsias, Michalis K.
Shi, Jiaxin
author_facet Zhang, Yangtian
Wang, Zhe
Gretton, Arthur
Ying, Rex
van Dijk, David
Titsias, Michalis K.
Shi, Jiaxin
contents Non-monotonic sequence generation methods, such as masked diffusion models, provide a flexible alternative to left-to-right autoregressive modeling by allowing tokens to be generated in non-fixed and prescribed orders. Despite their practical advantages, most existing non-monotonic models are order-agnostic and rely on a fixed-length grid, limiting their ability to support variable-length generation and adaptive insertion order. In this work, we introduce a probabilistic framework for learning insertion order in variable-length insertion models. We formalize a bijective correspondence between insertion trajectories and permutations, which enables an exact reparameterization of the data likelihood as a sum over permutations. Building on this result, we propose the Insertion Process (IP), a stochastic generative model that jointly learns where to insert, what to insert, and when to terminate, trained via permutation-based variational inference. Unlike prior fixed-canvas approaches, IP natively supports variable-length generation and learns data-driven preferences over insertion orders. Experiments on goal-conditioned planning and molecular string generation demonstrate that learning insertion order improves both modeling quality and generalization in domains without a canonical left-to-right structure.
format Preprint
id arxiv_https___arxiv_org_abs_2606_02133
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Variational Learning for Insertion-based Generation
Zhang, Yangtian
Wang, Zhe
Gretton, Arthur
Ying, Rex
van Dijk, David
Titsias, Michalis K.
Shi, Jiaxin
Machine Learning
Artificial Intelligence
Non-monotonic sequence generation methods, such as masked diffusion models, provide a flexible alternative to left-to-right autoregressive modeling by allowing tokens to be generated in non-fixed and prescribed orders. Despite their practical advantages, most existing non-monotonic models are order-agnostic and rely on a fixed-length grid, limiting their ability to support variable-length generation and adaptive insertion order. In this work, we introduce a probabilistic framework for learning insertion order in variable-length insertion models. We formalize a bijective correspondence between insertion trajectories and permutations, which enables an exact reparameterization of the data likelihood as a sum over permutations. Building on this result, we propose the Insertion Process (IP), a stochastic generative model that jointly learns where to insert, what to insert, and when to terminate, trained via permutation-based variational inference. Unlike prior fixed-canvas approaches, IP natively supports variable-length generation and learns data-driven preferences over insertion orders. Experiments on goal-conditioned planning and molecular string generation demonstrate that learning insertion order improves both modeling quality and generalization in domains without a canonical left-to-right structure.
title Variational Learning for Insertion-based Generation
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2606.02133