paper-with-me

Papers

Insertion Language Models: Sequence Generation with Arbitrary-Position Insertions

2025-05-09 · Dhruvesh Patel, Aishwarya Sahoo, Avinash Amballa, Tahira Naseem, Tim G. J. Rudner, Andrew McCallum

Autoregressive models (ARMs), which predict subsequent tokens one-by-one ``from left to right,'' have achieved significant success across a wide range of sequence generation tasks. However, they struggle to accurately represent sequences that require satisfying sophisticated constraints or whose sequential dependencies are better addressed by out-of-order generation. Masked Diffusion Models (MDMs) address some of these limitations, but the process of unmasking multiple tokens simultaneously in MDMs can introduce incoherences, and MDMs cannot handle arbitrary infilling constraints when the number of tokens to be filled in is not known in advance. In this work, we introduce Insertion Language Models (ILMs), which learn to insert tokens at arbitrary positions in a sequence -- that is, they select jointly both the position and the vocabulary element to be inserted. By inserting tokens one at a time, ILMs can represent strong dependencies between tokens, and their ability to generate sequences in arbitrary order allows them to accurately model sequences where token dependencies do not follow a left-to-right sequential structure. To train ILMs, we propose a tailored network parameterization and use a simple denoising objective. Our empirical evaluation demonstrates that ILMs outperform both ARMs and MDMs on common planning tasks. Furthermore, we show that ILMs outperform MDMs and perform on par with ARMs in an unconditional text generation task while offering greater flexibility than MDMs in arbitrary-length text infilling.

📄 PDF Abstract BibTeX arXiv:2505.05755

Code (0)

등록된 구현이 없습니다.

Tasks

DenoisingPositionText GenerationText Infilling

Methods 이 논문이 사용한 방법론

Diffusion Diffusion models generate samples by gradually removing noise from a signal, and their training objective can be expressed as a reweighted variational lower-bound…

Similar Papers 제목 키워드 기반

Insertion-based Decoding with automatically Inferred Generation Order

2019-02-04 · TACL 2019 3 · Jiatao Gu, Qi Liu, Kyunghyun Cho

Conventional neural autoregressive decoding commonly assumes a fixed left-to-right generation order, which may be sub-optimal. In this work, we propose a novel decoding algorithm -- InDIGO -- which supports flexible sequ…

Code GenerationMachine TranslationTranslation

Insertion Transformer: Flexible Sequence Generation via Insertion Operations

2019-02-08 · Mitchell Stern, William Chan, Jamie Kiros, Jakob Uszkoreit

We present the Insertion Transformer, an iterative, partially autoregressive model for sequence generation based on insertion operations. Unlike typical autoregressive models which rely on a fixed, often left-to-right or…

Machine TranslationTranslationvalid

Towards More Efficient Insertion Transformer with Fractional Positional Encoding

2021-12-12 · Zhisong Zhang, Yizhe Zhang, Bill Dolan

Auto-regressive neural sequence models have been shown to be effective across text generation tasks. However, their left-to-right decoding order prevents generation from being parallelized. Insertion Transformer (Stern e…

Text Generation

Beyond Masks: Efficient, Flexible Diffusion Language Models via Deletion-Insertion Processes

2026-03-04 · Fangyu Ding, Ding Ding, Sijin Chen, Kaibo Wang 외 arxiv

While Masked Diffusion Language Models (MDLMs) relying on token masking and unmasking have shown promise in language modeling, their computational efficiency and generation flexibility remain constrained by the masking p…

Computational Efficiency

Big Bidirectional Insertion Representations for Documents

2019-10-29 · WS 2019 11 · Lala Li, William Chan

The Insertion Transformer is well suited for long form text generation due to its parallel generation capabilities, requiring $O(\log_2 n)$ generation steps to generate $n$ tokens. However, modeling long sequences is dif…

FormSentenceText GenerationTranslation