English
Related papers

Related papers: Discovering Non-monotonic Autoregressive Orderings…

200 papers

We explore deep autoregressive Transformer models in language modeling for speech recognition. We focus on two aspects. First, we revisit Transformer model configurations specifically for language modeling. We show that well configured…

Computation and Language · Computer Science 2019-09-25 Kazuki Irie , Albert Zeyer , Ralf Schlüter , Hermann Ney

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 sequence generation in…

Computation and Language · Computer Science 2019-10-29 Jiatao Gu , Qi Liu , Kyunghyun Cho

Existing sentence ordering approaches generally employ encoder-decoder frameworks with the pointer net to recover the coherence by recurrently predicting each sentence step-by-step. Such an autoregressive manner only leverages unilateral…

Computation and Language · Computer Science 2023-10-20 Yi Bin , Wenhao Shi , Bin Ji , Jipeng Zhang , Yujuan Ding , Yang Yang

Autoregressive speech synthesis often adopts a left-to-right order, yet generation order is a modelling choice. We investigate decoding order through masked diffusion framework, which progressively unmasks positions and allows arbitrary…

Sound · Computer Science 2026-01-14 Minghui Zhao , Anton Ragni

Neural language models process sequences of words, but the mathematical operations inside them are insensitive to the order in which words appear. Positional encodings are the component added to remedy this. Despite their importance,…

Machine Learning · Computer Science 2026-04-08 Giansalvo Cirrincione

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 ordering of the…

Computation and Language · Computer Science 2019-02-12 Mitchell Stern , William Chan , Jamie Kiros , Jakob Uszkoreit

Modern large language models (LLMs) excel at tasks that require storing and retrieving knowledge, such as factual recall and question answering. Transformers are central to this capability because they can encode information during training…

Machine Learning · Statistics 2026-03-18 Nuri Mert Vural , Alberto Bietti , Mahdi Soltanolkotabi , Denny Wu

A latent-variable model is introduced for text matching, inferring sentence representations by jointly optimizing generative and discriminative objectives. To alleviate typical optimization challenges in latent-variable models for text, we…

Computation and Language · Computer Science 2017-11-23 Dinghan Shen , Yizhe Zhang , Ricardo Henao , Qinliang Su , Lawrence Carin

Most large language models are autoregressive: they generate tokens one at a time. Discrete diffusion language models can generate multiple tokens in parallel, but sampling from them requires a denoising order: a strategy for deciding which…

Artificial Intelligence · Computer Science 2026-03-27 Daniel Israel , Tian Jin , Ellie Cheng , Guy Van den Broeck , Aditya Grover , Suvinay Subramanian , Michael Carbin

Attention-based encoder-decoder (AED) models have achieved promising performance in speech recognition. However, because the decoder predicts text tokens (such as characters or words) in an autoregressive manner, it is difficult for an AED…

Computation and Language · Computer Science 2021-08-31 Ye Bai , Jiangyan Yi , Jianhua Tao , Zhengkun Tian , Zhengqi Wen , Shuai Zhang

We introduce a new way of learning to encode position information for non-recurrent models, such as Transformer models. Unlike RNN and LSTM, which contain inductive bias by loading the input tokens sequentially, non-recurrent models are…

Machine Learning · Computer Science 2020-03-23 Xuanqing Liu , Hsiang-Fu Yu , Inderjit Dhillon , Cho-Jui Hsieh

The ability of machine learning models to store input information in hidden layer vector embeddings, analogous to the concept of `memory', is widely employed but not well characterized. We find that language model embeddings typically…

Computation and Language · Computer Science 2026-05-20 Benjamin L. Badger

Word order, an essential property of natural languages, is injected in Transformer-based neural language models using position encoding. However, recent experiments have shown that explicit position encoding is not always useful, since some…

Computation and Language · Computer Science 2022-11-09 Karim Lasri , Alessandro Lenci , Thierry Poibeau

Undirected neural sequence models have achieved performance competitive with the state-of-the-art directed sequence models that generate monotonically from left to right in machine translation tasks. In this work, we train a policy that…

Computation and Language · Computer Science 2021-12-17 Yichen Jiang , Mohit Bansal

In Transformer-based neural machine translation (NMT), the positional encoding mechanism helps the self-attention networks to learn the source representation with order dependency, which makes the Transformer-based NMT achieve…

Computation and Language · Computer Science 2020-04-09 Kehai Chen , Rui Wang , Masao Utiyama , Eiichiro Sumita

In sequence-to-sequence learning, e.g., natural language generation, the decoder relies on the attention mechanism to efficiently extract information from the encoder. While it is common practice to draw information from only the last…

Computation and Language · Computer Science 2022-08-30 Fenglin Liu , Xuancheng Ren , Guangxiang Zhao , Chenyu You , Xuewei Ma , Xian Wu , Xu Sun

We show that language models' activations linearly encode when information was learned during training. Our setup involves creating a model with a known training order by sequentially fine-tuning Llama-3.2-1B on six disjoint but otherwise…

Machine Learning · Computer Science 2025-09-23 Dmitrii Krasheninnikov , Richard E. Turner , David Krueger

Non-autoregressive models greatly improve decoding speed over typical sequence-to-sequence models, but suffer from degraded performance. Infilling and iterative refinement models make up some of this gap by editing the outputs of a…

Audio and Speech Processing · Electrical Eng. & Systems 2020-10-28 Ethan A. Chi , Julian Salazar , Katrin Kirchhoff

We propose a conditional non-autoregressive neural sequence model based on iterative refinement. The proposed model is designed based on the principles of latent variable models and denoising autoencoders, and is generally applicable to any…

Machine Learning · Computer Science 2018-08-29 Jason Lee , Elman Mansimov , Kyunghyun Cho

Autoregressive sequence models based on deep neural networks, such as RNNs, Wavenet and the Transformer attain state-of-the-art results on many tasks. However, they are difficult to parallelize and are thus slow at processing long…

Machine Learning · Computer Science 2018-06-11 Łukasz Kaiser , Aurko Roy , Ashish Vaswani , Niki Parmar , Samy Bengio , Jakob Uszkoreit , Noam Shazeer
‹ Prev 1 2 3 10 Next ›