KERMIT: Generative Insertion-Based Modeling for Sequences

by   William Chan, et al.

We present KERMIT, a simple insertion-based approach to generative modeling for sequences and sequence pairs. KERMIT models the joint distribution and its decompositions (i.e., marginals and conditionals) using a single neural network and, unlike much prior work, does not rely on a prespecified factorization of the data distribution. During training, one can feed KERMIT paired data (x, y) to learn the joint distribution p(x, y), and optionally mix in unpaired data x or y to refine the marginals p(x) or p(y). During inference, we have access to the conditionals p(x | y) and p(y | x) in both directions. We can also sample from the joint distribution or the marginals. The model supports both serial fully autoregressive decoding and parallel partially autoregressive decoding, with the latter exhibiting an empirically logarithmic runtime. We demonstrate through experiments in machine translation, representation learning, and zero-shot cloze question answering that our unified approach is capable of matching or exceeding the performance of dedicated state-of-the-art systems across a wide range of tasks without the need for problem-specific architectural adaptation.


page 1

page 2

page 3

page 4


Insertion Transformer: Flexible Sequence Generation via Insertion Operations

We present the Insertion Transformer, an iterative, partially autoregres...

FlowSeq: Non-Autoregressive Conditional Sequence Generation with Generative Flow

Most sequence-to-sequence (seq2seq) models are autoregressive; they gene...

Inference Strategies for Machine Translation with Conditional Masking

Conditional masked language model (CMLM) training has proven successful ...

Examining Scaling and Transfer of Language Model Architectures for Machine Translation

Natural language understanding and generation models follow one of the t...

Enhancing Generalized Zero-Shot Learning via Adversarial Visual-Semantic Interaction

The performance of generative zero-shot methods mainly depends on the qu...

A Unified Deep Model of Learning from both Data and Queries for Cardinality Estimation

Cardinality estimation is a fundamental problem in database systems. To ...

Mode recovery in neural autoregressive sequence modeling

Despite its wide use, recent studies have revealed unexpected and undesi...