An Imitation Learning Curriculum for Text Editing with Non-Autoregressive Models

• URL: http://arxiv.org/abs/2203.09486v1
• Date: Thu, 17 Mar 2022 17:36:23 GMT
• Title: An Imitation Learning Curriculum for Text Editing with Non-Autoregressive Models
• Authors: Sweta Agrawal and Marine Carpuat
• Abstract summary: We show that imitation learning algorithms for machine translation introduce mismatches between training and inference that lead to undertraining and poor generalization in editing scenarios. We show the efficacy of these strategies on two challenging English editing tasks: controllable text simplification and abstractive summarization.
• Score: 22.996178360362734
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: We propose a framework for training non-autoregressive sequence-to-sequence models for editing tasks, where the original input sequence is iteratively edited to produce the output. We show that the imitation learning algorithms designed to train such models for machine translation introduces mismatches between training and inference that lead to undertraining and poor generalization in editing scenarios. We address this issue with two complementary strategies: 1) a roll-in policy that exposes the model to intermediate training sequences that it is more likely to encounter during inference, 2) a curriculum that presents easy-to-learn edit operations first, gradually increasing the difficulty of training samples as the model becomes competent. We show the efficacy of these strategies on two challenging English editing tasks: controllable text simplification and abstractive summarization. Our approach significantly improves output quality on both tasks and controls output complexity better on the simplification task.

Related papers

• Instruction Position Matters in Sequence Generation with Large Language Models [67.87516654892343]
  Large language models (LLMs) are capable of performing conditional sequence generation tasks, such as translation or summarization. We propose enhancing the instruction-following capability of LLMs by shifting the position of task instructions after the input sentences.
  arXiv  Detail & Related papers  (2023-08-23T12:36:57Z)
• On Conditional and Compositional Language Model Differentiable Prompting [75.76546041094436]
  Prompts have been shown to be an effective method to adapt a frozen Pretrained Language Model (PLM) to perform well on downstream tasks. We propose a new model, Prompt Production System (PRopS), which learns to transform task instructions or input metadata, into continuous prompts.
  arXiv  Detail & Related papers  (2023-07-04T02:47:42Z)
• Curriculum-Guided Abstractive Summarization [45.57561926145256]
  Recent Transformer-based summarization models have provided a promising approach to abstractive summarization. These models have two shortcomings: (1) they often perform poorly in content selection, and (2) their training strategy is not quite efficient, which restricts model performance. In this paper, we explore two ways to compensate for these pitfalls. First, we augment the Transformer network with a sentence cross-attention module in the decoder, encouraging more abstraction of salient content.
  arXiv  Detail & Related papers  (2023-02-02T11:09:37Z)
• Learning from Bootstrapping and Stepwise Reinforcement Reward: A Semi-Supervised Framework for Text Style Transfer [30.622772801446132]
  We propose a semi-supervised framework for text style transfer. First, the learning process is bootstrapped with supervision guided by automatically constructed pseudo-parallel pairs. Then the model learns from unlabeled data via reinforcement rewards.
  arXiv  Detail & Related papers  (2022-05-19T05:18:06Z)
• Text Revision by On-the-Fly Representation Optimization [76.11035270753757]
  Current state-of-the-art methods formulate these tasks as sequence-to-sequence learning problems. We present an iterative in-place editing approach for text revision, which requires no parallel data. It achieves competitive and even better performance than state-of-the-art supervised methods on text simplification.
  arXiv  Detail & Related papers  (2022-04-15T07:38:08Z)
• OCHADAI-KYODAI at SemEval-2021 Task 1: Enhancing Model Generalization and Robustness for Lexical Complexity Prediction [8.066349353140819]
  We propose an ensemble model for predicting the lexical complexity of words and multiword expressions. The model receives as input a sentence with a target word or MWEand outputs its complexity score. Our model achieved competitive results and ranked among the top-10 systems in both sub-tasks.
  arXiv  Detail & Related papers  (2021-05-12T09:27:46Z)
• COCO-LM: Correcting and Contrasting Text Sequences for Language Model Pretraining [59.169836983883656]
  COCO-LM is a new self-supervised learning framework that pretrains Language Models by COrrecting challenging errors and COntrasting text sequences. COCO-LM employs an auxiliary language model to mask-and-predict tokens in original text sequences. Our analyses reveal that COCO-LM's advantages come from its challenging training signals, more contextualized token representations, and regularized sequence representations.
  arXiv  Detail & Related papers  (2021-02-16T22:24:29Z)
• Controllable Text Simplification with Explicit Paraphrasing [88.02804405275785]
  Text Simplification improves the readability of sentences through several rewriting transformations, such as lexical paraphrasing, deletion, and splitting. Current simplification systems are predominantly sequence-to-sequence models that are trained end-to-end to perform all these operations simultaneously. We propose a novel hybrid approach that leverages linguistically-motivated rules for splitting and deletion, and couples them with a neural paraphrasing model to produce varied rewriting styles.
  arXiv  Detail & Related papers  (2020-10-21T13:44:40Z)
• Q-learning with Language Model for Edit-based Unsupervised Summarization [19.332743860240264]
  We propose a new approach based on Q-learning with an edit-based summarization. The method combines two key modules to form an Editorial Agent and Language Model converter. Q-learning is leveraged to train the agent to produce proper edit actions.
  arXiv  Detail & Related papers  (2020-10-09T05:47:00Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.