natural-language-processing-5

Conditional Text Generation

3260 papers • 126 benchmarks • 313 datasets

The task of generating text according to some pre-specified conditioning (e.g. topic or sentiment or constraint)

(Image credit: Papersgraph)

Benchmarks

These leaderboards are used to track progress in conditional-text-generation-5

Trend

Dataset

Best Model

Actions

Lipogram-e

Libraries

i

Use these libraries to find conditional-text-generation-5 models and implementations

Datasets

VitaminC

WikiGraphs

Lipogram-e

Subtasks

No subtasks available.

Most implemented papers

Pragmatically Informative Text Generation

Sheng Shen, Jacob Andreas, D. Klein, Daniel Fried•Mon Apr 01 2019

This work considers two pragmatic modeling methods for text generation: one where pragmatics is imposed by information preservation, and another where prag matics isimposed by explicit modeling of distractors.

69

Content

0

Paper Graph

Extract, Denoise and Enforce: Evaluating and Improving Concept Preservation for Text-to-Text Generation

Jiawei Han, Yuning Mao, Xiang Ren, Wenchang Ma, Deren Lei•Sat Apr 17 2021

This paper proposes a simple yet effective framework to automatically extract, denoise, and enforce important input concepts as lexical constraints and performs comparably or better than its unconstrained counterpart on automatic metrics, demonstrates higher coverage for concept preservation, and receives better ratings in the human evaluation.

4 0

Paper Graph

GENIUS: Sketch-based Language Model Pre-training via Extreme and Selective Masking for Text Generation and Augmentation

Nan Duan, Yeyun Gong, Yelong Shen, Weizhu Chen, Biyang Guo, Songqiao Han, Hailiang Huang•Thu Nov 17 2022

GeniusAug is proposed, which first extracts the target-aware sketches from the original training set and then generates new samples based on the sketches, and is demonstrated to be a strong and ready-to-use data augmentation tool for various natural language processing (NLP) tasks.

24 0

Paper Graph

Generating Text through Adversarial Training Using Skip-Thought Vectors

Afroz Ahamad•Sun Aug 26 2018

The proposed architecture aims to reproduce writing style in the generated text by modelling the way of expression at a sentence level across all the works of an author.

11 0

Paper Graph

Encoder-Agnostic Adaptation for Conditional Language Generation

Alexander M. Rush, Sebastian Gehrmann, Zachary M. Ziegler, Luke Melas-Kyriazi•Sun Aug 18 2019

This work proposes an adaptation that directly injects arbitrary conditioning into self attention, an approach the authors call pseudo self attention that outperforms strong baselines, produces coherent generations, and is data efficient.

60 0

Paper Graph

Pre-train and Plug-in: Flexible Conditional Text Generation with Variational Auto-Encoders

Chenliang Li, Canwen Xu, Jiaxin Pei, Yu Duan•Sat Nov 09 2019

A new framework named Pre-train and Plug-in Variational Auto-Encoder (PPVAE) towards flexible conditional text generation that decouples the text generation module from the condition representation module to allow “one-to-many” conditional generation.

46 0

Paper Graph

ToTTo: A Controlled Table-To-Text Generation Dataset

Manaal Faruqui, Ankur P. Parikh, Bhuwan Dhingra, Dipanjan Das, Diyi Yang, Sebastian Gehrmann, Xuezhi Wang•Tue Apr 28 2020

An open-domain English table-to-text dataset with over 120,000 training examples that proposes a controlled generation task: given a Wikipedia table and a set of highlighted table cells, produce a one-sentence description.

421 0

Paper Graph

Token Manipulation Generative Adversarial Network for Text Generation

DaeJin Jo•Tue May 05 2020

This paper decomposes conditional text generation problem into two tasks, make-a-blank and fill-in-the-blank, and extends the former to handle more complex manipulations on the given tokens, and introduces a conditional adversarial learning that allows the agents to reach a goal, producing realistic texts, in cooperative setting.

1 0

Paper Graph

ETC-NLG: End-to-end Topic-Conditioned Natural Language Generation

Gabriele Sarti, Ginevra Carbone•Mon Aug 24 2020

This work presents ETC-NLG, an approach leveraging topic modeling annotations to enable fully-unsupervised End-to-end Topic-Conditioned Natural Language Generation over emergent topics in unlabeled document collections.

9 0

Paper Graph

Plug and Play Autoencoders for Conditional Text Generation

Noah A. Smith, Florian Mai, Nikolaos Pappas, Ivan Montero, James Henderson Idiap Research Institute, Epfl, U. Washington, Allen Institute for Artificial Intelligence•Mon Oct 05 2020

Evaluations on style transfer tasks both with and without sequence-to-sequence supervision show that the proposed plug and play Emb2Emb method performs better than or comparable to strong baselines while being up to four times faster.

38 0

Paper Graph

Adding a benchmark result helps the community track progress.