SOTAVerified

Text Generation

Text Generation is the task of generating text with the goal of appearing indistinguishable to human-written text. This task is more formally known as "natural language generation" in the literature.

Text generation can be addressed with Markov processes or deep generative models like LSTMs. Recently, some of the most advanced methods for text generation include BART, GPT and other GAN-based approaches. Text generation systems are evaluated either through human ratings or automatic evaluation metrics like METEOR, ROUGE, and BLEU.

Further readings:

( Image credit: Adversarial Ranking for Language Generation )

Papers

Showing 41014150 of 5335 papers

TitleStatusHype
PyOpenDial: A Python-based Domain-Independent Toolkit for Developing Spoken Dialogue Systems with Probabilistic Rules0
Deep Copycat Networks for Text-to-Text GenerationCode0
Enhancing Neural Data-To-Text Generation Models with External Background Knowledge0
Enhanced Transformer Model for Data-to-Text Generation0
A Recurrent BERT-based Model for Question GenerationCode0
Answering Naturally: Factoid to Full length Answer GenerationCode0
Fine-Grained Control of Sentence Segmentation and Entity Positioning in Neural NLG0
Generating Natural Anagrams: Towards Language Generation Under Hard Combinatorial Constraints0
Kernelized Bayesian Softmax for Text GenerationCode0
Enhancing Variational Autoencoders with Mutual Information Neural Estimation for Text Generation0
Stick to the Facts: Learning towards a Fidelity-oriented E-Commerce Product Description Generation0
Back-Translation as Strategy to Tackle the Lack of Corpus in Natural Language Generation from Semantic Representations0
Controlling Sequence-to-Sequence Models - A Demonstration on Neural-based Acrostic Generator0
Natural Language Generation for Effective Knowledge DistillationCode0
Set to Ordered Text: Generating Discharge Instructions from Medical Billing Codes0
Hello, It's GPT-2 - How Can I Help You? Towards the Use of Pretrained Language Models for Task-Oriented Dialogue Systems0
Multi-task Learning for Natural Language Generation in Task-Oriented Dialogue0
Neural Assistant: Joint Action Prediction, Response Generation, and Latent Knowledge Reasoning0
Can adversarial training learn image captioning ?Code0
Naver Labs Europe's Systems for the Document-Level Generation and Translation Task at WNGT 20190
An Augmented Transformer Architecture for Natural Language Generation Tasks0
Big Bidirectional Insertion Representations for Documents0
Training ASR models by Generation of Contextual Information0
ViGGO: A Video Game Corpus for Data-To-Text Generation in Open-Domain Conversation0
Sticking to the Facts: Confident Decoding for Faithful Data-to-Text Generation0
PLATO: Pre-trained Dialogue Generation Model with Discrete Latent VariableCode0
Analyzing the Forgetting Problem in the Pretrain-Finetuning of Dialogue Response Models0
Automatic Quality Estimation for Natural Language Generation: Ranting (Jointly Rating and Ranking)Code0
Is Multilingual BERT Fluent in Language Generation?Code0
SMArT: Training Shallow Memory-aware Transformers for Robotic Explainability0
Learning from Fact-checkers: Analysis and Generation of Fact-checking LanguageCode0
Mapping Natural-language Problems to Formal-language Solutions Using Structured Neural RepresentationsCode0
Controlled Text Generation for Data Augmentation in Intelligent Artificial Agents0
Template-free Data-to-Text Generation of Finnish Sports NewsCode0
Clinical Text Generation through Leveraging Medical Concept and Relations0
Tree-Structured Semantic Encoder with Knowledge Sharing for Domain Adaptation in Natural Language Generation0
QTUNA: A Corpus for Understanding How Speakers Use QuantificationCode0
BERT for Question Generation0
TMLab: Generative Enhanced Model (GEM) for adversarial attacks0
Read, Attend and Comment: A Deep Architecture for Automatic NewsComment Generation0
Procedural Text Generation from a Photo Sequence0
What goes into a word: generating image descriptions with top-down spatial knowledge0
Proceedings of the 12th International Conference on Natural Language Generation0
Low Level Linguistic Controls for Style Transfer and Content Preservation0
Generation of Hip-Hop Lyrics with Hierarchical Modeling and Conditional Templates0
A Closer Look at Recent Results of Verb Selection for Data-to-Text NLG0
Best practices for the human evaluation of automatically generated text0
Generating Quantified Descriptions of Abstract Visual Scenes0
Using NLG for speech synthesis of mathematical sentences0
Hotel Scribe: Generating High Variation Hotel Descriptions0
Show:102550
← PrevPage 83 of 107Next →

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1T5B BaselineBLEU48.74Unverified
2FactT5BBLEU48.37Unverified
3JointGT BaselineBLEU47.51Unverified
4FactJointGTBLEU47.39Unverified
5Control Prefixes (T5-large)METEOR0.41Unverified
6T5METEOR0.12Unverified
7BARTMETEOR0.11Unverified
#ModelMetricClaimedVerifiedStatus
1LeakGANBLEU-20.95Unverified
2partGANBLEU-20.91Unverified
3RankGANBLEU-20.85Unverified
4RelGAN (100)BLEU-20.85Unverified
5SeqGANBLEU-20.83Unverified
#ModelMetricClaimedVerifiedStatus
1LeakGANBLEU-20.96Unverified
2PPOGANBLEU-20.91Unverified
3RelGANBLEU-20.88Unverified
4SeqGANBLEU-20.86Unverified
5RankGANBLEU-20.78Unverified
#ModelMetricClaimedVerifiedStatus
1UniCRSDistinct-30.65Unverified
2CRFRDistinct-30.52Unverified
3KGSFDistinct-30.43Unverified
4C2CRSDistinct-30.33Unverified
5KBRDDistinct-30.3Unverified
#ModelMetricClaimedVerifiedStatus
1UniLMCIDEr14.92Unverified
2BART (TextBox 2.0)CIDEr12.98Unverified
3BARTMETEOR0.3Unverified
4T5METEOR0.29Unverified
#ModelMetricClaimedVerifiedStatus
1Beam search + A*esque (beam)BLEU-134.4Unverified
2Beam search + A*esque (sample)BLEU-134.4Unverified
3Beam search + A*esque (greedy)BLEU-134.3Unverified
4Beam searchBLEU-133.7Unverified
#ModelMetricClaimedVerifiedStatus
1RankGANBLEU-20.81Unverified
2SeqGANBLEU-20.74Unverified
3LeakGANBLEU-20.46Unverified
#ModelMetricClaimedVerifiedStatus
1TGen++METEOR0.17Unverified
2TGenMETEOR0.15Unverified
3TGen+METEOR0.15Unverified
#ModelMetricClaimedVerifiedStatus
1GPT2-124Meval_loss3.12Unverified
2GPT2-81M-LOOPeval_loss3.11Unverified
3GPT2-Hermiteeval_loss2.91Unverified
#ModelMetricClaimedVerifiedStatus
1LLaMA-65B+CFG (zero-shot)Accuracy96.6Unverified
2LLaMA-30B+CFG (zero-shot)Accuracy96.4Unverified
3LLaMA-13B+CFG (zero-shot)Accuracy95.1Unverified
#ModelMetricClaimedVerifiedStatus
1CNN-VAENLL332.1Unverified
2SA-VAENLL327.5Unverified
3Aggressive VAENLL326.7Unverified
#ModelMetricClaimedVerifiedStatus
1BART (TextBox 2.0)BLEU-410.2Unverified
#ModelMetricClaimedVerifiedStatus
1STWGAN-GPBLEU-30.62Unverified
#ModelMetricClaimedVerifiedStatus
1PALMROUGE-L41.41Unverified
#ModelMetricClaimedVerifiedStatus
1BART (TextBox 2.0)ROUGE-L64.34Unverified
#ModelMetricClaimedVerifiedStatus
1AEM+AttentionBLEU-114.17Unverified
#ModelMetricClaimedVerifiedStatus
1GPT-4ASR65.1Unverified
#ModelMetricClaimedVerifiedStatus
1BART (TextBox 2.0)ROUGE-L42.96Unverified
#ModelMetricClaimedVerifiedStatus
1Graph2SeqBLEU22Unverified
#ModelMetricClaimedVerifiedStatus
1WGANGP + DGflowJS-40.19Unverified