SOTAVerified

Conditional Text Generation

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

Papers

Showing 2130 of 67 papers

TitleStatusHype
Token Manipulation Generative Adversarial Network for Text GenerationCode0
Generating Text through Adversarial Training using Skip-Thought VectorsCode0
Semantic Sensitivities and Inconsistent Predictions: Measuring the Fragility of NLI ModelsCode0
Exploring Conditional Text Generation for Aspect-Based Sentiment AnalysisCode0
Pragmatically Informative Text GenerationCode0
Evaluation of Language Models in the Medical Context Under Resource-Constrained SettingsCode0
ETC-NLG: End-to-end Topic-Conditioned Natural Language GenerationCode0
Encoder-Agnostic Adaptation for Conditional Language GenerationCode0
IPAD: Iterative, Parallel, and Diffusion-based Network for Scene Text RecognitionCode0
Pre-train and Plug-in: Flexible Conditional Text Generation with Variational Auto-EncodersCode0
Show:102550
← PrevPage 3 of 7Next →

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1GPT-2-no-fine-tuningIgnored Constraint Error Rate28.2Unverified
2GPT-2-fine-tuned-5-epochsIgnored Constraint Error Rate0.5Unverified
3GPT-2-fine-tuned-20-epochsIgnored Constraint Error Rate0.3Unverified
4GPT-2-with-filterIgnored Constraint Error Rate0Unverified