HyperAI
HomeNewsLatest PapersTutorialsDatasetsWikiSOTALLM ModelsGPU LeaderboardEvents
Search
About
HyperAI

Text Simplification On Asset

Metrics

BLEU
SARI (EASSEu003e=0.2.1)

Results

Performance results of various models on this benchmark

Model Name
BLEU
SARI (EASSEu003e=0.2.1)
Paper TitleRepository
ACCESS75.99*40.13Controllable Sentence Simplification
BART--The GEM Benchmark: Natural Language Generation, its Evaluation and Metrics-
T5--The GEM Benchmark: Natural Language Generation, its Evaluation and Metrics-
PBMT-R79.39*34.63--
Control Prefixes (BART)-43.58Control Prefixes for Parameter-Efficient Text Generation
TST-43.21Text Simplification by Tagging
Dress-LS86.39*36.59Sentence Simplification with Deep Reinforcement Learning
GPT-175B (15 SARI-selected examples, random ordering)73.9247.94Metric-Based In-context Learning: A Case Study in Text Simplification
UNTS (Unsupervised)76.14*35.19Unsupervised Neural Text Simplification
MUSS (BART+ACCESS Unsupervised)-42.65MUSS: Multilingual Unsupervised Sentence Simplification by Mining Paraphrases
MUSS (BART+ACCESS Supervised)72.9844.15MUSS: Multilingual Unsupervised Sentence Simplification by Mining Paraphrases
DMASS-DCSS71.44*38.67Integrating Transformer and Paraphrase Rules for Sentence Simplification
0 of 12 row(s) selected.