TGEA 2.0: A Large-Scale Diagnostically Annotated Dataset with Benchmark Tasks for Text Generation of Pretrained Language Models.
Huibin GeXiaohu ZhaoChuang LiuYulong ZengQun LiuDeyi XiongPublished in: NeurIPS (2022)
Keyphrases
- language model
- text generation
- language modeling
- n gram
- information retrieval
- probabilistic model
- document retrieval
- speech recognition
- retrieval model
- test collection
- language modelling
- smoothing methods
- statistical language models
- language model for information retrieval
- query expansion
- natural language generation
- natural language
- pseudo relevance feedback
- vector space model
- context sensitive
- query terms
- passage retrieval
- translation model
- term dependencies
- cross lingual
- language models for information retrieval
- spoken term detection
- co occurrence