29 citations · 152 across the 38 of their papers we have counts for
5 papers · 1 filter
Learning Instructions with Unlabeled Data for Zero-Shot Cross-Task Generalization
Yuxian Gu, Pei Ke, Xiaoyan Zhu +1
Training language models to learn from human instructions for zero-shot cross-task generalization has attracted much attention in NLP communities. Recently, instruction tuning (IT)…
Curriculum-Based Self-Training Makes Better Few-Shot Learners for Data-to-Text Generation
Pei Ke, Haozhe Ji, Zhenyu Yang +4
Despite the success of text-to-text pre-trained models in various natural language generation (NLG) tasks, the generation performance is largely restricted by the number of labeled…
Rethinking and Refining the Distinct Metric
Siyang Liu, Sahand Sabour, Yinhe Zheng +3
Distinct- score\cite{Li2016} is a widely used automatic metric for evaluating diversity in language generation tasks. However, we observed that the original approach for calcula…
CTRLEval: An Unsupervised Reference-Free Metric for Evaluating Controlled Text Generation
Pei Ke, Hao Zhou, Yankai Lin +4
Existing reference-free metrics have obvious limitations for evaluating controlled text generation models. Unsupervised metrics can only provide a task-agnostic evaluation result w…
EVA2.0: Investigating Open-Domain Chinese Dialogue Systems with Large-Scale Pre-Training
Yuxian Gu, Jiaxin Wen, Hao Sun +8
Large-scale pre-training has shown remarkable performance in building open-domain dialogue systems. However, previous works mainly focus on showing and evaluating the conversationa…