← Search

Huibin Ge

2 accepted papers

2022

TGEA 2.0: A Large-Scale Diagnostically Annotated Dataset with Benchmark Tasks for Text Generation of Pretrained Language Models

NeurIPS 2022accept

In order to diagnostically analyze and improve the capability of pretrained language models (PLMs) in text generation, we propose TGEA 2.0, to date the largest dataset built on machine-authored texts by PLMs with fine-grained semantic annotations on a wide variety of pathological generation errors.…

2021

Chinese WPLC: A Chinese Dataset for Evaluating Pretrained Language Models on Word Prediction Given Long-Range Context

EMNLP 2021main

This paper presents a Chinese dataset for evaluating pretrained language models on Word Prediction given Long-term Context (Chinese WPLC). We propose both automatic and manual selection strategies tailored to Chinese to guarantee that target words in passages collected from over 69K novels can only…