id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
cord-025523-6ttps1nx	Barlas, Georgios	Cross-Domain Authorship Attribution Using Pre-trained Language Models	2020-05-06		.txt	text/plain	3611	193	56	title: Cross-Domain Authorship Attribution Using Pre-trained Language Models An especially challenging but very realistic scenario is cross-domain attribution where texts of known authorship (training set) differ from texts of disputed authorship (test set) in topic or genre. Recently, the use of pre-trained language models (e.g., BERT, ELMo, ULM-FiT, has been demonstrated to obtain significant gains in several text classification tasks including sentiment analysis, emotion classification, and topic classification [2, 7, 13, 14] . This method is based on a character-level recurrent (RNN) neural network language model and a multiheaded classifier (MHC) [1] . We examine the use of pre-trained language models (e.g., BERT, ELMo, ULMFiT, GPT-2) in AA and the potentials of MHC. Based on Bagnall's model [1] , originally proposed for authorship verification, we compare the performance when we use either the original characterlevel RNN trained from scratch in the small-size AA corpus or pre-trained tokenbased language models obtained from general-domain corpora.	./cache/cord-025523-6ttps1nx.txt	./txt/cord-025523-6ttps1nx.txt