اشترك بالحزمة الذهبية واحصل على وصول غير محدود شمرا أكاديميا
تسجيل مستخدم جديدAutomatic Detection of Generated Text is Easiest when Humans are Fooled
88
0
0.0
(
0
)
اسأل ChatGPT حول البحث
ﻻ يوجد ملخص باللغة العربية
Recent advancements in neural language modelling make it possible to rapidly generate vast amounts of human-sounding text. The capabilities of humans and automatic discriminators to detect machine-generated text have been a large source of research interest, but humans and machines rely on different cues to make their decisions. Here, we perform careful benchmarking and analysis of three popular sampling-based decoding strategies---top-$k$, nucleus sampling, and untruncated random sampling---and show that improvements in decoding methods have primarily optimized for fooling humans. This comes at the expense of introducing statistical abnormalities that make detection easy for automatic systems. We also show that though both human and automatic detector performance improve with longer excerpt length, even multi-sentence excerpts can fool expert human raters over 30% of the time. Our findings reveal the importance of using both human and automatic detectors to assess the humanness of text generation systems.
قيم البحث
اقرأ أيضاً
In recent years, large neural networks for natural language generation (NLG) have made leaps and bounds in their ability to generate fluent text. However, the tasks of evaluating quality differences between NLG systems and understanding how humans pe
rceive the generated text remain both crucial and difficult. In this system demonstration, we present Real or Fake Text (RoFT), a website that tackles both of these challenges by inviting users to try their hand at detecting machine-generated text in a variety of domains. We introduce a novel evaluation task based on detecting the boundary at which a text passage that starts off human-written transitions to being machine-generated. We show preliminary results of using RoFT to evaluate detection of machine-generated news articles.
A wide variety of NLP applications, such as machine translation, summarization, and dialog, involve text generation. One major challenge for these applications is how to evaluate whether such generated texts are actually fluent, accurate, or effectiv
e. In this work, we conceptualize the evaluation of generated text as a text generation problem, modeled using pre-trained sequence-to-sequence models. The general idea is that models trained to convert the generated text to/from a reference output or the source text will achieve higher scores when the generated text is better. We operationalize this idea using BART, an encoder-decoder based pre-trained model, and propose a metric BARTScore with a number of variants that can be flexibly applied in an unsupervised fashion to evaluation of text from different perspectives (e.g. informativeness, fluency, or factuality). BARTScore is conceptually simple and empirically effective. It can outperform existing top-scoring metrics in 16 of 22 test settings, covering evaluation of 16 datasets (e.g., machine translation, text summarization) and 7 different perspectives (e.g., informativeness, factuality). Code to calculate BARTScore is available at https://github.com/neulab/BARTScore, and we have released an interactive leaderboard for meta-evaluation at http://explainaboard.nlpedia.ai/leaderboard/task-meval/ on the ExplainaBoard platform, which allows us to interactively understand the strengths, weaknesses, and complementarity of each metric.
Argument mining is often addressed by a pipeline method where segmentation of text into argumentative units is conducted first and proceeded by an argument component identification task. In this research, we apply a token-level classification to iden
tify claim and premise tokens from a new corpus of argumentative essays written by middle school students. To this end, we compare a variety of state-of-the-art models such as discrete features and deep learning architectures (e.g., BiLSTM networks and BERT-based architectures) to identify the argument components. We demonstrate that a BERT-based multi-task learning architecture (i.e., token and sentence level classification) adaptively pretrained on a relevant unlabeled dataset obtains the best results
Manual Summarization of large bodies of text involves a lot of human effort and time, especially in the legal domain. Lawyers spend a lot of time preparing legal briefs of their clients case files. Automatic Text summarization is a constantly evolvin
g field of Natural Language Processing(NLP), which is a subdiscipline of the Artificial Intelligence Field. In this paper a hybrid method for automatic text summarization of legal cases using k-means clustering technique and tf-idf(term frequency-inverse document frequency) word vectorizer is proposed. The summary generated by the proposed method is compared using ROGUE evaluation parameters with the case summary as prepared by the lawyer for appeal in court. Further, suggestions for improving the proposed method are also presented.
A snowclone is a customizable phrasal template that can be realized in multiple, instantly recognized variants. For example, ``* is the new * (Orange is the new black, 40 is the new 30). Snowclones are extensively used in social media. In this paper,
we study snowclones originating from pop-culture quotes; our goal is to automatically detect cultural references in text. We introduce a new, publicly available data set of pop-culture quotes and their corresponding snowclone usages and train models on them. We publish code for Catchphrase, an internet browser plugin to automatically detect and mark references in real-time, and examine its performance via a user study. Aside from assisting people to better comprehend cultural references, we hope that detecting snowclones can complement work on paraphrasing and help to tackle long-standing questions in social science about the dynamics of information propagation.
سجل دخول لتتمكن من نشر تعليقات
التعليقات
جاري جلب التعليقات
سجل دخول لتتمكن من متابعة معايير البحث التي قمت باختيارها
