While pretrained models such as BERT have shown large gains across natural language understanding tasks, their performance can be improved by further training the model on a data-rich intermediate task, before fine-tuning it on a target task. However, it is still poorly understood when and why intermediate-task training is beneficial for a given target task. To investigate this, we perform a large-scale study on the pretrained RoBERTa model with 110 intermediate-target task combinations. We further evaluate all trained models with 25 probing tasks meant to reveal the specific skills that drive transfer. We observe that intermediate tasks requiring high-level inference and reasoning abilities tend to work best. We also observe that target task performance is strongly correlated with higher-level abilities such as coreference resolution. However, we fail to observe more granular correlations between probing and target task performance, highlighting the need for further work on broad-coverage probing benchmarks. We also observe evidence that the forgetting of knowledge learned during pretraining may limit our analysis, highlighting the need for further work on transfer learning methods in these settings.
University of Mannheim（曼海姆大学）；University of Cambridge（剑桥大学）
Traditional NLP has long held (supervised) syntactic parsing necessary for successful higher-level semantic language understanding (LU). The recent advent of end-to-end neural models, self-supervised via language modeling (LM), and their success on a wide range of LU tasks, however, questions this belief. In this work, we empirically investigate the usefulness of supervised parsing for semantic LU in the context of LM-pretrained transformer networks. Relying on the established fine-tuning paradigm, we first couple a pretrained transformer with a biaffine parsing head, aiming to infuse explicit…
School of Data and Computer Science（数据与计算机学院）；Sun Yat-sen University（中山大学）；Guangzhou, China（中国广州（注：此为地址信息，非机构，根据规则应忽略，此处原文本可能存在排版问题，正确机构应为前两项））
Aspect based sentiment analysis aims to identify the sentimental tendency towards a given aspect in text. Fine-tuning of pretrained BERT performs excellent on this task and achieves state-of-the-art performances. Existing BERT-based works only utilize the last output layer of BERT and ignore the semantic knowledge in the intermediate layers. This paper explores the potential of utilizing BERT intermediate layers to enhance the performance of fine-tuning of BERT. To the best of our knowledge, no existing work has been done on this research. To show the generality, we also apply this approach to…
Academia Sinica（中央研究院）；Institute of Information Science, Academia Sinica（中央研究院资讯科学研究所）
Supplementary Training on Intermediate Labeled-data Tasks (STILTs) is a widely applied technique, which first fine-tunes the pretrained language models on an intermediate task before on the target task of interest. While STILTs is able to further improve the performance of pretrained language models, it is still unclear why and when it works. Previous research shows that those intermediate tasks involving complex inference, such as commonsense reasoning, work especially well for RoBERTa. In this paper, we discover that the improvement from an intermediate task could be orthogonal to it contain…
Korea Advanced Institute of Science and Technology（韩国科学技术院）；Ulsan National Institute of Science and Technology（蔚山国立科学技术院）
Currently, contextualized word representations are learned by intricate neural network models, such as masked neural language models (MNLMs). The new representations significantly enhanced the performance in automated question answering by reading paragraphs. However, identifying the detailed knowledge trained in the MNLMs is difficult owing to numerous and intermingled parameters. This paper provides empirical but insightful analyses on the pretrained MNLMs with respect to common sense knowledge. First, we propose a test that measures what types of common sense knowledge do pretrained MNLMs u…
There is growing evidence that pretrained language models improve task-specific fine-tuning not just for the languages seen in pretraining, but also for new languages and even non-linguistic data. What is the nature of this surprising cross-domain transfer? We offer a partial answer via a systematic exploration of how much transfer occurs when models are denied any information about word identity via random scrambling. In four classification tasks and two sequence labeling tasks, we evaluate baseline models, LSTMs using GloVe embeddings, and BERT. We find that only BERT shows high rates of tra…
Indian Institute of Technology, Bombay（印度理工学院孟买分校）
While recent benchmarks have spurred a lot of new work on improving the generalization of pretrained multilingual language models on multilingual tasks, techniques to improve code-switched natural language understanding tasks have been far less explored. In this work, we propose the use of bilingual intermediate pretraining as a reliable technique to derive large and consistent performance gains on three different NLP tasks using code-switched text. We achieve substantial absolute improvements of 7.87%, 20.15%, and 10.99%, on the mean accuracies and F1 scores over previous state-of-the-art sys…
Princeton University（普林斯顿大学）；Institute for Advanced Study（高等研究院）
Autoregressive language models, pretrained using large text corpora to do well on next word prediction, have been successful at solving many downstream tasks, even with zero-shot usage. However, there is little theoretical understanding of this success. This paper initiates a mathematical study of this phenomenon for the downstream task of text classification by considering the following questions: (1) What is the intuitive connection between the pretraining task of next word prediction and text classification? (2) How can we mathematically formalize this connection and quantify the benefit of…
With the growth of computing power neural machine translation (NMT) models also grow accordingly and become better. However, they also become harder to deploy on edge devices due to memory constraints. To cope with this problem, a common practice is to distill knowledge from a large and accurately-trained teacher network (T) into a compact student network (S). Although knowledge distillation (KD) is useful in most cases, our study shows that existing KD techniques might not be suitable enough for deep NMT engines, so we propose a novel alternative. In our model, besides matching T and S predic…
National Technical University of Athens（雅典国立技术大学）；Athena Research Center（雅典娜研究中心）
Multimodal learning pipelines have benefited from the success of pretrained language models. However, this comes at the cost of increased model parameters. In this work, we propose Adapted Multimodal BERT (AMB), a BERT-based architecture for multimodal tasks that uses a combination of adapter modules and intermediate fusion layers. The adapter adjusts the pretrained language model for the task at hand, while the fusion layers perform task-specific, layer-wise fusion of audio-visual information with textual BERT representations. During the adaptation process the pre-trained language model param…
We study the problem of multilingual masked language modeling, i.e. the training of a single model on concatenated text from multiple languages, and present a detailed study of several factors that influence why these models are so effective for cross-lingual transfer. We show, contrary to what was previously hypothesized, that transfer is possible even when there is no shared vocabulary across the monolingual corpora and also when the text comes from very different domains. The only requirement is that there are some shared parameters in the top layers of the multi-lingual encoder. To better …
Yale University（耶鲁大学）；Facebook AI（脸书人工智能研究院）；Nanyang Technological University（南洋理工大学）；Renmin University of China（中国人民大学）
Models pretrained with self-supervised objectives on large text corpora achieve state-of-the-art performance on English text summarization tasks. However, these models are typically fine-tuned on hundreds of thousands of data points, an infeasible requirement when applying summarization to new, niche domains. In this work, we introduce a novel and generalizable method, called WikiTransfer, for fine-tuning pretrained models for summarization in an unsupervised, dataset-specific manner. WikiTransfer fine-tunes pretrained models on pseudo-summaries, produced from generic Wikipedia data, which con…
Pretrained language models have achieved a new state of the art on many NLP tasks, but there are still many open questions about how and why they work so well. We investigate the contextualization of words in BERT. We quantify the amount of contextualization, i.e., how well words are interpreted in context, by studying the extent to which semantic classes of a word can be inferred from its contextualized embeddings. Quantifying contextualization helps in understanding and utilizing pretrained language models. We show that top layer representations achieve high accuracy inferring semantic class…
New York University（纽约大学）；ILLC, University of Amsterdam（阿姆斯特丹大学逻辑、语言与计算研究所）；University of Colorado Boulder（科罗拉多大学博尔德分校）
Intermediate-task training---fine-tuning a pretrained model on an intermediate task before fine-tuning again on the target task---often improves model performance substantially on language understanding tasks in monolingual English settings. We investigate whether English intermediate-task training is still helpful on non-English target tasks. Using nine intermediate language-understanding tasks, we evaluate intermediate-task transfer in a zero-shot cross-lingual setting on the XTREME benchmark. We see large improvements from intermediate training on the BUCC and Tatoeba sentence retrieval tas…
Korea Advanced Institute of Science and Technology（韩国科学技术院）
Many contextualized word representations are now learned by intricate neural network models, such as masked neural language models (MNLMs) which are made up of huge neural network structures and trained to restore the masked text. Such representations demonstrate superhuman performance in some reading comprehension (RC) tasks which extract a proper answer in the context given a question. However, identifying the detailed knowledge trained in MNLMs is challenging owing to numerous and intermingled model parameters. This paper provides new insights and empirical analyses on commonsense knowledge…
Beijing Institute of Technology（北京理工大学）；Meituan（美团）
Prompt-tuning has shown appealing performance in few-shot classification by virtue of its capability in effectively exploiting pre-trained knowledge. This motivates us to check the hypothesis that prompt-tuning is also a promising choice for long-tailed classification, since the tail classes are intuitively few-shot ones. To achieve this aim, we conduct empirical studies to examine the hypothesis. The results demonstrate that prompt-tuning makes pretrained language models at least good long-tailed learners. For intuitions on why prompt-tuning can achieve good performance in long-tailed classif…
Geographic Adaptation of Pretrained Language Models
CAIDAS, University of Würzburg（维尔茨堡大学CAIDAS）；Jožef Stefan Institute（约热夫·斯特凡研究所）；University of Ljubljana（卢布尔雅那大学）；University of Oxford（牛津大学）；LMU Munich（慕尼黑大学）；Allen Institute for AI（艾伦人工智能研究所）
A Systematic Survey of Semantic Role Labeling in the Era of Pretrained Language Models
Harbin Institute of Technology (Shenzhen)（哈尔滨工业大学（深圳））；Shenzhen Loop Area Institute (SLAI)（深圳南山区研究院）；University of Oslo（奥斯陆大学）；Charles University（查尔斯大学）
Adaptation Odyssey in LLMs: Why Does Additional Pretraining Sometimes Fail to Improve?
Concordia University（康考迪亚大学）；Mila-Quebec AI Institute（米拉-魁北克人工智能研究所）；University of Tübingen（蒂宾根大学）；Tübingen AI Center（蒂宾根人工智能中心）；Cohere For AI（Cohere人工智能项目）；Laval University（拉瓦尔大学）
Previous studies observed that finetuned models may be better base models than the vanilla pretrained model. Such a model, finetuned on some source dataset, may provide a better starting point for a new finetuning process on a desired target dataset. Here, we perform a systematic analysis of this intertraining scheme, over a wide range of English classification tasks. Surprisingly, our analysis suggests that the potential intertraining gain can be analyzed independently for the target dataset under consideration, and for a base model being considered as a starting point. This is in contrast to…
