Recent work has demonstrated the effectiveness of cross-lingual language model pretraining for cross-lingual understanding. In this study, we present the results of two larger multilingual masked language models, with 3.5B and 10.7B parameters. Our two new models dubbed XLM-R XL and XLM-R XXL outperform XLM-R by 1.8% and 2.4% average accuracy on XNLI. Our model also outperforms the RoBERTa-Large model on several English tasks of the GLUE benchmark by 0.3% on average while handling 99 more languages. This suggests pretrained models with larger capacity may obtain both strong performance on high-resource languages while greatly improving low-resource languages. We make our code and models publicly available.
Few-shot algorithms aim at learning new tasks provided only a handful of training examples. In this work we investigate few-shot learning in the setting where the data points are sequences of tokens and propose an efficient learning algorithm based on Transformers. In the simplest setting, we append a token to an input sequence which represents the particular task to be undertaken, and show that the embedding of this token can be optimized on the fly given few labeled examples. Our approach does not require complicated changes to the model architecture such as adapter layers nor computing second order derivatives as is currently popular in the meta-learning and few-shot learning literature. We demonstrate our approach on a variety of tasks, and analyze the generalization properties of several model variants and baseline approaches. In particular, we show that compositional task descriptors can improve performance. Experiments show that our approach works at least as well as other methods, while being more computationally efficient.
Building open-domain chatbots is a challenging area for machine learning research. While prior work has shown that scaling neural models in the number of parameters and the size of the data they are trained on gives improved results, we show that other ingredients are important for a high-performing chatbot. Good conversation requires a number of skills that an expert conversationalist blends in a seamless way: providing engaging talking points and listening to their partners, and displaying knowledge, empathy and personality appropriately, while maintaining a consistent persona. We show that large scale models can learn these skills when given appropriate training data and choice of generation strategy. We build variants of these recipes with 90M, 2.7B and 9.4B parameter models, and make our models and code publicly available. Human evaluations show our best models are superior to existing approaches in multi-turn dialogue in terms of engagingness and humanness measurements. We then discuss the limitations of this work by analyzing failure cases of our models.
The state of the art on many NLP tasks is currently achieved by large pre-trained language models, which require a considerable amount of computation. We explore a setting where many different predictions are made on a single piece of text. In that case, some of the computational cost during inference can be amortized over the different tasks using a shared text encoder. We compare approaches for training such an encoder and show that encoders pre-trained over multiple tasks generalize well to unseen tasks. We also compare ways of extracting fixed- and limited-size representations from this encoder, including different ways of pooling features extracted from multiple layers or positions. Our best approach compares favorably to knowledge distillation, achieving higher accuracy and lower computational cost once the system is handling around 7 tasks. Further, we show that through binary quantization, we can reduce the size of the extracted representations by a factor of 16 making it feasible to store them for later use. The resulting method offers a compelling solution for using large-scale pre-trained models at a fraction of the computational cost when multiple tasks are performed on the same text.
Text generation is ubiquitous in many NLP tasks, from summarization, to dialogue and machine translation. The dominant parametric approach is based on locally normalized models which predict one word at a time. While these work remarkably well, they are plagued by exposure bias due to the greedy nature of the generation process. In this work, we investigate un-normalized energy-based models (EBMs) which operate not at the token but at the sequence level. In order to make training tractable, we first work in the residual of a pretrained locally normalized language model and second we train using noise contrastive estimation. Furthermore, since the EBM works at the sequence level, we can leverage pretrained bi-directional contextual representations, such as BERT and RoBERTa. Our experiments on two large language modeling datasets show that residual EBMs yield lower perplexity compared to locally normalized baselines. Moreover, generation via importance sampling is very efficient and of higher quality than the baseline models according to human evaluation.
Current large-scale auto-regressive language models display impressive fluency and can generate convincing text. In this work we start by asking the question: Can the generations of these models be reliably distinguished from real text by statistical discriminators? We find experimentally that the answer is affirmative when we have access to the training data for the model, and guardedly affirmative even if we do not. This suggests that the auto-regressive models can be improved by incorporating the (globally normalized) discriminators into the generative process. We give a formalism for this using the Energy-Based Model framework, and show that it indeed improves the results of the generative models, measured both in terms of perplexity and in terms of human evaluation.
Language models are of considerable importance. They are used for pretraining, finetuning, and rescoring in downstream applications, and as is as a test-bed and benchmark for progress in natural language understanding. One fundamental question regards the way we should generate text from a language model. It is well known that different decoding strategies can have dramatic impact on the quality of the generated text and using the most likely sequence under the model distribution, e.g., via beam search, generally leads to degenerate and repetitive outputs. While generation strategies such as top-k and nucleus sampling lead to more natural and less repetitive generations, the true cost of avoiding the highest scoring solution is hard to quantify. In this paper, we argue that verifiability, i.e., the consistency of the generated text with factual knowledge, is a suitable metric for measuring this cost. We use an automatic fact-checking system to calculate new metrics as a function of the number of supported claims per sentence and find that sampling-based generation strategies, such as top-k, indeed lead to less verifiable text. This finding holds across various dimensions, such as model size, training data size and parameters of the generation strategy. Based on this finding, we introduce a simple and effective generation strategy for producing non-repetitive and more verifiable (in comparison to other methods) text.
This paper shows that pretraining multilingual language models at scale leads to significant performance gains for a wide range of cross-lingual transfer tasks. We train a Transformer-based masked language model on one hundred languages, using more than two terabytes of filtered CommonCrawl data. Our model, dubbed XLM-R, significantly outperforms multilingual BERT (mBERT) on a variety of cross-lingual benchmarks, including +13.8% average accuracy on XNLI, +12.3% average F1 score on MLQA, and +2.1% average F1 score on NER. XLM-R performs particularly well on low-resource languages, improving 11.8% in XNLI accuracy for Swahili and 9.2% for Urdu over the previous XLM model. We also present a detailed empirical evaluation of the key factors that are required to achieve these gains, including the trade-offs between (1) positive transfer and capacity dilution and (2) the performance of high and low resource languages at scale. Finally, we show, for the first time, the possibility of multilingual modeling without sacrificing per-language performance; XLM-Ris very competitive with strong monolingual models on the GLUE and XNLI benchmarks. We will make XLM-R code, data, and models publicly available.
In this work, we study how the large-scale pretrain-finetune framework changes the behavior of a neural language generator. We focus on the transformer encoder-decoder model for the open-domain dialogue response generation task. We find that after standard fine-tuning, the model forgets important language generation skills acquired during large-scale pre-training. We demonstrate the forgetting phenomenon through a detailed behavior analysis from the perspectives of context sensitivity and knowledge transfer. Adopting the concept of data mixing, we propose an intuitive fine-tuning strategy named "mix-review". We find that mix-review effectively regularize the fine-tuning process, and the forgetting problem is largely alleviated. Finally, we discuss interesting behavior of the resulting dialogue model and its implications.