Weight sharing is a popular approach to reduce the training cost of neural architecture search (NAS) by reusing the weights of shared operators from previously trained child models. However, the rank correlation between the estimated accuracy and ground truth accuracy of those child models is low due to the interference among different child models caused by weight sharing. In this paper, we investigate the interference issue by sampling different child models and calculating the gradient similarity of shared operators, and observe that: 1) the interference on a shared operator between two child models is positively correlated with the number of different operators between them; 2) the interference is smaller when the inputs and outputs of the shared operator are more similar. Inspired by these two observations, we propose two approaches to mitigate the interference: 1) rather than randomly sampling child models for optimization, we propose a gradual modification scheme by modifying one operator between adjacent optimization steps to minimize the interference on the shared operators; 2) forcing the inputs and outputs of the operator across all child models to be similar to reduce the interference. Experiments on a BERT search space verify that mitigating interference via each of our proposed methods improves the rank correlation of super-pet and combining both methods can achieve better results. Our discovered architecture outperforms RoBERTa$_{\rm base}$ by 1.1 and 0.6 points and ELECTRA$_{\rm base}$ by 1.6 and 1.1 points on the dev and test set of GLUE benchmark. Extensive results on the BERT compression, reading comprehension and large-scale image classification tasks also demonstrate the effectiveness and generality of our proposed methods.

Pre-trained language models have attracted increasing attention in the biomedical domain, inspired by their great success in the general natural language domain. Among the two main branches of pre-trained language models in the general language domain, i.e. BERT (and its variants) and GPT (and its variants), the first one has been extensively studied in the biomedical domain, such as BioBERT and PubMedBERT. While they have achieved great success on a variety of discriminative downstream biomedical tasks, the lack of generation ability constrains their application scope. In this paper, we propose BioGPT, a domain-specific generative Transformer language model pre-trained on large-scale biomedical literature. We evaluate BioGPT on six biomedical natural language processing tasks and demonstrate that our model outperforms previous models on most tasks. Especially, we get 44.98%, 38.42% and 40.76% F1 score on BC5CDR, KD-DTI and DDI end-to-end relation extraction tasks, respectively, and 78.2% accuracy on PubMedQA, creating a new record. Our case study on text generation further demonstrates the advantage of BioGPT on biomedical literature to generate fluent descriptions for biomedical terms.

Text to speech (TTS) has been broadly used to synthesize natural and intelligible speech in different scenarios. Deploying TTS in various end devices such as mobile phones or embedded devices requires extremely small memory usage and inference latency. While non-autoregressive TTS models such as FastSpeech have achieved significantly faster inference speed than autoregressive models, their model size and inference latency are still large for the deployment in resource constrained devices. In this paper, we propose LightSpeech, which leverages neural architecture search (NAS) to automatically design more lightweight and efficient models based on FastSpeech. We first profile the components of current Fast-Speech model and carefully design a novel search space containing various lightweight and potentially effective architectures. Then NAS is utilized to automatically discover well performing architectures within the search space. Experiments show that the model discovered by our method achieves 15x model compression ratio and 6.5x inference speedup on CPU with on par voice quality. Audio demos are provided at https://speechresearch.github.io/lightspeech.

Neural approaches have achieved state-of-the-art accuracy on machine translation but suffer from the high cost of collecting large scale parallel data. Thus, a lot of research has been conducted for neural machine translation (NMT) with very limited parallel data, i.e., the low-resource setting. In this paper, we provide a survey for low-resource NMT and classify related works into three categories according to the auxiliary data they used: (1) exploiting monolingual data of source and/or target languages, (2) exploiting data from auxiliary languages, and (3) exploiting multi-modal data. We hope that our survey can help researchers to better understand this field and inspire them to design better algorithms, and help industry practitioners to choose appropriate algorithms for their applications.

While pre-trained language models (e.g., BERT) have achieved impressive results on different natural language processing tasks, they have large numbers of parameters and suffer from big computational and memory costs, which make them difficult for real-world deployment. Therefore, model compression is necessary to reduce the computation and memory cost of pre-trained models. In this work, we aim to compress BERT and address the following two challenging practical issues: (1) The compression algorithm should be able to output multiple compressed models with different sizes and latencies, in order to support devices with different memory and latency limitations; (2) The algorithm should be downstream task agnostic, so that the compressed models are generally applicable for different downstream tasks. We leverage techniques in neural architecture search (NAS) and propose NAS-BERT, an efficient method for BERT compression. NAS-BERT trains a big supernet on a carefully designed search space containing a variety of architectures and outputs multiple compressed models with adaptive sizes and latency. Furthermore, the training of NAS-BERT is conducted on standard self-supervised pre-training tasks (e.g., masked language model) and does not depend on specific downstream tasks. Thus, the compressed models can be used across various downstream tasks. The technical challenge of NAS-BERT is that training a big supernet on the pre-training task is extremely costly. We employ several techniques including block-wise search, search space pruning, and performance approximation to improve search efficiency and accuracy. Extensive experiments on GLUE and SQuAD benchmark datasets demonstrate that NAS-BERT can find lightweight models with better accuracy than previous approaches, and can be directly applied to different downstream tasks with adaptive model sizes for different requirements of memory or latency.

Error correction techniques have been used to refine the output sentences from automatic speech recognition (ASR) models and achieve a lower word error rate (WER) than original ASR outputs. Previous works usually use a sequence-to-sequence model to correct an ASR output sentence autoregressively, which causes large latency and cannot be deployed in online ASR services. A straightforward solution to reduce latency, inspired by non-autoregressive (NAR) neural machine translation, is to use an NAR sequence generation model for ASR error correction, which, however, comes at the cost of significantly increased ASR error rate. In this paper, observing distinctive error patterns and correction operations (i.e., insertion, deletion, and substitution) in ASR, we propose FastCorrect, a novel NAR error correction model based on edit alignment. In training, FastCorrect aligns each source token from an ASR output sentence to the target tokens from the corresponding ground-truth sentence based on the edit distance between the source and target sentences, and extracts the number of target tokens corresponding to each source token during edition/correction, which is then used to train a length predictor and to adjust the source tokens to match the length of the target sentence for parallel generation. In inference, the token number predicted by the length predictor is used to adjust the source tokens for target sequence generation. Experiments on the public AISHELL-1 dataset and an internal industrial-scale ASR dataset show the effectiveness of FastCorrect for ASR error correction: 1) it speeds up the inference by 6-9 times and maintains the accuracy (8-14% WER reduction) compared with the autoregressive correction model; and 2) it outperforms the popular NAR models adopted in neural machine translation and text edition by a large margin.

Neural architecture search (NAS) relies on a good controller to generate better architectures or predict the accuracy of given architectures. However, training the controller requires both abundant and high-quality pairs of architectures and their accuracy, while it is costly to evaluate an architecture and obtain its accuracy. In this paper, we propose SemiNAS, a semi-supervised NAS approach that leverages numerous unlabeled architectures (without evaluation and thus nearly no cost). Specifically, SemiNAS 1) trains an initial accuracy predictor with a small set of architecture-accuracy data pairs; 2) uses the trained accuracy predictor to predict the accuracy of large amount of architectures (without evaluation); and 3) adds the generated data pairs to the original data to further improve the predictor. The trained accuracy predictor can be applied to various NAS algorithms by predicting the accuracy of candidate architectures for them. SemiNAS has two advantages: 1) It reduces the computational cost under the same accuracy guarantee. On NASBench-101 benchmark dataset, it achieves comparable accuracy with gradient-based method while using only 1/7 architecture-accuracy pairs. 2) It achieves higher accuracy under the same computational cost. It achieves 94.02% test accuracy on NASBench-101, outperforming all the baselines when using the same number of architectures. On ImageNet, it achieves 23.5% top-1 error rate (under 600M FLOPS constraint) using 4 GPU-days for search. We further apply it to LJSpeech text to speech task and it achieves 97% intelligibility rate in the low-resource setting and 15% test error rate in the robustness setting, with 9%, 7% improvements over the baseline respectively.

Automatic neural architecture design has shown its potential in discovering powerful neural network architectures. Existing methods, no matter based on reinforcement learning or evolutionary algorithms (EA), conduct architecture search in a discrete space, which is highly inefficient. In this paper, we propose a simple and efficient method to automatic neural architecture design based on continuous optimization. We call this new approach neural architecture optimization (NAO). There are three key components in our proposed approach: (1) An encoder embeds/maps neural network architectures into a continuous space. (2) A predictor takes the continuous representation of a network as input and predicts its accuracy. (3) A decoder maps a continuous representation of a network back to its architecture. The performance predictor and the encoder enable us to perform gradient based optimization in the continuous space to find the embedding of a new architecture with potentially better accuracy. Such a better embedding is then decoded to a network by the decoder. Experiments show that the architecture discovered by our method is very competitive for image classification task on CIFAR-10 and language modeling task on PTB, outperforming or on par with the best results of previous architecture search methods with a significantly reduction of computational resources. Specifically we obtain $2.11\%$ test set error rate for CIFAR-10 image classification task and $56.0$ test set perplexity of PTB language modeling task. The best discovered architectures on both tasks are successfully transferred to other tasks such as CIFAR-100 and WikiText-2. Furthermore, combined with the recent proposed weight sharing mechanism, we discover powerful architecture on CIFAR-10 (with error rate $3.53\%$) and on PTB (with test set perplexity $56.6$), with very limited computational resources (less than $10$ GPU hours) for both tasks.

Highly flexible software network functions (NFs) are crucial components to enable multi-tenancy in the clouds. However, software packet processing on a commodity server has limited capacity and induces high latency. While software NFs could scale out using more servers, doing so adds significant cost. This paper focuses on accelerating NFs with programmable hardware, i.e., FPGA, which is now a mature technology and inexpensive for datacenters. However, FPGA is predominately programmed using low-level hardware description languages (HDLs), which are hard to code and difficult to debug. More importantly, HDLs are almost inaccessible for most software programmers. This paper presents ClickNP, a FPGA-accelerated platform for highly flexible and high-performance NFs with commodity servers. ClickNP is highly flexible as it is completely programmable using high-level C-like languages, and exposes a modular programming abstraction that resembles Click Modular Router. ClickNP is also high performance. Our prototype NFs show that they can process traffic at up to 200 million packets per second with ultra-low latency ($< 2\mu$s). Compared to existing software counterparts, with FPGA, ClickNP improves throughput by 10x, while reducing latency by 10x. To the best of our knowledge, ClickNP is the first FPGA-accelerated platform for NFs, written completely in high-level language and achieving 40 Gbps line rate at any packet size.