Di He

Zhenyu He, Zexuan Zhong, Tianle Cai et al.

We introduce Retrieval-Based Speculative Decoding (REST), a novel algorithm designed to speed up language model generation. The key insight driving the development of REST is the observation that the process of text generation often includes certain common phases and patterns. Unlike previous methods that rely on a draft language model for speculative decoding, REST harnesses the power of retrieval to generate draft tokens. This method draws from the reservoir of existing knowledge, retrieving and employing relevant tokens based on the current context. Its plug-and-play nature allows for seamless integration and acceleration of any language models, all without necessitating additional training. When benchmarked on 7B and 13B language models in a single-batch setting, REST achieves a significant speedup of 1.62X to 2.36X on code or text generation. The code of REST is available at https://github.com/FasterDecoding/REST.

Bohang Zhang, Du Jiang, Di He et al.

Recently, Zhang et al. (2021) developed a new neural network architecture based on $\ell_\infty$-distance functions, which naturally possesses certified $\ell_\infty$ robustness by its construction. Despite the novel design and theoretical foundation, so far the model only achieved comparable performance to conventional networks. In this paper, we make the following two contributions: $\mathrm{(i)}$ We demonstrate that $\ell_\infty$-distance nets enjoy a fundamental advantage in certified robustness over conventional networks (under typical certification approaches); $\mathrm{(ii)}$ With an improved training process we are able to significantly boost the certified accuracy of $\ell_\infty$-distance nets. Our training approach largely alleviates the optimization problem that arose in the previous training scheme, in particular, the unexpected large Lipschitz constant due to the use of a crucial trick called $\ell_p$-relaxation. The core of our training approach is a novel objective function that combines scaled cross-entropy loss and clipped hinge loss with a decaying mixing coefficient. Experiments show that using the proposed training strategy, the certified accuracy of $\ell_\infty$-distance net can be dramatically improved from 33.30% to 40.06% on CIFAR-10 ($ε=8/255$), meanwhile outperforming other approaches in this area by a large margin. Our results clearly demonstrate the effectiveness and potential of $\ell_\infty$-distance net for certified robustness. Codes are available at https://github.com/zbh2047/L_inf-dist-net-v2.

Tianyu Pang, Huishuai Zhang, Di He et al.

Correctly classifying adversarial examples is an essential but challenging requirement for safely deploying machine learning models. As reported in RobustBench, even the state-of-the-art adversarially trained models struggle to exceed 67% robust test accuracy on CIFAR-10, which is far from practical. A complementary way towards robustness is to introduce a rejection option, allowing the model to not return predictions on uncertain inputs, where confidence is a commonly used certainty proxy. Along with this routine, we find that confidence and a rectified confidence (R-Con) can form two coupled rejection metrics, which could provably distinguish wrongly classified inputs from correctly classified ones. This intriguing property sheds light on using coupling strategies to better detect and reject adversarial examples. We evaluate our rectified rejection (RR) module on CIFAR-10, CIFAR-10-C, and CIFAR-100 under several attacks including adaptive ones, and demonstrate that the RR module is compatible with different adversarial training frameworks on improving robustness, with little extra computation. The code is available at https://github.com/P2333/Rectified-Rejection.

Shuqi Lu, Di He, Chenyan Xiong et al.

Dense retrieval requires high-quality text sequence embeddings to support effective search in the representation space. Autoencoder-based language models are appealing in dense retrieval as they train the encoder to output high-quality embedding that can reconstruct the input texts. However, in this paper, we provide theoretical analyses and show empirically that an autoencoder language model with a low reconstruction loss may not provide good sequence representations because the decoder may take shortcuts by exploiting language patterns. To address this, we propose a new self-learning method that pre-trains the autoencoder using a \textit{weak} decoder, with restricted capacity and attention flexibility to push the encoder to provide better text representations. Our experiments on web search, news recommendation, and open domain question answering show that our pre-trained model significantly boosts the effectiveness and few-shot ability of dense retrieval models. Our code is available at https://github.com/microsoft/SEED-Encoder/.

Shengjie Luo, Kaiyuan Gao, Shuxin Zheng et al.

Transformer has demonstrated its great power to learn contextual word representations for multiple languages in a single model. To process multilingual sentences in the model, a learnable vector is usually assigned to each language, which is called "language embedding". The language embedding can be either added to the word embedding or attached at the beginning of the sentence. It serves as a language-specific signal for the Transformer to capture contextual representations across languages. In this paper, we revisit the use of language embedding and identify several problems in the existing formulations. By investigating the interaction between language embedding and word embedding in the self-attention module, we find that the current methods cannot reflect the language-specific word correlation well. Given these findings, we propose a new approach called Cross-lingual Language Projection (XLP) to replace language embedding. For a sentence, XLP projects the word embeddings into language-specific semantic space, and then the projected embeddings will be fed into the Transformer model to process with their language-specific meanings. In such a way, XLP achieves the purpose of appropriately encoding "language" in a multilingual Transformer model. Experimental results show that XLP can freely and significantly boost the model performance on extensive multilingual benchmark datasets. Codes and models will be released at https://github.com/lsj2408/XLP.

Di He, Lei Xie

Accurate and robust prediction of patient's response to drug treatments is critical for developing precision medicine. However, it is often difficult to obtain a sufficient amount of coherent drug response data from patients directly for training a generalized machine learning model. Although the utilization of rich cell line data provides an alternative solution, it is challenging to transfer the knowledge obtained from cell lines to patients due to various confounding factors. Few existing transfer learning methods can reliably disentangle common intrinsic biological signals from confounding factors in the cell line and patient data. In this paper, we develop a Coherent Deconfounding Autoencoder (CODE-AE) that can extract both common biological signals shared by incoherent samples and private representations unique to each data set, transfer knowledge learned from cell line data to tissue data, and separate confounding factors from them. Extensive studies on multiple data sets demonstrate that CODE-AE significantly improves the accuracy and robustness over state-of-the-art methods in both predicting patient drug response and de-confounding biological signals. Thus, CODE-AE provides a useful framework to take advantage of in vitro omics data for developing generalized patient predictive models. The source code is available at https://github.com/XieResearchGroup/CODE-AE.

Guolin Ke, Di He, Tie-Yan Liu

In this work, we investigate the positional encoding methods used in language pre-training (e.g., BERT) and identify several problems in the existing formulations. First, we show that in the absolute positional encoding, the addition operation applied on positional embeddings and word embeddings brings mixed correlations between the two heterogeneous information resources. It may bring unnecessary randomness in the attention and further limit the expressiveness of the model. Second, we question whether treating the position of the symbol \texttt{[CLS]} the same as other words is a reasonable design, considering its special role (the representation of the entire sentence) in the downstream tasks. Motivated from above analysis, we propose a new positional encoding method called \textbf{T}ransformer with \textbf{U}ntied \textbf{P}ositional \textbf{E}ncoding (TUPE). In the self-attention module, TUPE computes the word contextual correlation and positional correlation separately with different parameterizations and then adds them together. This design removes the mixed and noisy correlations over heterogeneous embeddings and offers more expressiveness by using different projection matrices. Furthermore, TUPE unties the \texttt{[CLS]} symbol from other positions, making it easier to capture information from all positions. Extensive experiments and ablation studies on GLUE benchmark demonstrate the effectiveness of the proposed method. Codes and models are released at https://github.com/guolinke/TUPE.

Jinhua Zhu, Yingce Xia, Lijun Wu et al.

The recently proposed BERT has shown great power on a variety of natural language understanding tasks, such as text classification, reading comprehension, etc. However, how to effectively apply BERT to neural machine translation (NMT) lacks enough exploration. While BERT is more commonly used as fine-tuning instead of contextual embedding for downstream language understanding tasks, in NMT, our preliminary exploration of using BERT as contextual embedding is better than using for fine-tuning. This motivates us to think how to better leverage BERT for NMT along this direction. We propose a new algorithm named BERT-fused model, in which we first use BERT to extract representations for an input sequence, and then the representations are fused with each layer of the encoder and decoder of the NMT model through attention mechanisms. We conduct experiments on supervised (including sentence-level and document-level translations), semi-supervised and unsupervised machine translation, and achieve state-of-the-art results on seven benchmark datasets. Our code is available at \url{https://github.com/bert-nmt/bert-nmt}.

Yixian Xu, Shengjie Luo, Liwei Wang et al.

Diffusion models have achieved remarkable success in generative modeling. Despite more stable training, the loss of diffusion models is not indicative of absolute data-fitting quality, since its optimal value is typically not zero but unknown, leading to confusion between large optimal loss and insufficient model capacity. In this work, we advocate the need to estimate the optimal loss value for diagnosing and improving diffusion models. We first derive the optimal loss in closed form under a unified formulation of diffusion models, and develop effective estimators for it, including a stochastic variant scalable to large datasets with proper control of variance and bias. With this tool, we unlock the inherent metric for diagnosing the training quality of mainstream diffusion model variants, and develop a more performant training schedule based on the optimal loss. Moreover, using models with 120M to 1.5B parameters, we find that the power law is better demonstrated after subtracting the optimal loss from the actual training loss, suggesting a more principled setting for investigating the scaling law for diffusion models.

Andreas Schwarz, Di He, Maarten Van Segbroeck et al.

Streaming Automatic Speech Recognition (ASR) in voice assistants can utilize prefetching to partially hide the latency of response generation. Prefetching involves passing a preliminary ASR hypothesis to downstream systems in order to prefetch and cache a response. If the final ASR hypothesis after endpoint detection matches the preliminary one, the cached response can be delivered to the user, thus saving latency. In this paper, we extend this idea by introducing predictive automatic speech recognition, where we predict the full utterance from a partially observed utterance, and prefetch the response based on the predicted utterance. We introduce two personalization approaches and investigate the tradeoff between potential latency gains from successful predictions and the cost increase from failed predictions. We evaluate our methods on an internal voice assistant dataset as well as the public SLURP dataset.

Chengxuan Ying, Mingqi Yang, Shuxin Zheng et al.

In this technical report, we present our solution of KDD Cup 2021 OGB Large-Scale Challenge - PCQM4M-LSC Track. We adopt Graphormer and ExpC as our basic models. We train each model by 8-fold cross-validation, and additionally train two Graphormer models on the union of training and validation sets with different random seeds. For final submission, we use a naive ensemble for these 18 models by taking average of their outputs. Using our method, our team MachineLearning achieved 0.1200 MAE on test set, which won the first place in KDD Cup graph prediction track.

Samik Sadhu, Di He, Che-Wei Huang et al.

Wav2vec-C introduces a novel representation learning technique combining elements from wav2vec 2.0 and VQ-VAE. Our model learns to reproduce quantized representations from partially masked speech encoding using a contrastive loss in a way similar to Wav2vec 2.0. However, the quantization process is regularized by an additional consistency network that learns to reconstruct the input features to the wav2vec 2.0 network from the quantized representations in a way similar to a VQ-VAE model. The proposed self-supervised model is trained on 10k hours of unlabeled data and subsequently used as the speech encoder in a RNN-T ASR model and fine-tuned with 1k hours of labeled data. This work is one of only a few studies of self-supervised learning on speech tasks with a large volume of real far-field labeled data. The Wav2vec-C encoded representations achieves, on average, twice the error reduction over baseline and a higher codebook utilization in comparison to wav2vec 2.0

Bohang Zhang, Tianle Cai, Zhou Lu et al.

It is well-known that standard neural networks, even with a high classification accuracy, are vulnerable to small $\ell_\infty$-norm bounded adversarial perturbations. Although many attempts have been made, most previous works either can only provide empirical verification of the defense to a particular attack method, or can only develop a certified guarantee of the model robustness in limited scenarios. In this paper, we seek for a new approach to develop a theoretically principled neural network that inherently resists $\ell_\infty$ perturbations. In particular, we design a novel neuron that uses $\ell_\infty$-distance as its basic operation (which we call $\ell_\infty$-dist neuron), and show that any neural network constructed with $\ell_\infty$-dist neurons (called $\ell_{\infty}$-dist net) is naturally a 1-Lipschitz function with respect to $\ell_\infty$-norm. This directly provides a rigorous guarantee of the certified robustness based on the margin of prediction outputs. We then prove that such networks have enough expressive power to approximate any 1-Lipschitz function with robust generalization guarantee. We further provide a holistic training strategy that can greatly alleviate optimization difficulties. Experimental results show that using $\ell_{\infty}$-dist nets as basic building blocks, we consistently achieve state-of-the-art performance on commonly used datasets: 93.09% certified accuracy on MNIST ($ε=0.3$), 35.42% on CIFAR-10 ($ε=8/255$) and 16.31% on TinyImageNet ($ε=1/255$).

Di He, Lei Xie

An unsolved fundamental problem in biology and ecology is to predict observable traits (phenotypes) from a new genetic constitution (genotype) of an organism under environmental perturbations (e.g., drug treatment). The emergence of multiple omics data provides new opportunities but imposes great challenges in the predictive modeling of genotype-phenotype associations. Firstly, the high-dimensionality of genomics data and the lack of labeled data often make the existing supervised learning techniques less successful. Secondly, it is a challenging task to integrate heterogeneous omics data from different resources. Finally, the information transmission from DNA to phenotype involves multiple intermediate levels of RNA, protein, metabolite, etc. The higher-level features (e.g., gene expression) usually have stronger discriminative power than the lower level features (e.g., somatic mutation). To address above issues, we proposed a novel Cross-LEvel Information Transmission network (CLEIT) framework. CLEIT aims to explicitly model the asymmetrical multi-level organization of the biological system. Inspired by domain adaptation, CLEIT first learns the latent representation of high-level domain then uses it as ground-truth embedding to improve the representation learning of the low-level domain in the form of contrastive loss. In addition, we adopt a pre-training-fine-tuning approach to leveraging the unlabeled heterogeneous omics data to improve the generalizability of CLEIT. We demonstrate the effectiveness and performance boost of CLEIT in predicting anti-cancer drug sensitivity from somatic mutations via the assistance of gene expressions when compared with state-of-the-art methods.

Zhenhui Xu, Linyuan Gong, Guolin Ke et al.

Pre-trained contextual representations (e.g., BERT) have become the foundation to achieve state-of-the-art results on many NLP tasks. However, large-scale pre-training is computationally expensive. ELECTRA, an early attempt to accelerate pre-training, trains a discriminative model that predicts whether each input token was replaced by a generator. Our studies reveal that ELECTRA's success is mainly due to its reduced complexity of the pre-training task: the binary classification (replaced token detection) is more efficient to learn than the generation task (masked language modeling). However, such a simplified task is less semantically informative. To achieve better efficiency and effectiveness, we propose a novel meta-learning framework, MC-BERT. The pre-training task is a multi-choice cloze test with a reject option, where a meta controller network provides training input and candidates. Results over GLUE natural language understanding benchmark demonstrate that our proposed method is both efficient and effective: it outperforms baselines on GLUE semantic tasks given the same computational budget.

Runtian Zhai, Chen Dan, Di He et al.

Adversarial training is one of the most popular ways to learn robust models but is usually attack-dependent and time costly. In this paper, we propose the MACER algorithm, which learns robust models without using adversarial training but performs better than all existing provable l2-defenses. Recent work shows that randomized smoothing can be used to provide a certified l2 radius to smoothed classifiers, and our algorithm trains provably robust smoothed classifiers via MAximizing the CErtified Radius (MACER). The attack-free characteristic makes MACER faster to train and easier to optimize. In our experiments, we show that our method can be applied to modern deep neural networks on a wide range of datasets, including Cifar-10, ImageNet, MNIST, and SVHN. For all tasks, MACER spends less training time than state-of-the-art adversarial training algorithms, and the learned models achieve larger average certified radius.

Zhiqing Sun, Zhuohan Li, Haoqing Wang et al.

Autoregressive sequence models achieve state-of-the-art performance in domains like machine translation. However, due to the autoregressive factorization nature, these models suffer from heavy latency during inference. Recently, non-autoregressive sequence models were proposed to reduce the inference time. However, these models assume that the decoding process of each token is conditionally independent of others. Such a generation process sometimes makes the output sentence inconsistent, and thus the learned non-autoregressive models could only achieve inferior accuracy compared to their autoregressive counterparts. To improve then decoding consistency and reduce the inference cost at the same time, we propose to incorporate a structured inference module into the non-autoregressive models. Specifically, we design an efficient approximation for Conditional Random Fields (CRF) for non-autoregressive sequence models, and further propose a dynamic transition technique to model positional contexts in the CRF. Experiments in machine translation show that while increasing little latency (8~14ms), our model could achieve significantly better translation performance than previous non-autoregressive models on different translation datasets. In particular, for the WMT14 En-De dataset, our model obtains a BLEU score of 26.80, which largely outperforms the previous non-autoregressive baselines and is only 0.61 lower in BLEU than purely autoregressive models.

Jun Gao, Di He, Xu Tan et al.

We study an interesting problem in training neural network-based models for natural language generation tasks, which we call the \emph{representation degeneration problem}. We observe that when training a model for natural language generation tasks through likelihood maximization with the weight tying trick, especially with big training datasets, most of the learnt word embeddings tend to degenerate and be distributed into a narrow cone, which largely limits the representation power of word embeddings. We analyze the conditions and causes of this problem and propose a novel regularization method to address it. Experiments on language modeling and machine translation show that our method can largely mitigate the representation degeneration problem and achieve better performance than baseline algorithms.

Xu Tan, Yi Ren, Di He et al.

Multilingual machine translation, which translates multiple languages with a single model, has attracted much attention due to its efficiency of offline training and online serving. However, traditional multilingual translation usually yields inferior accuracy compared with the counterpart using individual models for each language pair, due to language diversity and model capacity limitations. In this paper, we propose a distillation-based approach to boost the accuracy of multilingual machine translation. Specifically, individual models are first trained and regarded as teachers, and then the multilingual model is trained to fit the training data and match the outputs of individual models simultaneously through knowledge distillation. Experiments on IWSLT, WMT and Ted talk translation datasets demonstrate the effectiveness of our method. Particularly, we show that one model is enough to handle multiple languages (up to 44 languages in our experiment), with comparable or even better accuracy than individual models.

Yiren Wang, Fei Tian, Di He et al.

As a new neural machine translation approach, Non-Autoregressive machine Translation (NAT) has attracted attention recently due to its high efficiency in inference. However, the high efficiency has come at the cost of not capturing the sequential dependency on the target side of translation, which causes NAT to suffer from two kinds of translation errors: 1) repeated translations (due to indistinguishable adjacent decoder hidden states), and 2) incomplete translations (due to incomplete transfer of source side information via the decoder hidden states). In this paper, we propose to address these two problems by improving the quality of decoder hidden representations via two auxiliary regularization terms in the training process of an NAT model. First, to make the hidden states more distinguishable, we regularize the similarity between consecutive hidden states based on the corresponding target tokens. Second, to force the hidden states to contain all the information in the source sentence, we leverage the dual nature of translation tasks (e.g., English to German and German to English) and minimize a backward reconstruction error to ensure that the hidden states of the NAT decoder are able to recover the source side sentence. Extensive experiments conducted on several benchmark datasets show that both regularization strategies are effective and can alleviate the issues of repeated translations and incomplete translations in NAT models. The accuracy of NAT models is therefore improved significantly over the state-of-the-art NAT models with even better efficiency for inference.

Chengyue Gong, Xu Tan, Di He et al.

Maximum-likelihood estimation (MLE) is widely used in sequence to sequence tasks for model training. It uniformly treats the generation/prediction of each target token as multi-class classification, and yields non-smooth prediction probabilities: in a target sequence, some tokens are predicted with small probabilities while other tokens are with large probabilities. According to our empirical study, we find that the non-smoothness of the probabilities results in low quality of generated sequences. In this paper, we propose a sentence-wise regularization method which aims to output smooth prediction probabilities for all the tokens in the target sequence. Our proposed method can automatically adjust the weights and gradients of each token in one sentence to ensure the predictions in a sequence uniformly well. Experiments on three neural machine translation tasks and one text summarization task show that our method outperforms conventional MLE loss on all these tasks and achieves promising BLEU scores on WMT14 English-German and WMT17 Chinese-English translation task.

Chengyue Gong, Di He, Xu Tan et al.

Continuous word representation (aka word embedding) is a basic building block in many neural network-based models used in natural language processing tasks. Although it is widely accepted that words with similar semantics should be close to each other in the embedding space, we find that word embeddings learned in several tasks are biased towards word frequency: the embeddings of high-frequency and low-frequency words lie in different subregions of the embedding space, and the embedding of a rare word and a popular word can be far from each other even if they are semantically similar. This makes learned word embeddings ineffective, especially for rare words, and consequently limits the performance of these neural network models. In this paper, we develop a neat, simple yet effective way to learn \emph{FRequency-AGnostic word Embedding} (FRAGE) using adversarial training. We conducted comprehensive studies on ten datasets across four natural language processing tasks, including word similarity, language modeling, machine translation and text classification. Results show that with FRAGE, we achieve higher performance than the baselines in all tasks.

Zhuohan Li, Di He, Fei Tian et al.

Long Short-Term Memory (LSTM) is one of the most widely used recurrent structures in sequence modeling. It aims to use gates to control information flow (e.g., whether to skip some information or not) in the recurrent computations, although its practical implementation based on soft gates only partially achieves this goal. In this paper, we propose a new way for LSTM training, which pushes the output values of the gates towards 0 or 1. By doing so, we can better control the information flow: the gates are mostly open or closed, instead of in a middle state, which makes the results more interpretable. Empirical studies show that (1) Although it seems that we restrict the model capacity, there is no performance drop: we achieve better or comparable performances due to its better generalization ability; (2) The outputs of gates are not sensitive to their inputs: we can easily compress the LSTM unit in multiple ways, e.g., low-rank approximation and low-precision approximation. The compressed models are even better than the baseline models without compression.

Fei Tian, Bin Gao, Di He et al.

We propose Sentence Level Recurrent Topic Model (SLRTM), a new topic model that assumes the generation of each word within a sentence to depend on both the topic of the sentence and the whole history of its preceding words in the sentence. Different from conventional topic models that largely ignore the sequential order of words or their topic coherence, SLRTM gives full characterization to them by using a Recurrent Neural Networks (RNN) based framework. Experimental results have shown that SLRTM outperforms several strong baselines on various tasks. Furthermore, SLRTM can automatically generate sentences given a topic (i.e., topics to sentences), which is a key technology for real world applications such as personalized short text conversation.

Di He, Wei Chen, Liwei Wang et al.

Sponsored search is an important monetization channel for search engines, in which an auction mechanism is used to select the ads shown to users and determine the prices charged from advertisers. There have been several pieces of work in the literature that investigate how to design an auction mechanism in order to optimize the revenue of the search engine. However, due to some unrealistic assumptions used, the practical values of these studies are not very clear. In this paper, we propose a novel \emph{game-theoretic machine learning} approach, which naturally combines machine learning and game theory, and learns the auction mechanism using a bilevel optimization framework. In particular, we first learn a Markov model from historical data to describe how advertisers change their bids in response to an auction mechanism, and then for any given auction mechanism, we use the learnt model to predict its corresponding future bid sequences. Next we learn the auction mechanism through empirical revenue maximization on the predicted bid sequences. We show that the empirical revenue will converge when the prediction period approaches infinity, and a Genetic Programming algorithm can effectively optimize this empirical revenue. Our experiments indicate that the proposed approach is able to produce a much more effective auction mechanism than several baselines.

Yining Wang, Liwei Wang, Yuanzhi Li et al.

A central problem in ranking is to design a ranking measure for evaluation of ranking functions. In this paper we study, from a theoretical perspective, the widely used Normalized Discounted Cumulative Gain (NDCG)-type ranking measures. Although there are extensive empirical studies of NDCG, little is known about its theoretical properties. We first show that, whatever the ranking function is, the standard NDCG which adopts a logarithmic discount, converges to 1 as the number of items to rank goes to infinity. On the first sight, this result is very surprising. It seems to imply that NDCG cannot differentiate good and bad ranking functions, contradicting to the empirical success of NDCG in many applications. In order to have a deeper understanding of ranking measures in general, we propose a notion referred to as consistent distinguishability. This notion captures the intuition that a ranking measure should have such a property: For every pair of substantially different ranking functions, the ranking measure can decide which one is better in a consistent manner on almost all datasets. We show that NDCG with logarithmic discount has consistent distinguishability although it converges to the same limit for all ranking functions. We next characterize the set of all feasible discount functions for NDCG according to the concept of consistent distinguishability. Specifically we show that whether NDCG has consistent distinguishability depends on how fast the discount decays, and 1/r is a critical point. We then turn to the cut-off version of NDCG, i.e., NDCG@k. We analyze the distinguishability of NDCG@k for various choices of k and the discount functions. Experimental results on real Web search datasets agree well with the theory.