Searcharxiv⌕ Search

arXiv subjects

Yu Yan

Publications and source records attributed to Yu Yan.

At least 73 records · Page 4Linked to original sources
arXiv2023

Tunable Coupling Architectures with Capacitively Connecting Pads for Large-Scale Superconducting Multi-Qubit Processors

We have proposed and experimentally verified a tunable inter-qubit coupling scheme for large-scale integration of superconducting qubits. The key feature of the scheme is the insertion of connecting pads between qubit and tunable coupling element. In such a way, the distance between two qubits can be increased considerably to a few millimeters, leaving enough space for arranging control lines, readout resonators and other necessary structures. The increased inter-qubit distance provides more wiring space for flip-chip process and reduces crosstalk between qubits and from control lines to qubits. We use the term Tunable Coupler with Capacitively Connecting Pad (TCCP) to name the tunable coupling part that consists of a transmon coupler and capacitively connecting pads. With the different placement of connecting pads, different TCCP architectures can be realized. We have designed and fabricated a few multi-qubit devices in which TCCP is used for coupling. The measured results show that the performance of the qubits coupled by the TCCP, such as $T_1$ and $T_2$, was similar to that of the traditional transmon qubits without TCCP. Meanwhile, our TCCP also exhibited a wide tunable range of the effective coupling strength and a low residual ZZ interaction between the qubits by properly tuning the parameters on the design. Finally, we successfully implemented an adiabatic CZ gate with TCCP. Furthermore, by introducing TCCP, we also discuss the realization of the flip-chip process and tunable coupling qubits between different chips.

quant-ph↗
arXiv2022

AAE: An Active Auto-Estimator for Improving Graph Storage

Nowadays, graph becomes an increasingly popular model in many real applications. The efficiency of graph storage is crucial for these applications. Generally speaking, the tune tasks of graph storage rely on the database administrators (DBAs) to find the best graph storage. However, DBAs make the tune decisions by mainly relying on their experiences and intuition. Due to the limitations of DBAs's experiences, the tunes may have an uncertain performance and conduct worse efficiency. In this paper, we observe that an estimator of graph workload has the potential to guarantee the performance of tune operations. Unfortunately, because of the complex characteristics of graph evaluation task, there exists no mature estimator for graph workload. We formulate the evaluation task of graph workload as a classification task and carefully design the feature engineering process, including graph data features, graph workload features and graph storage features. Considering the complex features of graph and the huge time consumption in graph workload execution, it is difficult for the graph workload estimator to obtain enough training set. So, we propose an active auto-estimator (AAE) for the graph workload evaluation by combining the active learning and deep learning. AAE could achieve good evaluation efficiency with limited training set. We test the time efficiency and evaluation accuracy of AAE with two open source graph data, LDBC and Freebase. Experimental results show that our estimator could efficiently complete the graph workload evaluation in milliseconds.

cs.DB↗
arXiv2022

A Self-Paced Mixed Distillation Method for Non-Autoregressive Generation

Non-Autoregressive generation is a sequence generation paradigm, which removes the dependency between target tokens. It could efficiently reduce the text generation latency with parallel decoding in place of token-by-token sequential decoding. However, due to the known multi-modality problem, Non-Autoregressive (NAR) models significantly under-perform Auto-regressive (AR) models on various language generation tasks. Among the NAR models, BANG is the first large-scale pre-training model on English un-labeled raw text corpus. It considers different generation paradigms as its pre-training tasks including Auto-regressive (AR), Non-Autoregressive (NAR), and semi-Non-Autoregressive (semi-NAR) information flow with multi-stream strategy. It achieves state-of-the-art performance without any distillation techniques. However, AR distillation has been shown to be a very effective solution for improving NAR performance. In this paper, we propose a novel self-paced mixed distillation method to further improve the generation quality of BANG. Firstly, we propose the mixed distillation strategy based on the AR stream knowledge. Secondly, we encourage the model to focus on the samples with the same modality by self-paced learning. The proposed self-paced mixed distillation algorithm improves the generation quality and has no influence on the inference latency. We carry out extensive experiments on summarization and question generation tasks to validate the effectiveness. To further illustrate the commercial value of our approach, we conduct experiments on three generation tasks in real-world advertisements applications. Experimental results on commercial data show the effectiveness of the proposed model. Compared with BANG, it achieves significant BLEU score improvement. On the other hand, compared with auto-regressive generation method, it achieves more than 7x speedup.

cs.CL↗
arXiv2021

Factorisation-based Image Labelling

Segmentation of brain magnetic resonance images (MRI) into anatomical regions is a useful task in neuroimaging. Manual annotation is time consuming and expensive, so having a fully automated and general purpose brain segmentation algorithm is highly desirable. To this end, we propose a patched-based label propagation approach based on a generative model with latent variables. Once trained, our Factorisation-based Image Labelling (FIL) model is able to label target images with a variety of image contrasts. We compare the effectiveness of our proposed model against the state-of-the-art using data from the MICCAI 2012 Grand Challenge and Workshop on Multi-Atlas Labeling. As our approach is intended to be general purpose, we also assess how well it can handle domain shift by labelling images of the same subjects acquired with different MR contrasts.

cs.CV↗
arXiv2021

Experience-Enhanced Learning: One Size Still does not Fit All in Automatic Database

Recent years, the database committee has attempted to develop automatic database management systems. Although some researches show that the applying AI to data management is a significant and promising direction, there still exists many problems in implementing these techniques to real applications (long training time, various environments and unstable performance). In this paper, we discover that traditional rule based methods have the potential to solve the above problems. We propose three methodologies for improving learned methods, i.e. label collection for efficiently pre-training, knowledge base for model transfer and theoretical guarantee for stable performance. We implement our methodologies on two widely used learning approaches, deep learning and reinforcement learning. Firstly, the novel experience enhanced deep learning (EEDL) could achieve efficient training and stable performance. We evaluate EEDL with cardinality estimation, an essential database management. The experimental results on four real dataset [1] show that our EEDL could outperforms the general DL model [2]. Secondly, we design a novel experience-enhanced reinforcement learning (EERL), which could efficiently converge and has better performance than general RL models [3]. We test EERL with online index tuning task. The experiments on TPC-H shows that EERL could accelerate the convergence of agent and generate better solution that generalizes the reinforcement learning.

cs.DB↗
arXiv2021

FastSeq: Make Sequence Generation Faster

Transformer-based models have made tremendous impacts in natural language generation. However the inference speed is a bottleneck due to large model size and intensive computing involved in auto-regressive decoding process. We develop FastSeq framework to accelerate sequence generation without accuracy loss. The proposed optimization techniques include an attention cache optimization, an efficient algorithm for detecting repeated n-grams, and an asynchronous generation pipeline with parallel I/O. These optimizations are general enough to be applicable to Transformer-based models (e.g., T5, GPT2, and UniLM). Our benchmark results on a set of widely used and diverse models demonstrate 4-9x inference speed gain. Additionally, FastSeq is easy to use with a simple one-line code change. The source code is available at https://github.com/microsoft/fastseq.

cs.CL↗
arXiv2021

ProphetNet-X: Large-Scale Pre-training Models for English, Chinese, Multi-lingual, Dialog, and Code Generation

Now, the pre-training technique is ubiquitous in natural language processing field. ProphetNet is a pre-training based natural language generation method which shows powerful performance on English text summarization and question generation tasks. In this paper, we extend ProphetNet into other domains and languages, and present the ProphetNet family pre-training models, named ProphetNet-X, where X can be English, Chinese, Multi-lingual, and so on. We pre-train a cross-lingual generation model ProphetNet-Multi, a Chinese generation model ProphetNet-Zh, two open-domain dialog generation models ProphetNet-Dialog-En and ProphetNet-Dialog-Zh. And also, we provide a PLG (Programming Language Generation) model ProphetNet-Code to show the generation performance besides NLG (Natural Language Generation) tasks. In our experiments, ProphetNet-X models achieve new state-of-the-art performance on 10 benchmarks. All the models of ProphetNet-X share the same model structure, which allows users to easily switch between different models. We make the code and models publicly available, and we will keep updating more pre-training models and finetuning scripts.

cs.CL↗
arXiv2021

BANG: Bridging Autoregressive and Non-autoregressive Generation with Large Scale Pretraining

In this paper, we propose BANG, a new pretraining model to Bridge the gap between Autoregressive (AR) and Non-autoregressive (NAR) Generation. AR and NAR generation can be uniformly regarded as to what extent previous tokens can be attended, and BANG bridges AR and NAR generation by designing a novel model structure for large-scale pretraining. The pretrained BANG model can simultaneously support AR, NAR and semi-NAR generation to meet different requirements. Experiments on question generation (SQuAD 1.1), summarization (XSum) and dialogue generation (PersonaChat) show that BANG improves NAR and semi-NAR performance significantly as well as attaining comparable performance with strong AR pretrained models. Compared with the semi-NAR strong baselines, BANG achieves absolute improvements of 14.01 and 5.24 in the overall scores of SQuAD 1.1 and XSum, respectively. In addition, BANG achieves absolute improvements of 10.73, 6.39 and 5.90 in the overall scores of SQuAD, XSUM and PersonaChat respectively compared with the strong NAR baselines.

cs.CL↗
arXiv2021

EL-Attention: Memory Efficient Lossless Attention for Generation

Transformer model with multi-head attention requires caching intermediate results for efficient inference in generation tasks. However, cache brings new memory-related costs and prevents leveraging larger batch size for faster speed. We propose memory-efficient lossless attention (called EL-attention) to address this issue. It avoids heavy operations for building multi-head keys and values, cache for them is not needed. EL-attention constructs an ensemble of attention results by expanding query while keeping key and value shared. It produces the same result as multi-head attention with less GPU memory and faster inference speed. We conduct extensive experiments on Transformer, BART, and GPT-2 for summarization and question generation tasks. The results show EL-attention speeds up existing models by 1.6x to 5.3x without accuracy loss.

cs.CL↗
arXiv2021

GLGE: A New General Language Generation Evaluation Benchmark

Multi-task benchmarks such as GLUE and SuperGLUE have driven great progress of pretraining and transfer learning in Natural Language Processing (NLP). These benchmarks mostly focus on a range of Natural Language Understanding (NLU) tasks, without considering the Natural Language Generation (NLG) models. In this paper, we present the General Language Generation Evaluation (GLGE), a new multi-task benchmark for evaluating the generalization capabilities of NLG models across eight language generation tasks. For each task, we continue to design three subtasks in terms of task difficulty (GLGE-Easy, GLGE-Medium, and GLGE-Hard). This introduces 24 subtasks to comprehensively compare model performance. To encourage research on pretraining and transfer learning on NLG models, we make GLGE publicly available and build a leaderboard with strong baselines including MASS, BART, and ProphetNet (The source code and dataset are publicly available at https://github.com/microsoft/glge).

cs.CL↗
arXiv2021

Multi-SQL: An extensible multi-model data query language

Big data management aims to establish data hubs that support data in multiple models and types in an all-around way. Thus, the multi-model database system is a promising architecture for building such a multi-model data store. For an integrated data hub, a unified and flexible query language is incredibly necessary. In this paper, an extensible and practical query language--Multi-SQL is proposed to realize the unified management of multi-model data considering the co-processing of multi-model data. To the best of our knowledge, Multi-SQL is the first query language based on various data models. Multi-SQL can also be expanded to suit more complicated scenarios as it is flexible to support other data models. Moreover, we provide a formal semantic definition of the core features of Multi-SQL, including the multi-model definition, multi-model filters, multi-model joins, etc. Furthermore, we propose a two-level query implementation method to totally exploit the existing query optimization capabilities of the underlying engines which could largely improve the query excution efficiency.

cs.DB↗
arXiv2020

Clinical Temporal Relation Extraction with Probabilistic Soft Logic Regularization and Global Inference

There has been a steady need in the medical community to precisely extract the temporal relations between clinical events. In particular, temporal information can facilitate a variety of downstream applications such as case report retrieval and medical question answering. Existing methods either require expensive feature engineering or are incapable of modeling the global relational dependencies among the events. In this paper, we propose a novel method, Clinical Temporal ReLation Exaction with Probabilistic Soft Logic Regularization and Global Inference (CTRL-PG) to tackle the problem at the document level. Extensive experiments on two benchmark datasets, I2B2-2012 and TB-Dense, demonstrate that CTRL-PG significantly outperforms baseline methods for temporal relation extraction.

cs.CL↗
arXiv2020

ProphetNet: Predicting Future N-gram for Sequence-to-Sequence Pre-training

This paper presents a new sequence-to-sequence pre-training model called ProphetNet, which introduces a novel self-supervised objective named future n-gram prediction and the proposed n-stream self-attention mechanism. Instead of optimizing one-step-ahead prediction in the traditional sequence-to-sequence model, the ProphetNet is optimized by n-step ahead prediction that predicts the next n tokens simultaneously based on previous context tokens at each time step. The future n-gram prediction explicitly encourages the model to plan for the future tokens and prevent overfitting on strong local correlations. We pre-train ProphetNet using a base scale dataset (16GB) and a large-scale dataset (160GB), respectively. Then we conduct experiments on CNN/DailyMail, Gigaword, and SQuAD 1.1 benchmarks for abstractive summarization and question generation tasks. Experimental results show that ProphetNet achieves new state-of-the-art results on all these datasets compared to the models using the same scale pre-training corpus.

cs.CL↗
arXiv2020

ProphetNet-Ads: A Looking Ahead Strategy for Generative Retrieval Models in Sponsored Search Engine

In a sponsored search engine, generative retrieval models are recently proposed to mine relevant advertisement keywords for users' input queries. Generative retrieval models generate outputs token by token on a path of the target library prefix tree (Trie), which guarantees all of the generated outputs are legal and covered by the target library. In actual use, we found several typical problems caused by Trie-constrained searching length. In this paper, we analyze these problems and propose a looking ahead strategy for generative retrieval models named ProphetNet-Ads. ProphetNet-Ads improves the retrieval ability by directly optimizing the Trie-constrained searching space. We build a dataset from a real-word sponsored search engine and carry out experiments to analyze different generative retrieval models. Compared with Trie-based LSTM generative retrieval model proposed recently, our single model result and integrated result improve the recall by 15.58\% and 18.8\% respectively with beam size 5. Case studies further demonstrate how these problems are alleviated by ProphetNet-Ads clearly.

cs.IR↗
arXiv2020

Diverse, Controllable, and Keyphrase-Aware: A Corpus and Method for News Multi-Headline Generation

News headline generation aims to produce a short sentence to attract readers to read the news. One news article often contains multiple keyphrases that are of interest to different users, which can naturally have multiple reasonable headlines. However, most existing methods focus on the single headline generation. In this paper, we propose generating multiple headlines with keyphrases of user interests, whose main idea is to generate multiple keyphrases of interest to users for the news first, and then generate multiple keyphrase-relevant headlines. We propose a multi-source Transformer decoder, which takes three sources as inputs: (a) keyphrase, (b) keyphrase-filtered article, and (c) original article to generate keyphrase-relevant, high-quality, and diverse headlines. Furthermore, we propose a simple and effective method to mine the keyphrases of interest in the news article and build a first large-scale keyphrase-aware news headline corpus, which contains over 180K aligned triples of $<$news article, headline, keyphrase$>$. Extensive experimental comparisons on the real-world dataset show that the proposed method achieves state-of-the-art results in terms of quality and diversity

cs.CL↗
arXiv2020

Tell Me How to Ask Again: Question Data Augmentation with Controllable Rewriting in Continuous Space

In this paper, we propose a novel data augmentation method, referred to as Controllable Rewriting based Question Data Augmentation (CRQDA), for machine reading comprehension (MRC), question generation, and question-answering natural language inference tasks. We treat the question data augmentation task as a constrained question rewriting problem to generate context-relevant, high-quality, and diverse question data samples. CRQDA utilizes a Transformer autoencoder to map the original discrete question into a continuous embedding space. It then uses a pre-trained MRC model to revise the question representation iteratively with gradient-based optimization. Finally, the revised question representations are mapped back into the discrete space, which serve as additional question data. Comprehensive experiments on SQuAD 2.0, SQuAD 1.1 question generation, and QNLI tasks demonstrate the effectiveness of CRQDA

cs.CL↗
arXiv2020

S-DIGing: A Stochastic Gradient Tracking Algorithm for Distributed Optimization

In this paper, we study convex optimization problems where agents of a network cooperatively minimize the global objective function which consists of multiple local objective functions. Different from most of the existing works, the local objective function of each agent is presented as the average of finite instantaneous functions. The intention of this work is to solve large-scale optimization problems where the local objective function is complicated and numerous. Integrating the gradient tracking algorithm with stochastic averaging gradient technology, we propose a novel distributed stochastic gradient tracking (termed as S-DIGing) algorithm. At each time instant, only one randomly selected gradient of a instantaneous function is computed and applied to approximate the gradient of local objection function. Based on a primal-dual interpretation of the S-DIGing algorithm, we show that the S-DIGing algorithm linearly converges to the global optimal solution when step-size lies in an explicit internal under the assumptions that the instantaneous functions are strongly convex and have Lipschitz-continuous gradient. Numerical experiments on the logistic regression problem are presented to demonstrate the practicability of the algorithm and correctness of the theoretical results.

math.OC↗