Search

Existing research has shown that a multilingual pre-trained language model fine-tuned with one (source) language also performs well on downstream tasks for non-source languages, even though no fine-tuning is done on these languages. However, there is a clear gap between the performance of the source language and that of the non-source languages. This paper analyzes the fine-tuning process, discovers when the performance gap changes and identifies which network weights affect the overall performance most. Additionally, the paper seeks to answer to what extent the gap can be reduced by reducing forgetting. Based on the analysis results, a method named Fine-tuning slow and fast with four training policies is proposed to address these issues. Experimental results show the proposed method outperforms baselines by a clear margin., Comment: Accepted by ACL2023 (Long paper)

Grammatical Error Correction (GEC) is the task of correcting errorful sentences into grammatically correct, semantically consistent, and coherent sentences. Popular GEC models either use large-scale synthetic corpora or use a large number of human-designed rules. The former is costly to train, while the latter requires quite a lot of human expertise. In recent years, AMR, a semantic representation framework, has been widely used by many natural language tasks due to its completeness and flexibility. A non-negligible concern is that AMRs of grammatically incorrect sentences may not be exactly reliable. In this paper, we propose the AMR-GEC, a seq-to-seq model that incorporates denoised AMR as additional knowledge. Specifically, We design a semantic aggregated GEC model and explore denoising methods to get AMRs more reliable. Experiments on the BEA-2019 shared task and the CoNLL-2014 shared task have shown that AMR-GEC performs comparably to a set of strong baselines with a large number of synthetic data. Compared with the T5 model with synthetic data, AMR-GEC can reduce the training time by 32\% while inference time is comparable. To the best of our knowledge, we are the first to incorporate AMR for grammatical error correction., Comment: 7 pages, 3 figures, Accepted by ACL findings 2023

Although many large-scale knowledge bases (KBs) claim to contain multilingual information, their support for many non-English languages is often incomplete. This incompleteness gives birth to the task of cross-lingual question answering over knowledge base (xKBQA), which aims to answer questions in languages different from that of the provided KB. One of the major challenges facing xKBQA is the high cost of data annotation, leading to limited resources available for further exploration. Another challenge is mapping KB schemas and natural language expressions in the questions under cross-lingual settings. In this paper, we propose a novel approach for xKBQA in a reading comprehension paradigm. We convert KB subgraphs into passages to narrow the gap between KB schemas and questions, which enables our model to benefit from recent advances in multilingual pre-trained language models (MPLMs) and cross-lingual machine reading comprehension (xMRC). Specifically, we use MPLMs, with considerable knowledge of cross-lingual mappings, for cross-lingual reading comprehension. Existing high-quality xMRC datasets can be further utilized to finetune our model, greatly alleviating the data scarcity issue in xKBQA. Extensive experiments on two xKBQA datasets in 12 languages show that our approach outperforms various baselines and achieves strong few-shot and zero-shot performance. Our dataset and code are released for further research., Comment: 14 pages, 4 figures, EACL 2023 (findings)

Video-grounded dialogue understanding is a challenging problem that requires machine to perceive, parse and reason over situated semantics extracted from weakly aligned video and dialogues. Most existing benchmarks treat both modalities the same as a frame-independent visual understanding task, while neglecting the intrinsic attributes in multimodal dialogues, such as scene and topic transitions. In this paper, we present Video-grounded Scene&Topic AwaRe dialogue (VSTAR) dataset, a large scale video-grounded dialogue understanding dataset based on 395 TV series. Based on VSTAR, we propose two benchmarks for video-grounded dialogue understanding: scene segmentation and topic segmentation, and one benchmark for video-grounded dialogue generation. Comprehensive experiments are performed on these benchmarks to demonstrate the importance of multimodal information and segments in video-grounded dialogue understanding and generation., To appear at ACL 2023

Pre-trained language models(PLM) have made impressive results in various NLP tasks. It has been revealed that one of the key factors to their success is the parameters of these models implicitly learn all kinds of knowledge during pre-training. However, encoding knowledge implicitly in the model parameters has two fundamental drawbacks. First, the knowledge is neither editable nor scalable once the model is trained, which is especially problematic in that knowledge is consistently evolving. Second, it lacks interpretability and prevents humans from understanding which knowledge PLM requires for a certain problem. In this paper, we introduce PlugLM, a pre-training model with differentiable plug-in memory(DPM). The key intuition is to decouple the knowledge storage from model parameters with an editable and scalable key-value memory and leverage knowledge in an explainable manner by knowledge retrieval in the DPM. To justify this design choice, we conduct evaluations in three settings including: (1) domain adaptation. PlugLM obtains 3.95 F1 improvements across four domains on average without any in-domain pre-training. (2) knowledge update. PlugLM could absorb new knowledge in a training-free way after pre-training is done. (3) in-task knowledge learning. PlugLM could be further improved by incorporating training samples into DPM with knowledge prompting., Comment: ACL2023 Findings

The multi-answer phenomenon, where a question may have multiple answers scattered in the document, can be well handled by humans but is challenging enough for machine reading comprehension (MRC) systems. Despite recent progress in multi-answer MRC, there lacks a systematic analysis of how this phenomenon arises and how to better address it. In this work, we design a taxonomy to categorize commonly-seen multi-answer MRC instances, with which we inspect three multi-answer datasets and analyze where the multi-answer challenge comes from. We further analyze how well different paradigms of current multi-answer MRC models deal with different types of multi-answer instances. We find that some paradigms capture well the key information in the questions while others better model the relationship between questions and contexts. We thus explore strategies to make the best of the strengths of different paradigms. Experiments show that generation models can be a promising platform to incorporate different paradigms. Our annotations and code are released for further research., Comment: Findings of ACL 2023

Nowadays, time-stamped web documents related to a general news query floods spread throughout the Internet, and timeline summarization targets concisely summarizing the evolution trajectory of events along the timeline. Unlike traditional document summarization, timeline summarization needs to model the time series information of the input events and summarize important events in chronological order. To tackle this challenge, in this paper, we propose a Unified Timeline Summarizer (UTS) that can generate abstractive and extractive timeline summaries in time order. Concretely, in the encoder part, we propose a graph-based event encoder that relates multiple events according to their content dependency and learns a global representation of each event. In the decoder part, to ensure the chronological order of the abstractive summary, we propose to extract the feature of event-level attention in its generation process with sequential information remained and use it to simulate the evolutionary attention of the ground truth summary. The event-level attention can also be used to assist in extracting summary, where the extracted summary also comes in time sequence. We augment the previous Chinese large-scale timeline summarization dataset and collect a new English timeline dataset. Extensive experiments conducted on these datasets and on the out-of-domain Timeline 17 dataset show that UTS achieves state-of-the-art performance in terms of both automatic and human evaluations., Comment: 30 pages, 12 figures, accepted by TOIS 2022

Enhancing word usage is a desired feature for writing assistance. To further advance research in this area, this paper introduces "Smart Word Suggestions" (SWS) task and benchmark. Unlike other works, SWS emphasizes end-to-end evaluation and presents a more realistic writing assistance scenario. This task involves identifying words or phrases that require improvement and providing substitution suggestions. The benchmark includes human-labeled data for testing, a large distantly supervised dataset for training, and the framework for evaluation. The test data includes 1,000 sentences written by English learners, accompanied by over 16,000 substitution suggestions annotated by 10 native speakers. The training dataset comprises over 3.7 million sentences and 12.7 million suggestions generated through rules. Our experiments with seven baselines demonstrate that SWS is a challenging task. Based on experimental analysis, we suggest potential directions for future research on SWS. The dataset and related codes is available at https://github.com/microsoft/SmartWordSuggestions., Comment: Accepted by Findings of ACL23

With direct access to human-written reference as memory, retrieval-augmented generation has achieved much progress in a wide range of text generation tasks. Since better memory would typically prompt better generation~(we define this as primal problem). The traditional approach for memory retrieval involves selecting memory that exhibits the highest similarity to the input. However, this method is constrained by the quality of the fixed corpus from which memory is retrieved. In this paper, by exploring the duality of the primal problem: better generation also prompts better memory, we propose a novel framework, selfmem, which addresses this limitation by iteratively employing a retrieval-augmented generator to create an unbounded memory pool and using a memory selector to choose one output as memory for the subsequent generation round. This enables the model to leverage its own output, referred to as self-memory, for improved generation. We evaluate the effectiveness of selfmem on three distinct text generation tasks: neural machine translation, abstractive text summarization, and dialogue generation, under two generation paradigms: fine-tuned small model and few-shot LLM. Our approach achieves state-of-the-art results in four directions in JRC-Acquis, XSum (50.3 ROUGE-1), and BigPatent (62.9 ROUGE-1), demonstrating the potential of self-memory in enhancing retrieval-augmented generation models. Furthermore, we conduct thorough analyses of each component in the selfmem framework to identify bottlenecks and provide insights for future research.

Large Language Models (LLMs) have shown remarkable performance in various basic natural language tasks, which raises hope for achieving Artificial General Intelligence. For completing the complex task, we still need a program for the task first and then ask LLMs to follow the program to generate the specific solution. We propose using natural language as a new programming language to describe task procedures, making them easily understandable to both humans and LLMs. ~The LLM is capable of directly generating natural language programs, but these programs may still contain factual errors or incomplete steps. Therefore, we further propose the Learning to Program (\text{LP}) method to ask LLMs themselves to learn the natural language program based on the training dataset of the complex task first and then use the learned program to guide the inference. Our experiments on the reasoning tasks of five different reasoning types (8 datasets) demonstrate the effectiveness of our approach. Further, our analysis experiment shows that the learned program can be directly used to guide another LLM to improve its performance, which reveals a new transfer learning paradigm., Comment: Work in progress

Position embeddings, encoding the positional relationships among tokens in text sequences, make great contributions to modeling local context features in Transformer-based pre-trained language models. However, in Extractive Question Answering, position embeddings trained with instances of varied context lengths may not perform well as we expect. Since the embeddings of rear positions are updated fewer times than the front position embeddings, the rear ones may not be properly trained. In this paper, we propose a simple but effective strategy, Random Padding, without any modifications to architectures of existing pre-trained language models. We adjust the token order of input sequences when fine-tuning, to balance the number of updating times of every position embedding. Experiments show that Random Padding can significantly improve model performance on the instances whose answers are located at rear positions, especially when models are trained on short contexts but evaluated on long contexts. Our code and data will be released for future research.

Breeding Insight was founded to empower breeding by adding genomics tools to small and specialty species breeding programs across USDA ARS.

Spatial commonsense, the knowledge about spatial position and relationship between objects (like the relative size of a lion and a girl, and the position of a boy relative to a bicycle when cycling), is an important part of commonsense knowledge. Although pretrained language models (PLMs) succeed in many NLP tasks, they are shown to be ineffective in spatial commonsense reasoning. Starting from the observation that images are more likely to exhibit spatial commonsense than texts, we explore whether models with visual signals learn more spatial commonsense than text-based PLMs. We propose a spatial commonsense benchmark that focuses on the relative scales of objects, and the positional relationship between people and objects under different actions. We probe PLMs and models with visual signals, including vision-language pretrained models and image synthesis models, on this benchmark, and find that image synthesis models are more capable of learning accurate and consistent spatial knowledge than other models. The spatial knowledge from image synthesis models also helps in natural language understanding tasks that require spatial commonsense., Comment: Accepted by ACL 2022 main conference

With the availability of massive general-domain dialogue data, pre-trained dialogue generation appears to be super appealing to transfer knowledge from the general domain to downstream applications. In most existing work, such transferable ability is mainly obtained by fitting a large model with hundreds of millions of parameters on massive data in an exhaustive way, leading to inefficient running and poor interpretability. This paper proposes a novel dialogue generation model with a latent structure that is easily transferable from the general domain to downstream tasks in a lightweight and transparent way. Experiments on two benchmarks validate the effectiveness of the proposed model. Thanks to the transferable latent structure, our model is able to yield better dialogue responses than four strong baselines in terms of both automatic and human evaluations, and our model with about 22% parameters particularly delivers a 5x speedup in running time compared with the strongest baseline. Moreover, the proposed model is explainable by interpreting the discrete latent variables., Comment: To appear at EMNLP 2022 main conference

To improve the performance of the dual-encoder retriever, one effective approach is knowledge distillation from the cross-encoder ranker. Existing works construct the candidate passages following the supervised learning setting where a query is paired with a positive passage and a batch of negatives. However, through empirical observation, we find that even the hard negatives from advanced methods are still too trivial for the teacher to distinguish, preventing the teacher from transferring abundant dark knowledge to the student through its soft label. To alleviate this issue, we propose ADAM, a knowledge distillation framework that can better transfer the dark knowledge held in the teacher with Adaptive Dark exAMples. Different from previous works that only rely on one positive and hard negatives as candidate passages, we create dark examples that all have moderate relevance to the query through mixing-up and masking in discrete space. Furthermore, as the quality of knowledge held in different training instances varies as measured by the teacher's confidence score, we propose a self-paced distillation strategy that adaptively concentrates on a subset of high-quality instances to conduct our dark-example-based knowledge distillation to help the student learn better. We conduct experiments on two widely-used benchmarks and verify the effectiveness of our method., Comment: 9 pages, 2 figures

Recently, retrieval models based on dense representations are dominant in passage retrieval tasks, due to their outstanding ability in terms of capturing semantics of input text compared to the traditional sparse vector space models. A common practice of dense retrieval models is to exploit a dual-encoder architecture to represent a query and a passage independently. Though efficient, such a structure loses interaction between the query-passage pair, resulting in inferior accuracy. To enhance the performance of dense retrieval models without loss of efficiency, we propose a GNN-encoder model in which query (passage) information is fused into passage (query) representations via graph neural networks that are constructed by queries and their top retrieved passages. By this means, we maintain a dual-encoder structure, and retain some interaction information between query-passage pairs in their representations, which enables us to achieve both efficiency and efficacy in passage retrieval. Evaluation results indicate that our method significantly outperforms the existing models on MSMARCO, Natural Questions and TriviaQA datasets, and achieves the new state-of-the-art on these datasets., Comment: Findings of EMNLP2022

Usually on the top of Smart Chip covered with active shielding layer to prevent invasive physical exploration tampering attacks on part of the chip’s function modules, to obtain the chip’s critical storage data and sensitive information. This paper introduces a design based on UMC55 technology, and applied to the safety chip active shielding layer method for layout design, the layout design from the two aspects of the metal shielding line and shielding layer detecting circuit, using the minimum size advantage and layout design process when the depth of hidden shielding line interface and port order connection method and greatly increased the difficulty of physical attack. The layout design can withstand most of the current FIB physical attack technology, and has been applied to the actual smart card design, and it has important practical significance for the security design and attack of the chip.

In this paper, we present a new verification style reading comprehension dataset named VGaokao from Chinese Language tests of Gaokao. Different from existing efforts, the new dataset is originally designed for native speakers' evaluation, thus requiring more advanced language understanding skills. To address the challenges in VGaokao, we propose a novel Extract-Integrate-Compete approach, which iteratively selects complementary evidence with a novel query updating mechanism and adaptively distills supportive evidence, followed by a pairwise competition to push models to learn the subtle difference among similar text pieces. Experiments show that our methods outperform various baselines on VGaokao with retrieved complementary evidence, while having the merits of efficiency and explainability. Our dataset and code are released for further research., Comment: 11 pages, 2 figures, EMNLP 2021 (findings)

A novel UHF RFID communication link rate adjustment strategy based on the coprime factor algorithm was proposed, and chip-level physical implementation was performed using 0.18um CMOS process. The test result shows that the new reverse link bit rate adjustment scheme not only controls the BLF violation within 5%, effectively avoids electromagnetic interference. The implementation reduces the design complexity of the chip baseband circuit where reverse link realization clock is 640KHz, jitter is 0.8ns and power consumption is 140nA.

Responding with knowledge has been recognized as an important capability for an intelligent conversational agent. Yet knowledge-grounded dialogues, as training data for learning such a response generation model, are difficult to obtain. Motivated by the challenge in practice, we consider knowledge-grounded dialogue generation under a natural assumption that only limited training examples are available. In such a low-resource setting, we devise a disentangled response decoder in order to isolate parameters that depend on knowledge-grounded dialogues from the entire generation model. By this means, the major part of the model can be learned from a large number of ungrounded dialogues and unstructured documents, while the remaining small parameters can be well fitted using the limited training examples. Evaluation results on two benchmarks indicate that with only 1/8 training data, our model can achieve the state-of-the-art performance and generalize well on out-of-domain knowledge., Comment: Published in ICLR 2020

Energy Internet is a new hot topic in the research of industry and academia at home and abroad. It is also the development direction of the energy industry, and it is possible to innovate the way of human life. The paper summarized the concept and connotation by surveying the domestic and international development of the Internet of Energy. The information technology architecture of the Internet of Energy is reviewed from the perspective of the integration of energy system and information technology. This article described the characteristics of power line carrier communication, and analyzed the role and status of power line carrier communication technology in the Internet of Energy. The challenges and opportunities that power line carrier communication faces in the Internet of Energy are analyzed. In the end, this paper discussed the future development of Internet of Energy. Through the “Internet +”technology though, it will integrate advanced information technology and energy systems, coordinate multi-energy networks, innovate energy system operation modes, promote industrial upgrading, and generate new economic growth points.

This paper presents our semantic parsing system for the evaluation task of open domain semantic parsing in NLPCC 2019. Many previous works formulate semantic parsing as a sequence-to-sequence(seq2seq) problem. Instead, we treat the task as a sketch-based problem in a coarse-to-fine(coarse2fine) fashion. The sketch is a high-level structure of the logical form exclusive of low-level details such as entities and predicates. In this way, we are able to optimize each part individually. Specifically, we decompose the process into three stages: the sketch classification determines the high-level structure while the entity labeling and the matching network fill in missing details. Moreover, we adopt the seq2seq method to evaluate logical form candidates from an overall perspective. The co-occurrence relationship between predicates and entities contribute to the reranking as well. Our submitted system achieves the exactly matching accuracy of 82.53% on full test set and 47.83% on hard test subset, which is the 3rd place in NLPCC 2019 Shared Task 2. After optimizations for parameters, network structure and sampling, the accuracy reaches 84.47% on full test set and 63.08% on hard test subset(Our code and data are available at https://github.com/zechagl/NLPCC2019-Semantic-Parsing)., Comment: Accepted by NLPCC 2019

In this paper, we study context-response matching with pre-trained contextualized representations for multi-turn response selection in retrieval-based chatbots. Existing models, such as Cove and ELMo, are trained with limited context (often a single sentence or paragraph), and may not work well on multi-turn conversations, due to the hierarchical nature, informal language, and domain-specific words. To address the challenges, we propose pre-training hierarchical contextualized representations, including contextual word-level and sentence-level representations, by learning a dialogue generation model from large-scale conversations with a hierarchical encoder-decoder architecture. Then the two levels of representations are blended into the input and output layer of a matching model respectively. Experimental results on two benchmark conversation datasets indicate that the proposed hierarchical contextualized representations can bring significantly and consistently improvement to existing matching models for response selection., Comment: 6 pages, 1 figure

Sentence simplification reduces semantic complexity to benefit people with language impairments. Previous simplification studies on the sentence level and word level have achieved promising results but also meet great challenges. For sentence-level studies, sentences after simplification are fluent but sometimes are not really simplified. For word-level studies, words are simplified but also have potential grammar errors due to different usages of words before and after simplification. In this paper, we propose a two-step simplification framework by combining both the word-level and the sentence-level simplifications, making use of their corresponding advantages. Based on the two-step framework, we implement a novel constrained neural generation model to simplify sentences given simplified words. The final results on Wikipedia and Simple Wikipedia aligned datasets indicate that our method yields better performance than various baselines.

Private wireless network communications technology is a key technology of the smart grid, but the scarce spectrum resources greatly restrict the development of the private wireless network communications technology. At present, the 230MHz spectrum can only support a low transfer rate, makes it hard for some emerging business requirements, also restrict the development of traditional business. To solve the emergence of the cognitive radio spectrum resources, realize the dynamic management of spectrum and improve the utilization rate of spectrum, the spectrum sharing technology was proposed to effectively solve the spectrum scarcity. In order to further promote the implementation of the smart grid, this paper puts forward a kind of transmission scheme based on the technology of OFDM and power allocation, and then puts forward a new spectrum planning scheme based on spectrum sharing. The simulation results show that the scheme can effectively increase the system capacity, give a planning scheme of 230MHz spectrum to the State Radio Regulatory Commission of the People's Republic of China(SRRC), and provides a good reference to promote the development of smart grid provides.

Open-domain human-computer conversation has attracted much attention in the field of NLP. Contrary to rule- or template-based domain-specific dialog systems, open-domain conversation usually requires data-driven approaches, which can be roughly divided into two categories: retrieval-based and generation-based systems. Retrieval systems search a user-issued utterance (called a query) in a large database, and return a reply that best matches the query. Generative approaches, typically based on recurrent neural networks (RNNs), can synthesize new replies, but they suffer from the problem of generating short, meaningless utterances. In this paper, we propose a novel ensemble of retrieval-based and generation-based dialog systems in the open domain. In our approach, the retrieved candidate, in addition to the original query, is fed to an RNN-based reply generator, so that the neural model is aware of more information. The generated reply is then fed back as a new candidate for post-reranking. Experimental results show that such ensemble outperforms each single part of it by a large margin.

In this paper, we propose a data structure, a quadruple neighbor list (QN-list, for short), to support real time queries of all longest increasing subsequence (LIS) and LIS with constraints over sequential data streams. The QN-List built by our algorithm requires $O(w)$ space, where $w$ is the time window size. The running time for building the initial QN-List takes $O(w\log w)$ time. Applying the QN-List, insertion of the new item takes $O(\log w)$ time and deletion of the first item takes $O(w)$ time. To the best of our knowledge, this is the first work to support both LIS enumeration and LIS with constraints computation by using a single uniform data structure for real time sequential data streams. Our method outperforms the state-of-the-art methods in both time and space cost, not only theoretically, but also empirically., Comment: 20 pages (12+8)

As the volume of the RDF data becomes increasingly large, it is essential for us to design a distributed database system to manage it. For distributed RDF data design, it is quite common to partition the RDF data into some parts, called fragments, which are then distributed. Thus, the distribution design consists of two steps: fragmentation and allocation. In this paper, we propose a method to explore the intrinsic similarities among the structures of queries in a workload for fragmentation and allocation, which aims to reduce the number of crossing matches and the communication cost during SPARQL query processing. Specifically, we mine and select some frequent access patterns to reflect the characteristics of the workload. Here, although we prove that selecting the optimal set of frequent access patterns is NP-hard, we propose a heuristic algorithm which guarantees both the data integrity and the approximation ratio. Based on the selected frequent access patterns, we propose two fragmentation strategies, vertical and horizontal fragmentation strategies, to divide RDF graphs while meeting different kinds of query processing objectives. Vertical fragmentation is for better throughput and horizontal fragmentation is for better performance. After fragmentation, we discuss how to allocate these fragments to various sites. Finally, we discuss how to process a query based on the results of fragmentation and allocation. Extensive experiments confirm the superior performance of our proposed solutions., Comment: 13 pages

Ship mathematical model and environmental disturbance model of a J-lay vessel are built, the model of pipe horizontal forces and moment acting on the ship is built by catenary theory. In order to avoid the ship's high frequency motion component disturbing the closed-loop control system, passive nonlinear observer is applied to estimate the vessel's low frequency variables during positioning process. Ship positioning operation controller is designed by the acceleration feedback and PID control strategy with pipeline action as feedforward term. Simulation results are used exclusively to demonstrate the observer and controller with established model for J-lay vessel.

In order to explore deep water oil and gas fields in South China Sea area, an automatic welding system applied in subsea pipeline laying has been developed, which has several features such as GMAW process, two welding carriages with dual torch, narrow gap and internal line up clamp fitted with copper backing rings. The welding carriage is specially designed for offshore application, so it is compact and lightweight. The control system of the automatic welding station is constructed on the CANOpen bus and Ethernet technologies, so the flexibility and reliability of the system are greatly improved, and the pipeline welding process also can be monitored remotely.

Replacing outdated productivity is an important goal of iron & steel industry during 11th five-year period. Based on the policies summary, this paper builds a mixed approach integrating bottom-up into top-down model by using MCP algorithm, and then designs the corresponding replaced rate. Results of simulation shows that: (1)Replacing outdated productivity of small iron & steel enterprises is remedied by super large iron & steel enterprises. (2)Replacing outdated productivity has less influence on whole economy, but great influence on resource-based industries. (3)Directly saving energy is more remarkable, but the total effect of policies saving energy less. According these, some suggestion is set up. Scale effect is not only important way to replace outdated productivity, but also the important means to promote enterprises saving energy.