Representation Learning on Knowledge Graphs (KGs) is essential for downstream tasks. The dominant approach, KG Embedding (KGE), represents entities with independent vectors and faces the scalability challenge. Recent studies propose an alternative way for parameter efficiency, which represents entities by composing entity-corresponding codewords matched from predefined small-scale codebooks. We refer to the process of obtaining corresponding codewords of each entity as entity quantization, for which previous works have designed complicated strategies. Surprisingly, this paper shows that simple random entity quantization can achieve similar results to current strategies. We analyze this phenomenon and reveal that entity codes, the quantization outcomes for expressing entities, have higher entropy at the code level and Jaccard distance at the codeword level under random entity quantization. Therefore, different entities become more easily distinguished, facilitating effective KG representation. The above results show that current quantization strategies are not critical for KG representation, and there is still room for improvement in entity distinguishability beyond current strategies.

Dense retrieval is widely used for entity linking to retrieve entities from large-scale knowledge bases. Mainstream techniques are based on a dual-encoder framework, which encodes mentions and entities independently and calculates their relevances via rough interaction metrics, resulting in difficulty in explicitly modeling multiple mention-relevant parts within entities to match divergent mentions. Aiming at learning entity representations that can match divergent mentions, this paper proposes a Multi-View Enhanced Distillation (MVD) framework, which can effectively transfer knowledge of multiple fine-grained and mention-relevant parts within entities from cross-encoders to dual-encoders. Each entity is split into multiple views to avoid irrelevant information being over-squashed into the mention-relevant view. We further design cross-alignment and self-alignment mechanisms for this framework to facilitate fine-grained knowledge distillation from the teacher model to the student model. Meanwhile, we reserve a global-view that embeds the entity as a whole to prevent dispersal of uniform information. Experiments show our method achieves state-of-the-art performance on several entity linking benchmarks.

Federated Multilingual Neural Machine Translation (Fed-MNMT) has emerged as a promising paradigm for institutions with limited language resources. This approach allows multiple institutions to act as clients and train a unified model through model synchronization, rather than collecting sensitive data for centralized training. This significantly reduces the cost of corpus collection and preserves data privacy. However, as pre-trained language models (PLMs) continue to increase in size, the communication cost for transmitting parameters during synchronization has become a training speed bottleneck. In this paper, we propose a communication-efficient Fed-MNMT framework that addresses this issue by keeping PLMs frozen and only transferring lightweight adapter modules between clients. Since different language pairs exhibit substantial discrepancies in data distributions, adapter parameters of clients may conflict with each other. To tackle this, we explore various clustering strategies to group parameters for integration and mitigate the negative effects of conflicting parameters. Experimental results demonstrate that our framework reduces communication cost by over 98% while achieving similar or even better performance compared to competitive baselines. Further analysis reveals that clustering strategies effectively solve the problem of linguistic discrepancy and pruning adapter modules further improves communication efficiency.

Demo: https://youtu.be/WQLL93TPB-cAbstract:We present DeepGen, a system deployed at web scale for automatically creating sponsored search advertisements (ads) for BingAds customers. We leverage state-of-the-art natural language generation (NLG) models to generate fluent ads from advertiser’s web pages in an abstractive fashion and solve practical issues such as factuality and inference speed. In addition, our system creates a customized ad in real-time in response to the user’s search query, therefore highlighting different aspects of the same product based on what the user is looking for. To achieve this, our system generates a diverse choice of smaller pieces of the ad ahead of time and, at query time, selects the most relevant ones to be stitched into a complete ad. We improve generation diversity by training a controllable NLG model to generate multiple ads for the same web page highlighting different selling points. Our system design further improves diversity horizontally by first running an ensemble of generation models trained with different objectives and then using a diversity sampling algorithm to pick a diverse subset of generation results for online selection. Experimental results show the effectiveness of our proposed system design. Our system is currently deployed in production, serving ~4% of global ads served in Bing.

Transforming the large amounts of unstructured text on the Internet into structured event knowledge is a critical, yet unsolved goal of NLP, especially when addressing document-level text. Existing methods struggle in Document-level Event Extraction (DEE) due to its two intrinsic challenges: (a) Nested arguments, which means one argument is the sub-string of another one. (b) Multiple events, which indicates we should identify multiple events and assemble the arguments for them. In this paper, we propose a role-interactive multi-event head attention network (CLIO) to solve these two challenges jointly. The key idea is to map different events to multiple subspaces (i.e. multi-event head). In each event subspace, we draw the semantic representation of each role closer to its corresponding arguments, then we determine whether the current event exists. To further optimize event representation, we propose an event representation enhancing strategy to regularize pre-trained embedding space to be more isotropic. Our experiments on two widely used DEE datasets show that CLIO achieves consistent improvements over previous methods.

Sarcasm is important to sentiment analysis on social media. Sarcasm Target Identification (STI) deserves further study to understand sarcasm in depth. However, text lacking context or missing sarcasm target makes target identification very difficult. In this paper, we introduce multimodality to STI and present Multimodal Sarcasm Target Identification (MSTI) task. We propose a novel multi-scale cross-modality model that can simultaneously perform textual target labeling and visual target detection. In the model, we extract multi-scale visual features to enrich spatial information for different sized visual sarcasm targets. We design a set of convolution networks to unify multi-scale visual features with textual features for cross-modal attention learning, and correspondingly a set of transposed convolution networks to restore multi-scale visual information. The results show that visual clues can improve the performance of TSTI by a large margin, and VSTI achieves good accuracy.

Copy mechanisms explicitly obtain unchanged tokens from the source (input) sequence to generate the target (output) sequence under the neural seq2seq framework. However, most of the existing copy mechanisms only consider single word copying from the source sentences, which results in losing essential tokens while copying long spans. In this work, we propose a plug-and-play architecture, namely BioCopy, to alleviate the problem aforementioned. Specifically, in the training stage, we construct a BIO tag for each token and train the original model with BIO tags jointly. In the inference stage, the model will firstly predict the BIO tag at each time step, then conduct different mask strategies based on the predicted BIO label to diminish the scope of the probability distributions over the vocabulary list. Experimental results on two separate generative tasks show that they all outperform the baseline models by adding our BioCopy to the original model structure.

In recent years, a variety of recurrent neural networks have been proposed, e.g LSTM. However, existing models only read the text once, it cannot describe the situation of repeated reading in reading comprehension. In fact, when reading or analyzing a text, we may read the text several times rather than once if we couldn’t well understand it. So, how to model this kind of the reading behavior? To address the issue, we propose a multi-glance mechanism (MGM) for modeling the habit of reading behavior. In the proposed framework, the actual reading process can be fully simulated, and then the obtained information can be consistent with the task. Based on the multi-glance mechanism, we design two types of recurrent neural network models for repeated reading: Glance Cell Model (GCM) and Glance Gate Model (GGM). Visualization analysis of the GCM and the GGM demonstrates the effectiveness of multi-glance mechanisms. Experiments results on the large-scale datasets show that the proposed methods can achieve better performance.

Human generates responses relying on semantic and functional dependencies, including coreference relation, among dialogue elements and their context. In this paper, we investigate matching a response with its multi-turn context using dependency information based entirely on attention. Our solution is inspired by the recently proposed Transformer in machine translation (Vaswani et al., 2017) and we extend the attention mechanism in two ways. First, we construct representations of text segments at different granularities solely with stacked self-attention. Second, we try to extract the truly matched segment pairs with attention across the context and response. We jointly introduce those two kinds of attention in one uniform neural network. Experiments on two large-scale multi-turn response selection tasks show that our proposed model significantly outperforms the state-of-the-art models.

Deep learning approaches for sentiment classification do not fully exploit sentiment linguistic knowledge. In this paper, we propose a Multi-sentiment-resource Enhanced Attention Network (MEAN) to alleviate the problem by integrating three kinds of sentiment linguistic knowledge (e.g., sentiment lexicon, negation words, intensity words) into the deep neural network via attention mechanisms. By using various types of sentiment resources, MEAN utilizes sentiment-relevant information from different representation sub-spaces, which makes it more effective to capture the overall semantics of the sentiment, negation and intensity words for sentiment prediction. The experimental results demonstrate that MEAN has robust superiority over strong competitors.

In this paper, we present the design, collection, transcription and analysis of a Mandarin Chinese Broadcast Collection of over 3000 hours. The data was collected by Hong Kong University of Science and Technology (HKUST) in China on a cable TV and satellite transmission platform established in support of the DARPA Global Autonomous Language Exploitation (GALE) program. The collection includes broadcast news (BN) and broadcast conversation (BC) including talk shows, roundtable discussions, call-in shows, editorials and other conversational programs that focus on news and current events. HKUST also collects detailed information about all recorded programs. A subset of BC and BN recordings are manually transcribed with standard Chinese characters in UTF-8 encoding, using specific mark-ups for a small set of spontaneous and conversational speech phenomena. The collection is among the largest and first of its kind for Mandarin Chinese Broadcast speech, providing abundant and diverse samples for Mandarin speech recognition and other application-dependent tasks, such as spontaneous speech processing and recognition, topic detection, information retrieval, and speaker recognition. HKUST’s acoustic analysis of 500 hours of the speech and transcripts demonstrates the positive impact this data could have on system performance.