Human dialogue contains evolving concepts, and speakers naturally associate multiple concepts to compose a response. However, current dialogue models with the seq2seq framework lack the ability to effectively manage concept transitions and can hardly
introduce multiple concepts to responses in a sequential decoding manner. To facilitate a controllable and coherent dialogue, in this work, we devise a concept-guided non-autoregressive model (CG-nAR) for open-domain dialogue generation. The proposed model comprises a multi-concept planning module that learns to identify multiple associated concepts from a concept graph and a customized Insertion Transformer that performs concept-guided non-autoregressive generation to complete a response. The experimental results on two public datasets show that CG-nAR can produce diverse and coherent responses, outperforming state-of-the-art baselines in both automatic and human evaluations with substantially faster inference speed.
With the rapid increase in the volume of dialogue data from daily life, there is a growing demand for dialogue summarization. Unfortunately, training a large summarization model is generally infeasible due to the inadequacy of dialogue data with anno
tated summaries. Most existing works for low-resource dialogue summarization directly pretrain models in other domains, e.g., the news domain, but they generally neglect the huge difference between dialogues and conventional articles. To bridge the gap between out-of-domain pretraining and in-domain fine-tuning, in this work, we propose a multi-source pretraining paradigm to better leverage the external summary data. Specifically, we exploit large-scale in-domain non-summary data to separately pretrain the dialogue encoder and the summary decoder. The combined encoder-decoder model is then pretrained on the out-of-domain summary data using adversarial critics, aiming to facilitate domain-agnostic summarization. The experimental results on two public datasets show that with only limited training data, our approach achieves competitive performance and generalizes well in different dialogue scenarios.
Regardless of their different types of progenitors and central engines, gamma-ray bursts (GRBs) were always assumed to be standalone systems after they formed. Little attention has been paid to the possibility that a stellar companion can still accom
pany a GRB itself. This paper investigates such a GRB-involved binary system and studies the effects of the stellar companion on the observed GRB emission when it is located inside the jet opening angle. Assuming a typical emission radius of $sim10^{15},$cm, we show that the blockage by a companion star with a radius of $R_mathrm{c}sim67,mathrm{R_odot}$ becomes non-negligible when it is located within a typical GRB jet opening angle (e.g., $sim10$ degrees) and beyond the GRB emission site. In such a case, an on-axis observer will see a GRB with a similar temporal behavior but 25% dimmer. On the other hand, an off-axis observer outside the jet opening angle (hence missed the original GRB) can see a delayed reflected GRB, which is much fainter in brightness, much wider in the temporal profile and slightly softer in energy. Our study can naturally explain the origin of some low-luminosity GRBs. Moreover, we also point out that the companion star may be shocked if it is located inside the GRB emission site, which can give rise to an X-ray transient or a GRB followed by a delayed X-ray bump on top of X-ray afterglows.
We propose HOI Transformer to tackle human object interaction (HOI) detection in an end-to-end manner. Current approaches either decouple HOI task into separated stages of object detection and interaction classification or introduce surrogate interac
tion problem. In contrast, our method, named HOI Transformer, streamlines the HOI pipeline by eliminating the need for many hand-designed components. HOI Transformer reasons about the relations of objects and humans from global image context and directly predicts HOI instances in parallel. A quintuple matching loss is introduced to force HOI predictions in a unified way. Our method is conceptually much simpler and demonstrates improved accuracy. Without bells and whistles, HOI Transformer achieves $26.61% $ $ AP $ on HICO-DET and $52.9%$ $AP_{role}$ on V-COCO, surpassing previous methods with the advantage of being much simpler. We hope our approach will serve as a simple and effective alternative for HOI tasks. Code is available at https://github.com/bbepoch/HoiTransformer .
In a customer service system, dialogue summarization can boost service efficiency by automatically creating summaries for long spoken dialogues in which customers and agents try to address issues about specific topics. In this work, we focus on topic
-oriented dialogue summarization, which generates highly abstractive summaries that preserve the main ideas from dialogues. In spoken dialogues, abundant dialogue noise and common semantics could obscure the underlying informative content, making the general topic modeling approaches difficult to apply. In addition, for customer service, role-specific information matters and is an indispensable part of a summary. To effectively perform topic modeling on dialogues and capture multi-role information, in this work we propose a novel topic-augmented two-stage dialogue summarizer (TDS) jointly with a saliency-aware neural topic model (SATM) for topic-oriented summarization of customer service dialogues. Comprehensive studies on a real-world Chinese customer service dataset demonstrated the superiority of our method against several strong baselines.
Automatic chat summarization can help people quickly grasp important information from numerous chat messages. Unlike conventional documents, chat logs usually have fragmented and evolving topics. In addition, these logs contain a quantity of elliptic
al and interrogative sentences, which make the chat summarization highly context dependent. In this work, we propose a novel unsupervised framework called RankAE to perform chat summarization without employing manually labeled data. RankAE consists of a topic-oriented ranking strategy that selects topic utterances according to centrality and diversity simultaneously, as well as a denoising auto-encoder that is carefully designed to generate succinct but context-informative summaries based on the selected utterances. To evaluate the proposed method, we collect a large-scale dataset of chat logs from a customer service environment and build an annotated set only for model evaluation. Experimental results show that RankAE significantly outperforms other unsupervised methods and is able to generate high-quality summaries in terms of relevance and topic coverage.
The booming online e-commerce platforms demand highly accurate approaches to segment queries that carry the product requirements of consumers. Recent works have shown that the supervised methods, especially those based on deep learning, are attractiv
e for achieving better performance on the problem of query segmentation. However, the lack of labeled data is still a big challenge for training a deep segmentation network, and the problem of Out-of-Vocabulary (OOV) also adversely impacts the performance of query segmentation. Different from query segmentation task in an open domain, e-commerce scenario can provide external documents that are closely related to these queries. Thus, to deal with the two challenges, we employ the idea of distant supervision and design a novel method to find contexts in external documents and extract features from these contexts. In this work, we propose a BiLSTM-CRF based model with an attention module to encode external features, such that external contexts information, which can be utilized naturally and effectively to help query segmentation. Experiments on two datasets show the effectiveness of our approach compared with several kinds of baselines.
This Letter, i.e. for the first time, proves that a general invariant velocity is originated from the principle of special relativity, namely, discovers the origin of the general invariant velocity, and when the general invariant velocity is taken as
the invariant light velocity in current theories, we get the corresponding special theory of relativity. Further, this Letter deduces triple special theories of relativity in cosmology, and cancels the invariant presumption of light velocity, it is proved that there exists a general constant velocity K determined by the experiments in cosmology, for K > 0, = 0 and < 0, they correspond to three kinds of possible relativistic theories in which the special theory of relativity is naturally contained for the special case of K > 0, and this Letter gives a prediction that, for K < 0, there is another likely case satisfying the principle of special relativity for some special physical systems in cosmology, in which the relativistic effects observed would be that the moving body would be lengthened, moving clock would be quickened. And the point of K = 0 is a bifurcation point, through which it gives out three types of possible universes in the cosmology (or multiverse). When a kind of matter with the maximally invariant velocity that may be superluminal or equal to light velocity is determined by experiments, then the invariant velocity can be taken as one of the general invariant velocity achieved in this Letter, then all results of current physical theories are consistent by utilizing this Letters theory.
