arXiv subjects

Yue Zhang

Publications and source records attributed to Yue Zhang.

At least 577 records · Page 32Linked to original sources

arXiv2022

Duality Regularization for Unsupervised Bilingual Lexicon Induction

Unsupervised bilingual lexicon induction naturally exhibits duality, which results from symmetry in back-translation. For example, EN-IT and IT-EN induction can be mutually primal and dual problems. Current state-of-the-art methods, however, consider the two tasks independently. In this paper, we propose to train primal and dual models jointly, using regularizers to encourage consistency in back translation cycles. Experiments across 6 language pairs show that the proposed method significantly outperforms competitive baselines, obtaining the best-published results on a standard benchmark.

cs.CL↗

arXiv2022

Exploiting Sentiment and Common Sense for Zero-shot Stance Detection

The stance detection task aims to classify the stance toward given documents and topics. Since the topics can be implicit in documents and unseen in training data for zero-shot settings, we propose to boost the transferability of the stance detection model by using sentiment and commonsense knowledge, which are seldom considered in previous studies. Our model includes a graph autoencoder module to obtain commonsense knowledge and a stance detection module with sentiment and commonsense. Experimental results show that our model outperforms the state-of-the-art methods on the zero-shot and few-shot benchmark dataset--VAST. Meanwhile, ablation studies prove the significance of each module in our model. Analysis of the relations between sentiment, common sense, and stance indicates the effectiveness of sentiment and common sense.

cs.CL↗

arXiv2022

Human-in-the-loop Robotic Grasping using BERT Scene Representation

Current NLP techniques have been greatly applied in different domains. In this paper, we propose a human-in-the-loop framework for robotic grasping in cluttered scenes, investigating a language interface to the grasping process, which allows the user to intervene by natural language commands. This framework is constructed on a state-of-the-art rasping baseline, where we substitute a scene-graph representation with a text representation of the scene using BERT. Experiments on both simulation and physical robot show that the proposed method outperforms conventional object-agnostic and scene-graph based methods in the literature. In addition, we find that with human intervention, performance can be significantly improved.

cs.RO↗

arXiv2022

The Axion Quality and Fifth Force

We investigate what it takes for the axion to address the strong CP problem in the presence of explicit Peccei-Quinn (PQ) symmetry breaking effects besides the strong interaction. In cases where the PQ-Higgsing scalar field directly couples to the Standard Model sector, it is pointed out that existing fifth force experiments can set better constraints on the axion quality over neutron electric dipole moment.

hep-ph↗

arXiv2022

LOViS: Learning Orientation and Visual Signals for Vision and Language Navigation

Understanding spatial and visual information is essential for a navigation agent who follows natural language instructions. The current Transformer-based VLN agents entangle the orientation and vision information, which limits the gain from the learning of each information source. In this paper, we design a neural agent with explicit Orientation and Vision modules. Those modules learn to ground spatial information and landmark mentions in the instructions to the visual environment more effectively. To strengthen the spatial reasoning and visual perception of the agent, we design specific pre-training tasks to feed and better utilize the corresponding modules in our final navigation model. We evaluate our approach on both Room2room (R2R) and Room4room (R4R) datasets and achieve the state of the art results on both benchmarks.

cs.CV↗

arXiv2022

Semantic-based Pre-training for Dialogue Understanding

Pre-trained language models have made great progress on dialogue tasks. However, these models are typically trained on surface dialogue text, thus are proven to be weak in understanding the main semantic meaning of a dialogue context. We investigate Abstract Meaning Representation (AMR) as explicit semantic knowledge for pre-training models to capture the core semantic information in dialogues during pre-training. In particular, we propose a semantic-based pre-training framework that extends the standard pre-training framework (Devlin et al., 2019) by three tasks for learning 1) core semantic units, 2) semantic relations and 3) the overall semantic representation according to AMR graphs. Experiments on the understanding of both chit-chats and task-oriented dialogues show the superiority of our model. To our knowledge, we are the first to leverage a deep semantic representation for dialogue pre-training.

cs.CL↗

arXiv2022

Can Offline Reinforcement Learning Help Natural Language Understanding?

Pre-training has been a useful method for learning implicit transferable knowledge and it shows the benefit of offering complementary features across different modalities. Recent work mainly focuses on the modalities such as image and text, for example, studies show that visual features learned from images can help visual-grounded language understanding. In this paper, we consider investigating the potential connection between offline reinforcement learning (RL) and language modeling (LM). Intuitively, RL and LM are similar in predicting the next states based on the current and previous states, which rely on both local and long-range dependency across states. To validate such an assumption, we pre-trained different offline RL tasks using Transformer and then evaluate these models on various language-related tasks. Experimental results show that our RL pre-trained models can give close performance compared with the models using the LM training objective, showing that there exist common useful features across these two modalities. To further explore the potential relationship, we investigate some factors such as Markov property and the sequential nature of RL trajectory.

cs.CL↗

arXiv2022

DialogSum Challenge: Results of the Dialogue Summarization Shared Task

We report the results of DialogSum Challenge, the shared task on summarizing real-life scenario dialogues at INLG 2022. Four teams participate in this shared task and three submit their system reports, exploring different methods to improve the performance of dialogue summarization. Although there is a great improvement over the baseline models regarding automatic evaluation metrics, such as Rouge scores, we find that there is a salient gap between model generated outputs and human annotated summaries by human evaluation from multiple aspects. These findings demonstrate the difficulty of dialogue summarization and suggest that more fine-grained evaluatuion metrics are in need.

cs.CL↗

arXiv2022

Recent Advances in Text-to-SQL: A Survey of What We Have and What We Expect

Text-to-SQL has attracted attention from both the natural language processing and database communities because of its ability to convert the semantics in natural language into SQL queries and its practical application in building natural language interfaces to database systems. The major challenges in text-to-SQL lie in encoding the meaning of natural utterances, decoding to SQL queries, and translating the semantics between these two forms. These challenges have been addressed to different extents by the recent advances. However, there is still a lack of comprehensive surveys for this task. To this end, we review recent progress on text-to-SQL for datasets, methods, and evaluation and provide this systematic survey, addressing the aforementioned challenges and discussing potential future directions. We hope that this survey can serve as quick access to existing work and motivate future research.

cs.CL↗

arXiv2022

Lost in Context? On the Sense-wise Variance of Contextualized Word Embeddings

Contextualized word embeddings in language models have given much advance to NLP. Intuitively, sentential information is integrated into the representation of words, which can help model polysemy. However, context sensitivity also leads to the variance of representations, which may break the semantic consistency for synonyms. We quantify how much the contextualized embeddings of each word sense vary across contexts in typical pre-trained models. Results show that contextualized embeddings can be highly consistent across contexts. In addition, part-of-speech, number of word senses, and sentence length have an influence on the variance of sense representations. Interestingly, we find that word representations are position-biased, where the first words in different contexts tend to be more similar. We analyze such a phenomenon and also propose a simple way to alleviate such bias in distance-based word sense disambiguation settings.

cs.CL↗

arXiv2022

Mere Contrastive Learning for Cross-Domain Sentiment Analysis

Cross-domain sentiment analysis aims to predict the sentiment of texts in the target domain using the model trained on the source domain to cope with the scarcity of labeled data. Previous studies are mostly cross-entropy-based methods for the task, which suffer from instability and poor generalization. In this paper, we explore contrastive learning on the cross-domain sentiment analysis task. We propose a modified contrastive objective with in-batch negative samples so that the sentence representations from the same class will be pushed close while those from the different classes become further apart in the latent space. Experiments on two widely used datasets show that our model can achieve state-of-the-art performance in both cross-domain and multi-domain sentiment analysis tasks. Meanwhile, visualizations demonstrate the effectiveness of transferring knowledge learned in the source domain to the target domain and the adversarial test verifies the robustness of our model.

cs.CL↗

arXiv2022

Exploiting Unlabeled Data for Target-Oriented Opinion Words Extraction

Target-oriented Opinion Words Extraction (TOWE) is a fine-grained sentiment analysis task that aims to extract the corresponding opinion words of a given opinion target from the sentence. Recently, deep learning approaches have made remarkable progress on this task. Nevertheless, the TOWE task still suffers from the scarcity of training data due to the expensive data annotation process. Limited labeled data increase the risk of distribution shift between test data and training data. In this paper, we propose exploiting massive unlabeled data to reduce the risk by increasing the exposure of the model to varying distribution shifts. Specifically, we propose a novel Multi-Grained Consistency Regularization (MGCR) method to make use of unlabeled data and design two filters specifically for TOWE to filter noisy data at different granularity. Extensive experimental results on four TOWE benchmark datasets indicate the superiority of MGCR compared with current state-of-the-art methods. The in-depth analysis also demonstrates the effectiveness of the different-granularity filters. Our codes are available at https://github.com/TOWESSL/TOWESSL.

cs.CL↗

arXiv2022

Dual Domain-Adversarial Learning for Audio-Visual Saliency Prediction

Both visual and auditory information are valuable to determine the salient regions in videos. Deep convolution neural networks (CNN) showcase strong capacity in coping with the audio-visual saliency prediction task. Due to various factors such as shooting scenes and weather, there often exists moderate distribution discrepancy between source training data and target testing data. The domain discrepancy induces to performance degradation on target testing data for CNN models. This paper makes an early attempt to tackle the unsupervised domain adaptation problem for audio-visual saliency prediction. We propose a dual domain-adversarial learning algorithm to mitigate the domain discrepancy between source and target data. First, a specific domain discrimination branch is built up for aligning the auditory feature distributions. Then, those auditory features are fused into the visual features through a cross-modal self-attention module. The other domain discrimination branch is devised to reduce the domain discrepancy of visual features and audio-visual correlations implied by the fused audio-visual features. Experiments on public benchmarks demonstrate that our method can relieve the performance degradation caused by domain discrepancy.

cs.CV↗

arXiv2022

Core-collapse Supernova Constraint on the Origin of Sterile Neutrino Dark Matter via Neutrino Self-interactions

Novel neutrino self-interaction can open up viable parameter space for the relic abundance of sterile-neutrino dark matter (S$ν$DM). In this work, we constrain the relic target using core-collapse supernova which features the same fundamental process and a similar environment to the early universe era when S$ν$DM is dominantly produced. We present a detailed calculation of the effects of a massive scalar mediated neutrino self-interaction on the supernova cooling rate, including the derivation of the thermal potential in the presence of non-zero chemical potentials from plasma species. Our results demonstrate that the supernova cooling argument can cover the neutrino self-interaction parameter space that complements terrestrial and cosmological probes.

hep-ph↗

arXiv2022

Modeling mandatory and discretionary lane changes using dynamic interaction networks

A quantitative understanding of dynamic lane-changing (LC) interaction patterns is indispensable for improving the decision-making of autonomous vehicles, especially in mixed traffic with human-driven vehicles. This paper develops a novel framework combining the hidden Markov model and graph structure to identify the difference in dynamic interaction networks between mandatory lane changes (MLC) and discretionary lane changes (DLC). A hidden Markov model is developed to decompose LC interactions into homogenous segments and reveal the temporal properties of these segments. Then, conditional mutual information is used to quantify the interaction intensity, and the graph structure is used to characterize the connectivity between vehicles. Finally, the critical vehicle in each dynamic interaction network is identified. Based on the LC events extracted from the INTERACTION dataset, the proposed analytical framework is applied to modeling MLC and DLC under congested traffic with levels of service E and F. The results show that there are multiple heterogeneous dynamic interaction network structures in an LC process. A comparison of MLC and DLC demonstrates that MLC are more complex, while DLC are more random. The complexity of MLC is attributed to the intense interaction and frequent transition of the interaction network structure, while the random DLC demonstrate no obvious evolution rules and dominant vehicles in interaction networks. The findings in this study are useful for understanding the connectivity structure between vehicles in LC interactions, and for designing appropriate and well-directed driving decision-making models for autonomous vehicles and advanced driver-assistance systems.

eess.SY↗

arXiv2022

Dodelson-Widrow Mechanism In the Presence of Self-Interacting Neutrinos

keV-scale gauge-singlet fermions, allowed to mix with the active neutrinos, are elegant dark matter (DM) candidates. They are produced in the early universe via the Dodelson-Widrow mechanism and can be detected as they decay very slowly, emitting X-rays. In the absence of new physics, this hypothesis is virtually ruled out by astrophysical observations. Here, we show that new interactions among the active neutrinos allow these sterile neutrinos to make up all the DM while safely evading all current experimental bounds. The existence of these new neutrino interactions may manifest itself in next-generation experiments, including DUNE.

hep-ph↗

arXiv2022

A General Contextualized Rewriting Framework for Text Summarization

The rewriting method for text summarization combines extractive and abstractive approaches, improving the conciseness and readability of extractive summaries using an abstractive model. Exiting rewriting systems take each extractive sentence as the only input, which is relatively focused but can lose necessary background knowledge and discourse context. In this paper, we investigate contextualized rewriting, which consumes the entire document and considers the summary context. We formalize contextualized rewriting as a seq2seq with group-tag alignments, introducing group-tag as a solution to model the alignments, identifying extractive sentences through content-based addressing. Results show that our approach significantly outperforms non-contextualized rewriting systems without requiring reinforcement learning, achieving strong improvements on ROUGE scores upon multiple extractors.

cs.CL↗

arXiv2022

Chaotic spin precession in antiferromagnetic domain walls

In contrast with rich investigations about the translation of an antiferromagnetic (AFM) texture, spin precession in an AFM texture is seldom concerned for lacking an effective driving method. In this work, however, we show that under an alternating spin-polarized current with spin along the AFM anisotropy axis, spin precession can be excited in an AFM DW. Especially, chaotic spin precession occurs at moderate interfacial Dzyaloshinskii-Moriya interaction (DMI), which contributes to a nonlinear term in the dynamic equation of DW precession. Also, crisis-induced intermittent chaos appears when the current density is higher than a critical value. This work not only paves a way to unravel rich spin precession behaviors in an AFM texture but also provides guidelines for developing ultrafast spintronic devices based on new physical principles.

cond-mat.mtrl-sci↗