arXiv subjects

Xiaodong He

Publications and source records attributed to Xiaodong He.

At least 55 records · Page 3Linked to original sources

arXiv2023

SegCLIP: Patch Aggregation with Learnable Centers for Open-Vocabulary Semantic Segmentation

Recently, the contrastive language-image pre-training, e.g., CLIP, has demonstrated promising results on various downstream tasks. The pre-trained model can capture enriched visual concepts for images by learning from a large scale of text-image data. However, transferring the learned visual knowledge to open-vocabulary semantic segmentation is still under-explored. In this paper, we propose a CLIP-based model named SegCLIP for the topic of open-vocabulary segmentation in an annotation-free manner. The SegCLIP achieves segmentation based on ViT and the main idea is to gather patches with learnable centers to semantic regions through training on text-image pairs. The gathering operation can dynamically capture the semantic groups, which can be used to generate the final segmentation results. We further propose a reconstruction loss on masked patches and a superpixel-based KL loss with pseudo-labels to enhance the visual representation. Experimental results show that our model achieves comparable or superior segmentation accuracy on the PASCAL VOC 2012 (+0.3% mIoU), PASCAL Context (+2.3% mIoU), and COCO (+2.2% mIoU) compared with baselines. We release the code at https://github.com/ArrowLuo/SegCLIP.

cs.CV↗

arXiv2023

MoNET: Tackle State Momentum via Noise-Enhanced Training for Dialogue State Tracking

Dialogue state tracking (DST) aims to convert the dialogue history into dialogue states which consist of slot-value pairs. As condensed structural information memorizing all history information, the dialogue state in the last turn is typically adopted as the input for predicting the current state by DST models. However, these models tend to keep the predicted slot values unchanged, which is defined as state momentum in this paper. Specifically, the models struggle to update slot values that need to be changed and correct wrongly predicted slot values in the last turn. To this end, we propose MoNET to tackle state momentum via noise-enhanced training. First, the previous state of each turn in the training data is noised via replacing some of its slot values. Then, the noised previous state is used as the input to learn to predict the current state, improving the model's ability to update and correct slot values. Furthermore, a contrastive context matching framework is designed to narrow the representation distance between a state and its corresponding noised variant, which reduces the impact of noised state and makes the model better understand the dialogue history. Experimental results on MultiWOZ datasets show that MoNET outperforms previous DST methods. Ablations and analysis verify the effectiveness of MoNET in alleviating state momentum and improving anti-noise ability.

cs.CL↗

arXiv2023

AUGUST: an Automatic Generation Understudy for Synthesizing Conversational Recommendation Datasets

High-quality data is essential for conversational recommendation systems and serves as the cornerstone of the network architecture development and training strategy design. Existing works contribute heavy human efforts to manually labeling or designing and extending recommender dialogue templates. However, they suffer from (i) the limited number of human annotators results in that datasets can hardly capture rich and large-scale cases in the real world, (ii) the limited experience and knowledge of annotators account for the uninformative corpus and inappropriate recommendations. In this paper, we propose a novel automatic dataset synthesis approach that can generate both large-scale and high-quality recommendation dialogues through a data2text generation process, where unstructured recommendation conversations are generated from structured graphs based on user-item information from the real world. In doing so, we comprehensively exploit: (i) rich personalized user profiles from traditional recommendation datasets, (ii) rich external knowledge from knowledge graphs, and (iii) the conversation ability contained in human-to-human conversational recommendation datasets. Extensive experiments validate the benefit brought by the automatically synthesized data under low-resource scenarios and demonstrate the promising potential to facilitate the development of a more effective conversational recommendation system.

cs.CL↗

arXiv2023

OTF: Optimal Transport based Fusion of Supervised and Self-Supervised Learning Models for Automatic Speech Recognition

Self-Supervised Learning (SSL) Automatic Speech Recognition (ASR) models have shown great promise over Supervised Learning (SL) ones in low-resource settings. However, the advantages of SSL are gradually weakened when the amount of labeled data increases in many industrial applications. To further improve the ASR performance when abundant labels are available, we first explore the potential of combining SL and SSL ASR models via analyzing their complementarity in recognition accuracy and optimization property. Then, we propose a novel Optimal Transport based Fusion (OTF) method for SL and SSL models without incurring extra computation cost in inference. Specifically, optimal transport is adopted to softly align the layer-wise weights to unify the two different networks into a single one. Experimental results on the public 1k-hour English LibriSpeech dataset and our in-house 2.6k-hour Chinese dataset show that OTF largely outperforms the individual models with lower error rates.

eess.AS↗

arXiv2023

MaskedSpeech: Context-aware Speech Synthesis with Masking Strategy

Humans often speak in a continuous manner which leads to coherent and consistent prosody properties across neighboring utterances. However, most state-of-the-art speech synthesis systems only consider the information within each sentence and ignore the contextual semantic and acoustic features. This makes it inadequate to generate high-quality paragraph-level speech which requires high expressiveness and naturalness. To synthesize natural and expressive speech for a paragraph, a context-aware speech synthesis system named MaskedSpeech is proposed in this paper, which considers both contextual semantic and acoustic features. Inspired by the masking strategy in the speech editing research, the acoustic features of the current sentence are masked out and concatenated with those of contextual speech, and further used as additional model input. The phoneme encoder takes the concatenated phoneme sequence from neighboring sentences as input and learns fine-grained semantic information from contextual text. Furthermore, cross-utterance coarse-grained semantic features are employed to improve the prosody generation. The model is trained to reconstruct the masked acoustic features with the augmentation of both the contextual semantic and acoustic features. Experimental results demonstrate that the proposed MaskedSpeech outperformed the baseline system significantly in terms of naturalness and expressiveness.

cs.SD↗

arXiv2023

Learning to Generate Poetic Chinese Landscape Painting with Calligraphy

In this paper, we present a novel system (denoted as Polaca) to generate poetic Chinese landscape painting with calligraphy. Unlike previous single image-to-image painting generation, Polaca takes the classic poetry as input and outputs the artistic landscape painting image with the corresponding calligraphy. It is equipped with three different modules to complete the whole piece of landscape painting artwork: the first one is a text-to-image module to generate landscape painting image, the second one is an image-to-image module to generate stylistic calligraphy image, and the third one is an image fusion module to fuse the two images into a whole piece of aesthetic artwork.

cs.CV↗

arXiv2023

A Novel Vector-Field-Based Motion Planning Algorithm for 3D Nonholonomic Robots

This paper focuses on the motion planning for mobile robots in 3D, which are modelled by 6-DOF rigid body systems with nonholonomic kinematics constraints. We not only specify the target position, but also bring in the requirement of the heading direction at the terminal time, which gives rise to a new and more challenging 3D motion planning problem. The proposed planning algorithm involves a novel velocity vector field (VF) over the workspace, and by following the VF, the robot can be navigated to the destination with the specified heading direction. In order to circumvent potential collisions with obstacles and other robots, a composite VF is designed by composing the navigation VF and an additional VF tangential to the boundary of the dangerous area. Moreover, we propose a priority-based algorithm to deal with the motion coupling issue among multiple robots. Finally, numerical simulations are conducted to verify the theoretical results.

cs.RO↗

arXiv2023

UFO2: A unified pre-training framework for online and offline speech recognition

In this paper, we propose a Unified pre-training Framework for Online and Offline (UFO2) Automatic Speech Recognition (ASR), which 1) simplifies the two separate training workflows for online and offline modes into one process, and 2) improves the Word Error Rate (WER) performance with limited utterance annotating. Specifically, we extend the conventional offline-mode Self-Supervised Learning (SSL)-based ASR approach to a unified manner, where the model training is conditioned on both the full-context and dynamic-chunked inputs. To enhance the pre-trained representation model, stop-gradient operation is applied to decouple the online-mode objectives to the quantizer. Moreover, in both the pre-training and the downstream fine-tuning stages, joint losses are proposed to train the unified model with full-weight sharing for the two modes. Experimental results on the LibriSpeech dataset show that UFO2 outperforms the SSL-based baseline method by 29.7% and 18.2% relative WER reduction in offline and online modes, respectively.

eess.AS↗

arXiv2023

Roto-Translation Invariant Formation of Fixed-Wing UAVs in 3D: Feasibility and Control

This paper investigates the formation of fixed-wing UAVs in 3D, which communicate via a directed acyclic graph. Different from common formation problems, we consider the roto-translation invariant (RTI) formation, where the "roto-translation" refers to a rigid-body motion obtained by composing rotation and translation. Besides, the fixed-wing UAV is modelled by a 3-D nonholonomic constrained rigid body instead of a particle agent. The main results of this paper include proposing the formation feasibility conditions and designing the formation controller. Firstly, we define the RTI formation and propose the conditions to guarantee that the formations are feasible for the fixed-wing UAVs under the nonholonomic and input saturation constraints. Secondly, given feasible formations, we design a formation controller by introducing a virtual leader and employing the compensation of rotation, followed by proving the stability of the closed-loop system. Finally, simulation examples are presented to verify the theoretical results.al with the motion coupling among multiple robots. Finally, numerical simulations are conducted to verify the theoretical results.

eess.SY↗

arXiv2022

MNER-QG: An End-to-End MRC framework for Multimodal Named Entity Recognition with Query Grounding

Multimodal named entity recognition (MNER) is a critical step in information extraction, which aims to detect entity spans and classify them to corresponding entity types given a sentence-image pair. Existing methods either (1) obtain named entities with coarse-grained visual clues from attention mechanisms, or (2) first detect fine-grained visual regions with toolkits and then recognize named entities. However, they suffer from improper alignment between entity types and visual regions or error propagation in the two-stage manner, which finally imports irrelevant visual information into texts. In this paper, we propose a novel end-to-end framework named MNER-QG that can simultaneously perform MRC-based multimodal named entity recognition and query grounding. Specifically, with the assistance of queries, MNER-QG can provide prior knowledge of entity types and visual regions, and further enhance representations of both texts and images. To conduct the query grounding task, we provide manual annotations and weak supervisions that are obtained via training a highly flexible visual grounding model with transfer learning. We conduct extensive experiments on two public MNER datasets, Twitter2015 and Twitter2017. Experimental results show that MNER-QG outperforms the current state-of-the-art models on the MNER task, and also improves the query grounding performance.

cs.CV↗

arXiv2022

Multi-Speaker Multi-Style Speech Synthesis with Timbre and Style Disentanglement

Disentanglement of a speaker's timbre and style is very important for style transfer in multi-speaker multi-style text-to-speech (TTS) scenarios. With the disentanglement of timbres and styles, TTS systems could synthesize expressive speech for a given speaker with any style which has been seen in the training corpus. However, there are still some shortcomings with the current research on timbre and style disentanglement. The current method either requires single-speaker multi-style recordings, which are difficult and expensive to collect, or uses a complex network and complicated training method, which is difficult to reproduce and control the style transfer behavior. To improve the disentanglement effectiveness of timbres and styles, and to remove the reliance on single-speaker multi-style corpus, a simple but effective timbre and style disentanglement method is proposed in this paper. The FastSpeech2 network is employed as the backbone network, with explicit duration, pitch, and energy trajectory to represent the style. Each speaker's data is considered as a separate and isolated style, then a speaker embedding and a style embedding are added to the FastSpeech2 network to learn disentangled representations. Utterance level pitch and energy normalization are utilized to improve the decoupling effect. Experimental results demonstrate that the proposed model could synthesize speech with any style seen during training with high style similarity while maintaining very high speaker similarity.

cs.SD↗

arXiv2022

P$^3$LM: Probabilistically Permuted Prophet Language Modeling for Generative Pre-Training

Conventional autoregressive left-to-right (L2R) sequence generation faces two issues during decoding: limited to unidirectional target sequence modeling, and constrained on strong local dependencies. To address the aforementioned problem, we propose P$^3$LM, a probabilistically permuted prophet language model, which strengthens the modeling of bidirectional information and long token dependencies for sequence generation. Specifically, P$^3$LM learns to generate tokens in permuted order upon an order-aware transformer decoder, as well as to generate the corresponding future $N$ tokens with a multi-stream attention mechanism. Extensive experiments are conducted on the GLGE benchmark, which includes four datasets for summarization, two for question generation, one for conversational question answering, and one for dialog response generation, where P$^3$LM achieves state-of-the-art results compared with strong publicly available generative pre-training methods.

cs.CL↗

arXiv2022

MuGER$^2$: Multi-Granularity Evidence Retrieval and Reasoning for Hybrid Question Answering

Hybrid question answering (HQA) aims to answer questions over heterogeneous data, including tables and passages linked to table cells. The heterogeneous data can provide different granularity evidence to HQA models, e.t., column, row, cell, and link. Conventional HQA models usually retrieve coarse- or fine-grained evidence to reason the answer. Through comparison, we find that coarse-grained evidence is easier to retrieve but contributes less to the reasoner, while fine-grained evidence is the opposite. To preserve the advantage and eliminate the disadvantage of different granularity evidence, we propose MuGER$^2$, a Multi-Granularity Evidence Retrieval and Reasoning approach. In evidence retrieval, a unified retriever is designed to learn the multi-granularity evidence from the heterogeneous data. In answer reasoning, an evidence selector is proposed to navigate the fine-grained evidence for the answer reader based on the learned multi-granularity evidence. Experiment results on the HybridQA dataset show that MuGER$^2$ significantly boosts the HQA performance. Further ablation analysis verifies the effectiveness of both the retrieval and reasoning designs.

cs.CL↗

arXiv2022

UniRPG: Unified Discrete Reasoning over Table and Text as Program Generation

Question answering requiring discrete reasoning, e.g., arithmetic computing, comparison, and counting, over knowledge is a challenging task. In this paper, we propose UniRPG, a semantic-parsing-based approach advanced in interpretability and scalability, to perform unified discrete reasoning over heterogeneous knowledge resources, i.e., table and text, as program generation. Concretely, UniRPG consists of a neural programmer and a symbolic program executor, where a program is the composition of a set of pre-defined general atomic and higher-order operations and arguments extracted from table and text. First, the programmer parses a question into a program by generating operations and copying arguments, and then the executor derives answers from table and text based on the program. To alleviate the costly program annotation issue, we design a distant supervision approach for programmer learning, where pseudo programs are automatically constructed without annotated derivations. Extensive experiments on the TAT-QA dataset show that UniRPG achieves tremendous improvements and enhances interpretability and scalability compared with state-of-the-art methods, even without derivation annotation. Moreover, it achieves promising performance on the textual dataset DROP without derivations.

cs.CL↗

arXiv2022

Simultaneous Position and Orientation Planning of Nonholonomic Multi-Robot Systems: A Dynamic Vector Field Approach

This paper considers the simultaneous position and orientation planning of nonholonomic multirobot systems. Different from common researches which only focus on final position constraints, we model the nonholonomic mobile robot as a rigid body and introduce the orientation as well as position constraints for the robot's final states. In other words, robots should not only reach the specified positions, but also point to the desired orientations simultaneously. The challenge of this problem lies in the underactuation of full-state motion planning, since three states need to be planned by mere two control inputs. To this end, we propose a dynamic vector field (DVF) based on the rigid body modeling. Specifically, the dynamics of the robot orientation are brought into the vector field, implying that the vector field is not static on the 2-D plane anymore, but a dynamic one varying with the attitude angle. Hence, each robot can move along the integral curve of the DVF to arrive at the desired position, and in the meantime, the attitude angle can converge to the specified value following the orientation dynamics. Subsequently, by designing a circular vector field under the framework of the DVF, we further study the obstacle avoidance and mutual-robot-collision avoidance in the motion planning. Finally, numerical simulation examples are provided to verify the effectiveness of the proposed methodology.

cs.RO↗

arXiv2022

AutoQGS: Auto-Prompt for Low-Resource Knowledge-based Question Generation from SPARQL

This study investigates the task of knowledge-based question generation (KBQG). Conventional KBQG works generated questions from fact triples in the knowledge graph, which could not express complex operations like aggregation and comparison in SPARQL. Moreover, due to the costly annotation of large-scale SPARQL-question pairs, KBQG from SPARQL under low-resource scenarios urgently needs to be explored. Recently, since the generative pre-trained language models (PLMs) typically trained in natural language (NL)-to-NL paradigm have been proven effective for low-resource generation, e.g., T5 and BART, how to effectively utilize them to generate NL-question from non-NL SPARQL is challenging. To address these challenges, AutoQGS, an auto-prompt approach for low-resource KBQG from SPARQL, is proposed. Firstly, we put forward to generate questions directly from SPARQL for the KBQG task to handle complex operations. Secondly, we propose an auto-prompter trained on large-scale unsupervised data to rephrase SPARQL into NL description, smoothing the low-resource transformation from non-NL SPARQL to NL question with PLMs. Experimental results on the WebQuestionsSP, ComlexWebQuestions 1.1, and PathQuestions show that our model achieves state-of-the-art performance, especially in low-resource settings. Furthermore, a corpus of 330k factoid complex question-SPARQL pairs is generated for further KBQG research.

cs.CL↗

arXiv2022

Flow Completion Network: Inferring the Fluid Dynamics from Incomplete Flow Information using Graph Neural Networks

This paper introduces a novel neural network - flow completion network (FCN) - to infer the fluid dynamics, includ-ing the flow field and the force acting on the body, from the incomplete data based on Graph Convolution AttentionNetwork. The FCN is composed of several graph convolution layers and spatial attention layers. It is designed to inferthe velocity field and the vortex force contribution of the flow field when combined with the vortex force map (VFM)method. Compared with other neural networks adopted in fluid dynamics, the FCN is capable of dealing with bothstructured data and unstructured data. The performance of the proposed FCN is assessed by the computational fluiddynamics (CFD) data on the flow field around a circular cylinder. The force coefficients predicted by our model arevalidated against those obtained directly from CFD. Moreover, it is shown that our model effectively utilizes the exist-ing flow field information and the gradient information simultaneously, giving a better performance than the traditionalconvolution neural network (CNN)-based and deep neural network (DNN)-based models. Specifically, among all thecases of different Reynolds numbers and different proportions of the training dataset, the results show that the proposedFCN achieves a maximum norm mean square error of 5.86% in the test dataset, which is much lower than those of thetraditional CNN-based and DNN-based models (42.32% and 15.63% respectively).

physics.flu-dyn↗

arXiv2022

Ultra-fast synthesis and thermodynamic analysis of MoAlB by self-propagating high temperature combustion synthesis

MoAlB as a typical member of the MAB phases has attracted much growing attention due to its unique properties. However, the low production of MoAlB powders limits its further development and potential applications. To respond this challenge, the ultra-fast preparation of high-purity MoAlB powders is achieved in a few seconds by self-propagating high temperature combustion synthesis (SHS) using the raw powder mixture at the atomic ratio of 1Mo/1.3Al/1B. The reaction mechanism in SHS process was obtained by analyzing the corresponding composition changes of starting materials. Furthermore, the thermodynamic prediction for the SHS reaction of Mo + (1+y)Al + B = MoAlB + yAl is consistent with the present experiments, where the preparation of MoAlB also conforms to two common self-propagating conditions of SHS. Enthalpy vs temperature curve shows that the adiabatic temperature of the reaction decreases with increasing the amount of excuse Al, but increase when pre-heating the reactants. The thermodynamic calculation method also provides a new idea for the preparation of other MAB phases by SHS.

cond-mat.mtrl-sci↗