arXiv subjects

Liang He

Publications and source records attributed to Liang He.

At least 235 records · Page 13Linked to original sources

arXiv2023

EduChat: A Large-Scale Language Model-based Chatbot System for Intelligent Education

EduChat (https://www.educhat.top/) is a large-scale language model (LLM)-based chatbot system in the education domain. Its goal is to support personalized, fair, and compassionate intelligent education, serving teachers, students, and parents. Guided by theories from psychology and education, it further strengthens educational functions such as open question answering, essay assessment, Socratic teaching, and emotional support based on the existing basic LLMs. Particularly, we learn domain-specific knowledge by pre-training on the educational corpus and stimulate various skills with tool use by fine-tuning on designed system prompts and instructions. Currently, EduChat is available online as an open-source project, with its code, data, and model parameters available on platforms (e.g., GitHub https://github.com/icalk-nlp/EduChat, Hugging Face https://huggingface.co/ecnu-icalk ). We also prepare a demonstration of its capabilities online (https://vimeo.com/851004454). This initiative aims to promote research and applications of LLMs for intelligent education.

cs.CL↗

arXiv2023

Robust Training for Speaker Verification against Noisy Labels

The deep learning models used for speaker verification rely heavily on large amounts of data and correct labeling. However, noisy (incorrect) labels often occur, which degrades the performance of the system. In this paper, we propose a novel two-stage learning method to filter out noisy labels from speaker datasets. Since a DNN will first fit data with clean labels, we first train the model with all data for several epochs. Then, based on this model, the model predictions are compared with the labels using our proposed the OR-Gate with top-k mechanism to select the data with clean labels and the selected data is used to train the model. This process is iterated until the training is completed. We have demonstrated the effectiveness of this method in filtering noisy labels through extensive experiments and have achieved excellent performance on the VoxCeleb (1 and 2) with different added noise rates.

cs.SD↗

arXiv2023

Scaling behavior of transient dynamics of vortex-like states in self-propelled particles

Nonequilibrium many-body transient dynamics play an important role in the adaptation of active matter systems environment changes. However, the generic universal behavior of such dynamics is usually elusive and left as open questions. Here, we investigate the transient dynamics of vortex-like states in a two-dimensional active matter system that consists of self-propelled particles with alignment interactions subjected to extrinsic environmental noise. We identify a universal power-law scaling for the average lifetime of vortex-like states with respect to the speed of the self-propelled particles. This universal scaling behavior manifests strong robustness against the noise, up to the level where influences from environmental fluctuations are large enough to directly randomize the moving directions of particles. Direct experimental observations can be readily performed by related experimental setups operated at a decently low noise level.

cond-mat.stat-mech↗

arXiv2023

Emergent Mott-insulators at non-integer fillings and devil's staircase induced by attractive interaction in many-body polarons

We investigate the ground state properties of an ultracold atom system consisting of many-body polarons, quasiparticles formed by impurity atoms in optical lattices immersing in a Bose-Einstein condensate. We find the nearest-neighbor attractive interaction between polarons can give rise to rich physics that is peculiar to this system. In a relatively shallow optical lattice, the attractive interaction can drive the system being in a self-bound superfluid phase with its particle density distribution manifesting a self-concentrated structure. While in a relatively deep optical lattice, the attractive interaction can drive the system forming the Mott-insulator phase even though the global filling factor is not integer. Interestingly, in the Mott-insulator regime, the system can support a series of different Mott-insulators with their effective density manifesting a devil's staircase structure with respect to the strength of attractive interaction. Detailed estimation on relevant experimental parameters shows that these rich physics can be readily observed in current experimental setups.

cond-mat.quant-gas↗

arXiv2023

MMFormer: Multimodal Transformer Using Multiscale Self-Attention for Remote Sensing Image Classification

To benefit the complementary information between heterogeneous data, we introduce a new Multimodal Transformer (MMFormer) for Remote Sensing (RS) image classification using Hyperspectral Image (HSI) accompanied by another source of data such as Light Detection and Ranging (LiDAR). Compared with traditional Vision Transformer (ViT) lacking inductive biases of convolutions, we first introduce convolutional layers to our MMFormer to tokenize patches from multimodal data of HSI and LiDAR. Then we propose a Multi-scale Multi-head Self-Attention (MSMHSA) module to address the problem of compatibility which often limits to fuse HSI with high spectral resolution and LiDAR with relatively low spatial resolution. The proposed MSMHSA module can incorporate HSI to LiDAR data in a coarse-to-fine manner enabling us to learn a fine-grained representation. Extensive experiments on widely used benchmarks (e.g., Trento and MUUFL) demonstrate the effectiveness and superiority of our proposed MMFormer for RS image classification.

cs.CV↗

arXiv2023

A Review on Machine Theory of Mind

Theory of Mind (ToM) is the ability to attribute mental states to others, the basis of human cognition. At present, there has been growing interest in the AI with cognitive abilities, for example in healthcare and the motoring industry. Beliefs, desires, and intentions are the early abilities of infants and the foundation of human cognitive ability, as well as for machine with ToM. In this paper, we review recent progress in machine ToM on beliefs, desires, and intentions. And we shall introduce the experiments, datasets and methods of machine ToM on these three aspects, summarize the development of different tasks and datasets in recent years, and compare well-behaved models in aspects of advantages, limitations and applicable conditions, hoping that this study can guide researchers to quickly keep up with latest trend in this field. Unlike other domains with a specific task and resolution framework, machine ToM lacks a unified instruction and a series of standard evaluation tasks, which make it difficult to formally compare the proposed models. We argue that, one method to address this difficulty is now to present a standard assessment criteria and dataset, better a large-scale dataset covered multiple aspects of ToM.

cs.AI↗

arXiv2023

Real-Time Decentralized Navigation of Nonholonomic Agents Using Shifted Yielding Areas

We present a lightweight, decentralized algorithm for navigating multiple nonholonomic agents through challenging environments with narrow passages. Our key idea is to allow agents to yield to each other in large open areas instead of narrow passages, to increase the success rate of conventional decentralized algorithms. At pre-processing time, our method computes a medial axis for the freespace. A reference trajectory is then computed and projected onto the medial axis for each agent. During run time, when an agent senses other agents moving in the opposite direction, our algorithm uses the medial axis to estimate a Point of Impact (POI) as well as the available area around the POI. If the area around the POI is not large enough for yielding behaviors to be successful, we shift the POI to nearby large areas by modulating the agent's reference trajectory and traveling speed. We evaluate our method on a row of 4 environments with up to 15 robots, and we find our method incurs a marginal computational overhead of 10-30 ms on average, achieving real-time performance. Afterward, our planned reference trajectories can be tracked using local navigation algorithms to achieve up to a $100\%$ higher success rate over local navigation algorithms alone.

cs.RO↗

arXiv2023

LoGoNet: Towards Accurate 3D Object Detection with Local-to-Global Cross-Modal Fusion

LiDAR-camera fusion methods have shown impressive performance in 3D object detection. Recent advanced multi-modal methods mainly perform global fusion, where image features and point cloud features are fused across the whole scene. Such practice lacks fine-grained region-level information, yielding suboptimal fusion performance. In this paper, we present the novel Local-to-Global fusion network (LoGoNet), which performs LiDAR-camera fusion at both local and global levels. Concretely, the Global Fusion (GoF) of LoGoNet is built upon previous literature, while we exclusively use point centroids to more precisely represent the position of voxel features, thus achieving better cross-modal alignment. As to the Local Fusion (LoF), we first divide each proposal into uniform grids and then project these grid centers to the images. The image features around the projected grid points are sampled to be fused with position-decorated point cloud features, maximally utilizing the rich contextual information around the proposals. The Feature Dynamic Aggregation (FDA) module is further proposed to achieve information interaction between these locally and globally fused features, thus producing more informative multi-modal features. Extensive experiments on both Waymo Open Dataset (WOD) and KITTI datasets show that LoGoNet outperforms all state-of-the-art 3D detection methods. Notably, LoGoNet ranks 1st on Waymo 3D object detection leaderboard and obtains 81.02 mAPH (L2) detection performance. It is noteworthy that, for the first time, the detection performance on three classes surpasses 80 APH (L2) simultaneously. Code will be available at \url{https://github.com/sankin97/LoGoNet}.

cs.CV↗

arXiv2023

Self-Attentive Sequential Recommendation with Cheap Causal Convolutions

Sequential Recommendation is a prominent topic in current research, which uses user behavior sequence as an input to predict future behavior. By assessing the correlation strength of historical behavior through the dot product, the model based on the self-attention mechanism can capture the long-term preference of the sequence. However, it has two limitations. On the one hand, it does not effectively utilize the items' local context information when determining the attention and creating the sequence representation. On the other hand, the convolution and linear layers often contain redundant information, which limits the ability to encode sequences. In this paper, we propose a self-attentive sequential recommendation model based on cheap causal convolution. It utilizes causal convolutions to capture items' local information for calculating attention and generating sequence embedding. It also uses cheap convolutions to improve the representations by lightweight structure. We evaluate the effectiveness of the proposed model in terms of both accurate and calibrated sequential recommendation. Experiments on benchmark datasets show that the proposed model can perform better in single- and multi-objective recommendation scenarios.

cs.IR↗

arXiv2023

Tell Model Where to Attend: Improving Interpretability of Aspect-Based Sentiment Classification via Small Explanation Annotations

Gradient-based explanation methods play an important role in the field of interpreting complex deep neural networks for NLP models. However, the existing work has shown that the gradients of a model are unstable and easily manipulable, which impacts the model's reliability largely. According to our preliminary analyses, we also find the interpretability of gradient-based methods is limited for complex tasks, such as aspect-based sentiment classification (ABSC). In this paper, we propose an \textbf{I}nterpretation-\textbf{E}nhanced \textbf{G}radient-based framework for \textbf{A}BSC via a small number of explanation annotations, namely \texttt{IEGA}. Particularly, we first calculate the word-level saliency map based on gradients to measure the importance of the words in the sentence towards the given aspect. Then, we design a gradient correction module to enhance the model's attention on the correct parts (e.g., opinion words). Our model is model agnostic and task agnostic so that it can be integrated into the existing ABSC methods or other tasks. Comprehensive experimental results on four benchmark datasets show that our \texttt{IEGA} can improve not only the interpretability of the model but also the performance and robustness.

cs.CL↗

arXiv2022

Incorporating Pre-training Paradigm for Antibody Sequence-Structure Co-design

Antibodies are versatile proteins that can bind to pathogens and provide effective protection for human body. Recently, deep learning-based computational antibody design has attracted popular attention since it automatically mines the antibody patterns from data that could be complementary to human experiences. However, the computational methods heavily rely on high-quality antibody structure data, which is quite limited. Besides, the complementarity-determining region (CDR), which is the key component of an antibody that determines the specificity and binding affinity, is highly variable and hard to predict. Therefore, the data limitation issue further raises the difficulty of CDR generation for antibodies. Fortunately, there exists a large amount of sequence data of antibodies that can help model the CDR and alleviate the reliance on structure data. By witnessing the success of pre-training models for protein modeling, in this paper, we develop the antibody pre-training language model and incorporate it into the (antigen-specific) antibody design model in a systemic way. Specifically, we first pre-train an antibody language model based on the sequence data, then propose a one-shot way for sequence and structure generation of CDR to avoid the heavy cost and error propagation from an autoregressive manner, and finally leverage the pre-trained antibody model for the antigen-specific antibody generation model with some carefully designed modules. Through various experiments, we show that our method achieves superior performances over previous baselines on different tasks, such as sequence and structure generation and antigen-binding CDR-H3 design.

q-bio.BM↗

arXiv2022

I4U System Description for NIST SRE'20 CTS Challenge

This manuscript describes the I4U submission to the 2020 NIST Speaker Recognition Evaluation (SRE'20) Conversational Telephone Speech (CTS) Challenge. The I4U's submission was resulted from active collaboration among researchers across eight research teams - I$^2$R (Singapore), UEF (Finland), VALPT (Italy, Spain), NEC (Japan), THUEE (China), LIA (France), NUS (Singapore), INRIA (France) and TJU (China). The submission was based on the fusion of top performing sub-systems and sub-fusion systems contributed by individual teams. Efforts have been spent on the use of common development and validation sets, submission schedule and milestone, minimizing inconsistency in trial list and score file format across sites.

eess.AS↗

arXiv2022

DACSR: Decoupled-Aggregated End-to-End Calibrated Sequential Recommendation

Sequential recommendations have made great strides in accurately predicting the future behavior of users. However, seeking accuracy alone may bring side effects such as unfair and overspecialized recommendation results. In this work, we focus on the calibrated recommendations for sequential recommendation, which is connected to both fairness and diversity. On the one hand, it aims to provide fairer recommendations whose preference distributions are consistent with users' historical behaviors. On the other hand, it can improve the diversity of recommendations to a certain degree. But existing methods for calibration have mainly relied on the post-processing on the candidate lists, which require more computation time in generating recommendations. In addition, they fail to establish the relationship between accuracy and calibration, leading to the limitation of accuracy. To handle these problems, we propose an end-to-end framework to provide both accurate and calibrated recommendations for sequential recommendation. We design an objective function to calibrate the interests between recommendation lists and historical behaviors. We also provide distribution modification approaches to improve the diversity and mitigate the effect of imbalanced interests. In addition, we design a decoupled-aggregated model to improve the recommendation. The framework assigns two objectives to two individual sequence encoders, and aggregates the outputs by extracting useful information. Experiments on benchmark datasets validate the effectiveness of our proposed model.

cs.IR↗

arXiv2022

Homogeneous Multi-modal Feature Fusion and Interaction for 3D Object Detection

Multi-modal 3D object detection has been an active research topic in autonomous driving. Nevertheless, it is non-trivial to explore the cross-modal feature fusion between sparse 3D points and dense 2D pixels. Recent approaches either fuse the image features with the point cloud features that are projected onto the 2D image plane or combine the sparse point cloud with dense image pixels. These fusion approaches often suffer from severe information loss, thus causing sub-optimal performance. To address these problems, we construct the homogeneous structure between the point cloud and images to avoid projective information loss by transforming the camera features into the LiDAR 3D space. In this paper, we propose a homogeneous multi-modal feature fusion and interaction method (HMFI) for 3D object detection. Specifically, we first design an image voxel lifter module (IVLM) to lift 2D image features into the 3D space and generate homogeneous image voxel features. Then, we fuse the voxelized point cloud features with the image features from different regions by introducing the self-attention based query fusion mechanism (QFM). Next, we propose a voxel feature interaction module (VFIM) to enforce the consistency of semantic information from identical objects in the homogeneous point cloud and image voxel representations, which can provide object-level alignment guidance for cross-modal feature fusion and strengthen the discriminative ability in complex backgrounds. We conduct extensive experiments on the KITTI and Waymo Open Dataset, and the proposed HMFI achieves better performance compared with the state-of-the-art multi-modal methods. Particularly, for the 3D detection of cyclist on the KITTI benchmark, HMFI surpasses all the published algorithms by a large margin.

cs.CV↗

arXiv2022

THUEE system description for NIST 2020 SRE CTS challenge

This paper presents the system description of the THUEE team for the NIST 2020 Speaker Recognition Evaluation (SRE) conversational telephone speech (CTS) challenge. The subsystems including ResNet74, ResNet152, and RepVGG-B2 are developed as speaker embedding extractors in this evaluation. We used combined AM-Softmax and AAM-Softmax based loss functions, namely CM-Softmax. We adopted a two-staged training strategy to further improve system performance. We fused all individual systems as our final submission. Our approach leads to excellent performance and ranks 1st in the challenge.

cs.SD↗

arXiv2022

An Information Minimization Based Contrastive Learning Model for Unsupervised Sentence Embeddings Learning

Unsupervised sentence embeddings learning has been recently dominated by contrastive learning methods (e.g., SimCSE), which keep positive pairs similar and push negative pairs apart. The contrast operation aims to keep as much information as possible by maximizing the mutual information between positive instances, which leads to redundant information in sentence embedding. To address this problem, we present an information minimization based contrastive learning (InforMin-CL) model to retain the useful information and discard the redundant information by maximizing the mutual information and minimizing the information entropy between positive instances meanwhile for unsupervised sentence representation learning. Specifically, we find that information minimization can be achieved by simple contrast and reconstruction objectives. The reconstruction operation reconstitutes the positive instance via the other positive instance to minimize the information entropy between positive instances. We evaluate our model on fourteen downstream tasks, including both supervised and unsupervised (semantic textual similarity) tasks. Extensive experimental results show that our InforMin-CL obtains a state-of-the-art performance.

cs.CL↗

arXiv2022

A Multi-Format Transfer Learning Model for Event Argument Extraction via Variational Information Bottleneck

Event argument extraction (EAE) aims to extract arguments with given roles from texts, which have been widely studied in natural language processing. Most previous works have achieved good performance in specific EAE datasets with dedicated neural architectures. Whereas, these architectures are usually difficult to adapt to new datasets/scenarios with various annotation schemas or formats. Furthermore, they rely on large-scale labeled data for training, which is unavailable due to the high labelling cost in most cases. In this paper, we propose a multi-format transfer learning model with variational information bottleneck, which makes use of the information especially the common knowledge in existing datasets for EAE in new datasets. Specifically, we introduce a shared-specific prompt framework to learn both format-shared and format-specific knowledge from datasets with different formats. In order to further absorb the common knowledge for EAE and eliminate the irrelevant noise, we integrate variational information bottleneck into our architecture to refine the shared representation. We conduct extensive experiments on three benchmark datasets, and obtain new state-of-the-art performance on EAE.

cs.CL↗

arXiv2022

Graph Neural Networks with Motif-aware for Tenuous Subgraph Finding

Tenuous subgraph finding aims to detect a subgraph with few social interactions and weak relationships among nodes. Despite significant efforts have been made on this task, they are mostly carried out in view of graph-structured data. These methods depend on calculating the shortest path and need to enumerate all the paths between nodes, which suffer the combinatorial explosion. Moreover, they all lack the integration of neighborhood information. To this end, we propose a novel model named Graph Neural Network with Motif-aware for tenuous subgraph finding (GNNM), a neighborhood aggregation based GNNs framework which can capture the latent relationship between nodes. Specially, we design a GNN module to project nodes into a low dimensional vector combining the higher-order correlation within nodes based on a motif-aware module. And then design greedy algorithms in vector space to obtain a tenuous subgraph whose size is greater than a specified constraint. Particularly, considering existing evaluation indicators cannot capture the latent friendship between nodes, we introduce a novel Potential Friend(PF) concept to measure the tenuity of a graph from a new perspective. Experimental results on the real-world and synthetic datasets demonstrate that our proposed method GNNM outperforms existing algorithms in efficiency and subgraph quality.

cs.SI↗