Author name cluster

Xing Wu

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

10 papers

1 author row

EAAI Journal 2026 Journal Article

Asynchronous multithreading reinforcement learning with attention-based significance measurement for collision-free robot navigation

Chao Sun
Jiang Wang
Xing Wu
Chaoxu Mu
Changyin Sun

Collision avoidance is one crucial technique to achieve safe and efficient robotic vehicle navigation in unknown environments. However, moving obstacles with unpredictability in dynamic scenarios, usually increase the difficulty and complexity in collision avoidance of robotic vehicles. To enhance the stability of collision avoidance and boost its adaptability to uncertain dynamic scenes, a new attention-based significance measurement actor–critic (ASMAC) architecture is proposed. It is an end-to-end robot navigation model that uses imperfect local observation to directly plan precise collision-free motion commands. Firstly, a significance-measured rollout replaybuffer (SMRR) is presented to categorize the experiences into different pools. It can prevent any overfitting or bias that may result from repeatedly sampling experience of a certain type during policy learning. Then, we enhance the traditional actor–critic network by integrating a multi-head local attention module to extract the local information at entity level. This way, the collision avoidance system can focus on key environmental features to compute more lightweight and respond more swiftly to dynamic changes in environment. Besides, a multi-step lookahead prediction (MLP) reward function is designed in the ASMAC-based reinforcement learning (RL) framework to prevent the generation of unnatural, intrusive, and short-sighted motion decisions. Finally, the asynchronous multithreading (AM) mechanism and proximal policy optimization (PPO) algorithm are extended to ASMAC model to offload the expensive online computation to an offline training process, enhancing the exploration efficiency in navigation policy learning of robotic vehicles. Extensive simulation and real-world physical experiments show that our method can generate time-efficient and collision-free guide paths in complex dynamic scenes, to successfully dodge collisions while moving towards the goal.

Details DOI

AAAI Conference 2026 Conference Paper

LiteLong: Resource-Efficient Long-Context Data Synthesis for LLMs

Junlong Jia
Xing Wu
Chaochen Gao
Ziyang Chen
Zijia Lin
Zhongzhi Li
Weinong Wang
Haotian Xu

High-quality long-context data is essential for training large language models (LLMs) capable of processing extensive documents, yet existing synthesis approaches using relevance-based aggregation face challenges of computational efficiency. We present LiteLong, a resource-efficient method for synthesizing long-context data through structured topic organization and multi-agent debate. Our approach leverages the BISAC book classification system to provide a comprehensive hierarchical topic organization, and then employs a debate mechanism with multiple LLMs to generate diverse, high-quality topics within this structure. For each topic, we use lightweight BM25 retrieval to obtain relevant documents and concatenate them into 128K-token training samples. Experiments on HELMET and Ruler benchmarks demonstrate that LiteLong achieves competitive long-context performance and can seamlessly integrate with other long-dependency enhancement methods. LiteLong makes high-quality long-context data synthesis more accessible by reducing both computational and data engineering costs, facilitating further research in long-context language training.

PDF Details DOI

AAAI Conference 2026 Conference Paper

Put the Space of LoRA Initialization to the Extreme to Preserve Pre-trained Knowledge

Pengwei Tang
Xiaolin Hu
Yong Liu
Lizhong Ding
Dongjie Zhang
Xing Wu
Debing Zhang

Low-Rank Adaptation (LoRA) is the leading parameter-efficient fine-tuning method for Large Language Models (LLMs), but it still suffers from catastrophic forgetting. Recent work has shown that specialized LoRA initialization can alleviate catastrophic forgetting. There are currently two approaches to LoRA initialization aimed at preventing knowledge forgetting during fine-tuning: (1) making residual weights close to pre-trained weights, and (2) ensuring the space of LoRA initialization is orthogonal to pre-trained knowledge. The former is what current methods strive to achieve, while the importance of the latter is not sufficiently recognized. We find that the space of LoRA initialization is the key to preserving pre-trained knowledge rather than the residual weights. Existing methods like MiLoRA propose making the LoRA initialization space orthogonal to pre-trained weights. However, MiLoRA utilizes the null space of pre-trained weights. Compared to pre-trained weights, the input activations of pre-trained knowledge take into account the parameters of all previous layers as well as the input data, while pre-trained weights only contain information from the current layer. Moreover, we find that the effective ranks of input activations are much smaller than those of pre-trained weights. Thus, the null space of activations is more accurate and contains less pre-trained knowledge information compared to that of weights. Based on these, we introduce LoRA-Null, our proposed method that initializes LoRA in the null space of activations. Experimental results show that LoRA-Null effectively preserves the pre-trained world knowledge of LLMs while achieving good fine-tuning performance, as evidenced by extensive experiments.

PDF Details DOI

EAAI Journal 2025 Journal Article

A time and frequency convolutional Autoencoder for anomaly detection in industrial robots based on inertial measurement unit error calibration

Jianlong Li
Xiaoqin Liu
Xing Wu
Dongxiao Wang
Kai Xu
Yashan Li

In the realm of industrial robots, ensuring operational reliability and Long-Term Autonomy hinges on the accurate detection of anomalies. However, this sample difference due to noise, joint random errors and sensor errors increases the challenge of robot anomaly detection. To address this problem, an unsupervised deep learning method based on inertial measurement unit (IMU) error calibration is proposed. Firstly, the attitude signals acquired by the IMU from the end of the robot were calibrated using Kalman filtering. The three dimensional (3D) free acceleration was corrected based on the calibrated attitude signal and the calibrated 3D free acceleration signal was used as a signal sample. Secondly, a time and frequency convolutional autoencoder model (TFCAE) is proposed. And the distribution of the different component signals is fitted by stacking multiple encoder modules and 3D-TFCAE is used for 3D free acceleration signal reconstruction model. Then, the error sphere radius is calculated based on the reconstruction error of the 3D free acceleration signal. And the error sphere radius is used as the anomaly detection threshold to realize the robust detection of different types of anomalies. The model was evaluated on a constructed anomaly dataset. This study contributes an innovative 3D-TFCAE architecture, integrating Kalman filtering with time-frequency feature fusion, markedly enhancing anomaly detection in complex signal environments. Experimental findings reveal that 3D-TFCAE significantly outperforms 18 baseline models, improving detection accuracy by about 20 %–40 %, offering an effective solution for high-precision anomaly detection in industrial robots. The code for this project is available at https: //github. com/LJlong977/3DTFCAE.

Details DOI

AAAI Conference 2025 Conference Paper

Task-level Distributionally Robust Optimization for Large Language Model-based Dense Retrieval

Guangyuan Ma
Yongliang Ma
Xing Wu
Zhenpeng Su
Ming Zhou
Songlin Hu

Large Language Model-based Dense Retrieval (LLM-DR) optimizes over numerous heterogeneous fine-tuning collections from different domains. However, the discussion about its training data distribution is still minimal. Previous studies rely on empirically assigned dataset choices or sampling ratios, which inevitably lead to sub-optimal retrieval performances. In this paper, we propose a new task-level Distributionally Robust Optimization (tDRO) algorithm for LLM-DR fine-tuning, targeted at improving the universal domain generalization ability by end-to-end reweighting the data distribution of each task. The tDRO parameterizes the domain weights and updates them with scaled domain gradients. The optimized weights are then transferred to the LLM-DR fine-tuning to train more robust retrievers. Experiments show optimal improvements in large-scale retrieval benchmarks and reduce up to 30% dataset usage after applying our optimization algorithm with a series of different-sized LLM-DR models.

PDF Details DOI

IS Journal 2024 Journal Article

A Text-Enhanced Transformer Fusion Network for Multimodal Knowledge Graph Completion

Jingchao Wang
Xiao Liu
Weimin Li
Fangfang Liu
Xing Wu
Qun Jin

Multimodal knowledge graphs (MKGs) organize multimodal facts in the form of entities and relations, and have been successfully applied to several downstream tasks. As most MKGs are incomplete, the MKG completion task has been proposed to address this problem, which aims to complete missing entities in MKGs. Most of the previous works obtain reasoning ability by capturing the correlation between target triplets and related images, but they ignore contextual semantic information and the reasoning process is not easily explainable. To address these issues, we propose a novel text-enhanced transformer fusion network, which converts the context path between head and tail entities into natural language text and fuses multimodal features from both coarse and fine granularities through a multigranularity fuser. It not only effectively enhances text semantic information but also improves the interpretability of the model by introducing paths. Experimental results on benchmark datasets demonstrate the effectiveness of our model.

Details DOI

EAAI Journal 2023 Journal Article

Better utilization of materials’ compositions for predicting their properties: Material composition visualization network

Yeyong Yu
Xing Wu
Quan Qian

Owing to the complexity and diversity of advanced high-performance materials, it is challenging to comprehensively understand a material’s composition–process–structure–performance relationship. Data-driven approaches have been regarded as the fourth paradigm of new materials R&D. However, the complexity of constituent elements in many material datasets leads to very sparse compositional features, posing a tremendous challenge to machine learning models. In this study, a data mapping scheme based on fundamental atomic features was used to visualize the chemical composition characteristics mapped into two-dimensional grayscale image data to solve the problem of sparse material composition matrix. Based on this, a material composition visualization network (MCVN) is proposed and applied to predict the mechanical properties of steel and classify amorphous alloy materials. We compared the MCVN to other machine learning methods. The MCVN had an average R 2 value improvement of 4% on the four targets in the National Institute for Materials Science’s (NIMS’s) steel dataset, where other models already get an average R 2 of 0. 92, and it achieved an R 2 of 0. 835 on the cross-sectional shrinkage target in the Shanghai Research Institute of Materials’(SRIM’s) steel dataset where the other models only had an average R 2 of 0. 64. For the unbalanced amorphous alloy material dataset, the MCVN improved the average R e c a l l of the small-class crystalline alloy (CRA) from 0. 58 to 0. 78. The method based on expanding the material chemical composition information is universal and provides a new paradigm for material property prediction.

Details DOI

AAAI Conference 2023 Conference Paper

ConTextual Masked Auto-Encoder for Dense Passage Retrieval

Xing Wu
Guangyuan Ma
Meng Lin
Zijia Lin
Zhongyuan Wang
Songlin Hu

Dense passage retrieval aims to retrieve the relevant passages of a query from a large corpus based on dense representations (i.e., vectors) of the query and the passages. Recent studies have explored improving pre-trained language models to boost dense retrieval performance. This paper proposes CoT-MAE (ConTextual Masked Auto-Encoder), a simple yet effective generative pre-training method for dense passage retrieval. CoT-MAE employs an asymmetric encoder-decoder architecture that learns to compress the sentence semantics into a dense vector through self-supervised and context-supervised masked auto-encoding. Precisely, self-supervised masked auto-encoding learns to model the semantics of the tokens inside a text span, and context-supervised masked auto-encoding learns to model the semantical correlation between the text spans. We conduct experiments on large-scale passage retrieval benchmarks and show considerable improvements over strong baselines, demonstrating the high efficiency of CoT-MAE. Our code is available at https://github.com/caskcsg/ir/tree/main/cotmae.

PDF Details DOI

JBHI Journal 2023 Journal Article

The Human Activity Radar Challenge: Benchmarking Based on the ‘Radar Signatures of Human Activities’ Dataset From Glasgow University

Shufan Yang
Julien Le Kernec
Olivier Romain
Francesco Fioranelli
Pierre Cadart
Jérémy Fix
Chenfang Ren
Giovanni Manfredi

Radar is an extremely valuable sensing technology for detecting moving targets and measuring their range, velocity, and angular positions. When people are monitored at home, radar is more likely to be accepted by end-users, as they already use WiFi, is perceived as privacy-preserving compared to cameras, and does not require user compliance as wearable sensors do. Furthermore, it is not affected by lighting conditions nor requires artificial lights that could cause discomfort in the home environment. So, radar-based human activities classification in the context of assisted living can empower an aging society to live at home independently longer. However, challenges remain as to the formulation of the most effective algorithms for radar-based human activities classification and their validation. To promote the exploration and cross-evaluation of different algorithms, our dataset released in 2019 was used to benchmark various classification approaches. The challenge was open from February 2020 to December 2020. A total of 23 organizations worldwide, forming 12 teams from academia and industry, participated in the inaugural Radar Challenge, and submitted 188 valid entries to the challenge. This paper presents an overview and evaluation of the approaches used for all primary contributions in this inaugural challenge. The proposed algorithms are summarized, and the main parameters affecting their performances are analyzed.

Details DOI

IJCAI Conference 2019 Conference Paper

Mask and Infill: Applying Masked Language Model for Sentiment Transfer

Xing Wu
Tao Zhang
Liangjun Zang
Jizhong Han
Songlin Hu

This paper focuses on the task of sentiment transfer on non-parallel text, which modifies sentiment attributes (e. g. , positive or negative) of sentences while preserving their attribute-independent contents. Existing methods adopt RNN encoder-decoder structure to generate a new sentence of a target sentiment word by word, which is trained on a particular dataset from scratch and have limited ability to produce satisfactory sentences. When people convert the sentiment attribute of a given sentence, a simple but effective approach is to only replace the sentiment tokens of the sentence with other expressions indicative of the target sentiment, instead of building a new sentence from scratch. Such a process is very similar to the task of Text Infilling or Cloze. With this intuition, we propose a two steps approach: Mask and Infill. In the \emph{mask} step, we identify and mask the sentiment tokens of a given sentence. In the \emph{infill} step, we utilize a pre-trained Masked Language Model (MLM) to infill the masked positions by predicting words or phrases conditioned on the context\footnote{In this paper, \emph{content} and \emph{context} are equivalent, \emph{style}, \emph{attribute} and \emph{label} are equivalent. }and target sentiment. We evaluate our model on two review datasets \emph{Yelp} and \emph{Amazon} by quantitative, qualitative, and human evaluations. Experimental results demonstrate that our model achieve state-of-the-art performance on both accuracy and BLEU scores.

PDF Details