Arrow Research search

Author name cluster

Wei Xia

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.

16 papers
2 author rows

Possible papers

16

AAAI Conference 2026 Conference Paper

AP2O-Coder: Adaptively Progressive Preference Optimization for Reducing Compilation and Runtime Errors in LLM-Generated Code

  • Jianqing Zhang
  • Wei Xia
  • Hande Dong
  • Qiang Lin
  • Jian Cao

LLM's code generation capabilities have yielded substantial improvements in the effectiveness of programming tasks. However, LLM-generated code still suffers from compilation and runtime errors. Existing offline preference optimization methods primarily focus on enhancing LLMs' coding abilities using pass/fail signals in the preference data, overlooking the deep-level error types in the failed codes. To address this, we propose Adaptively Progressive Preference Optimization (AP2O) for coding (i.e., AP2O-Coder), a method that guides LLMs adaptively and methodically to reduce code errors for code generation. Specifically, we construct an error notebook from failed codes and progressively optimize the LLM to correct errors type by type. Furthermore, we adaptively replay error types to tailor to the LLM's evolving weaknesses throughout training. Through extensive experiments on both code and general LLMs (Llama, Qwen, and DeepSeek series) with parameters ranging from 0.5B to 34B, our AP2O-Coder improves code generation performance by up to 3% in pass@k while using less preference data.

PDF Details DOI

AAAI Conference 2026 Conference Paper

SDA: Steering-Driven Distribution Alignment for Open LLMs Without Fine-Tuning

  • Wei Xia
  • Zhi-Hong Deng

With the rapid advancement of large language models (LLMs), their deployment in real-world applications has become increasingly widespread. LLMs are expected to deliver robust performance across diverse tasks, user preferences, and practical scenarios. However, as demands grow, ensuring that LLMs produce responses aligned with human intent remains a foundational challenge. In particular, aligning model behavior effectively and efficiently during inference, without costly retraining or extensive supervision, is both a critical requirement and a non-trivial technical endeavor. To address the challenge, we propose SDA (Steering-Driven Distribution Alignment), a training-free and model-agnostic alignment framework designed for open-source LLMs. SDA dynamically redistributes model output probabilities based on user-defined alignment instructions, enhancing alignment between model behavior and human intents without fine-tuning. The method is lightweight, resource-efficient, and compatible with a wide range of open-source LLMs. It can function independently during inference or be integrated with training-based alignment strategies. Moreover, SDA supports personalized preference alignment, enabling flexible control over the model’s response behavior. Empirical results demonstrate that SDA consistently improves alignment performance across 8 open-source LLMs with varying scales and diverse origins, evaluated on three key alignment dimensions, helpfulness, harmlessness, and honesty (3H). Specifically, SDA achieves average gains of 64.4% in helpfulness, 30% in honesty and 11.5% in harmlessness across the tested models, indicating its effectiveness and generalization across diverse models and application scenarios.

PDF Details DOI

YNIMG Journal 2025 Journal Article

White-Matter fiber tract and resting-state functional connectivity abnormalities in young children with autism spectrum disorder

  • Jia Wang
  • Natasha Y.S. Kawata
  • Xuan Cao
  • Jie Zhang
  • Takashi X. Fujisawa
  • Xinyi Zhang
  • Lili Fan
  • Wei Xia

Autism spectrum disorder (ASD) is a complex developmental disorder characterized by difficulties in social interaction and communication and repetitive behaviors. Although abnormal brain development has been shown to exist in children with ASD, the link between structural brain abnormalities and resting-state functional connectivity (rsFC) disruptions in children with ASD remains understudied. To address this limitation, we utilized the population-based bundle-to-region connectome, providing a detailed understanding of the connectivity between cortical regions and white matter (WM) tracts. By precisely indexing WM-Gray Matter (GM) interactions, we investigated the rsFC of the cortex-associated ROIs to explore the association between structural and rsFC abnormalities and clinical symptoms in young children with ASD. This MRI study identified significant differences in WM structure and rsFC between children with ASD (n = 34) and typically developing children (TD, n = 43). Our results showed that decreased fractional anisotropy (FA) and increased mean diffusivity (MD) and radial diffusivity (RD) in ASD WM tracts compared to TD, particularly in left hemisphere tracts (anterior thalamic radiation [ATR], cingulum, inferior fronto-occipital fasciculus [IFOF], inferior longitudinal fasciculus [ILF], superior longitudinal fasciculus [SLF], and uncinate fasciculus [UF]). Abnormal rsFC was observed in GM areas connected by ATR, cingulum, IFOF, ILF, and SLF. Furthermore, abnormalities in the structural and functional connectivity index (SFCI) within the SLF and cingulum were identified. An association has been observed between these abnormalities and clinical symptoms. Specifically, SLF structural and functional connectivity appear to be associated with repetitive and restrictive behavior (RRB), while cingulum connectivity is associated with communication abilities. In conclusion, young children with ASD exhibit abnormal WM tract structures and associated rsFC abnormalities. These differences highlight significant disruptions in rsFC mapped from WM tracts to cortical areas in ASD, correlating with the severity of ASD symptoms, and suggest the importance of multi-modal imaging in capturing these variations.

Details DOI

NeurIPS Conference 2024 Conference Paper

A Full-duplex Speech Dialogue Scheme Based On Large Language Model

  • Peng Wang
  • Songshuo Lu
  • Yaohua Tang
  • Sijie Yan
  • Wei Xia
  • Yuanjun Xiong

We present a generative dialogue system capable of operating in a full-duplex manner, allowing for seamless interaction. It is based on a large language model (LLM) carefully aligned to be aware of a perception module, a motor function module, and the concept of a simple finite state machine (called neural FSM) with two states. The perception and motor function modules operate in tandem, allowing the system to speak and listen to the user simultaneously. The LLM generates textual tokens for inquiry responses and makes autonomous decisions to start responding to, wait for, or interrupt the user by emitting control tokens to the neural FSM. All these tasks of the LLM are carried out as next token prediction on a serialized view of the dialogue in real-time. In automatic quality evaluations simulating real-life interaction, the proposed system reduces the average conversation response latency by more than threefold compared with LLM-based half-duplex dialogue systems while responding within less than 500 milliseconds in more than 50% of evaluated interactions. Running an LLM with only 8 billion parameters, our system exhibits an 8% higher interruption precision rate than the best available commercial LLM for voice-based dialogue.

PDF Details DOI

AAAI Conference 2024 Conference Paper

Tensorized Label Learning on Anchor Graph

  • Jing Li
  • Quanxue Gao
  • Qianqian Wang
  • Wei Xia

Graph-based multimedia data clustering has attracted much attention due to the impressive clustering performance for arbitrarily shaped multimedia data. However, existing graph-based clustering methods need post-processing to get labels for multimedia data with high computational complexity. Moreover, it is sub-optimal for label learning due to the fact that they exploit the complementary information embedded in data with different types pixel by pixel. To handle these problems, we present a novel label learning model with good interpretability for clustering. To be specific, our model decomposes anchor graph into the products of two matrices with orthogonal non-negative constraint to directly get soft label without any post-processing, which remarkably reduces the computational complexity. To well exploit the complementary information embedded in multimedia data, we introduce tensor Schatten p-norm regularization on the label tensor which is composed of soft labels of multimedia data. The solution can be obtained by iteratively optimizing four decoupled sub-problems, which can be solved more efficiently with good convergence. Experimental results on various datasets demonstrate the efficiency of our model.

PDF Details DOI

AAAI Conference 2023 Conference Paper

Centerless Multi-View K-means Based on the Adjacency Matrix

  • Han Lu
  • Quanxue Gao
  • Qianqian Wang
  • Ming Yang
  • Wei Xia

Although K-Means clustering has been widely studied due to its simplicity, these methods still have the following fatal drawbacks. Firstly, they need to initialize the cluster centers, which causes unstable clustering performance. Secondly, they have poor performance on non-Gaussian datasets. Inspired by the affinity matrix, we propose a novel multi-view K-Means based on the adjacency matrix. It maps the affinity matrix to the distance matrix according to the principle that every sample has a small distance from the points in its neighborhood and a large distance from the points outside of the neighborhood. Moreover, this method well exploits the complementary information embedded in different views by minimizing the tensor Schatten p-norm regularize on the third-order tensor which consists of cluster assignment matrices of different views. Additionally, this method avoids initializing cluster centroids to obtain stable performance. And there is no need to compute the means of clusters so that our model is not sensitive to outliers. Experiment on a toy dataset shows the excellent performance on non-Gaussian datasets. And other experiments on several benchmark datasets demonstrate the superiority of our proposed method.

PDF Details DOI

NeurIPS Conference 2023 Conference Paper

Orthogonal Non-negative Tensor Factorization based Multi-view Clustering

  • Jing Li
  • Quanxue Gao
  • Qianqian Wang
  • Ming Yang
  • Wei Xia

Multi-view clustering (MVC) based on non-negative matrix factorization (NMF) and its variants have attracted much attention due to their advantages in clustering interpretability. However, existing NMF-based multi-view clustering methods perform NMF on each view respectively and ignore the impact of between-view. Thus, they can't well exploit the within-view spatial structure and between-view complementary information. To resolve this issue, we present orthogonal non-negative tensor factorization (Orth-NTF) and develop a novel multi-view clustering based on Orth-NTF with one-side orthogonal constraint. Our model directly performs Orth-NTF on the 3rd-order tensor which is composed of anchor graphs of views. Thus, our model directly considers the between-view relationship. Moreover, we use the tensor Schatten $p$-norm regularization as a rank approximation of the 3rd-order tensor which characterizes the cluster structure of multi-view data and exploits the between-view complementary information. In addition, we provide an optimization algorithm for the proposed method and prove mathematically that the algorithm always converges to the stationary KKT point. Extensive experiments on various benchmark datasets indicate that our proposed method is able to achieve satisfactory clustering performance.

PDF Details

AAAI Conference 2023 Conference Paper

Set-to-Sequence Ranking-Based Concept-Aware Learning Path Recommendation

  • Xianyu Chen
  • Jian Shen
  • Wei Xia
  • Jiarui Jin
  • Yakun Song
  • Weinan Zhang
  • Weiwen Liu
  • Menghui Zhu

With the development of the online education system, personalized education recommendation has played an essential role. In this paper, we focus on developing path recommendation systems that aim to generating and recommending an entire learning path to the given user in each session. Noticing that existing approaches fail to consider the correlations of concepts in the path, we propose a novel framework named Set-to-Sequence Ranking-based Concept-aware Learning Path Recommendation (SRC), which formulates the recommendation task under a set-to-sequence paradigm. Specifically, we first design a concept-aware encoder module which can capture the correlations among the input learning concepts. The outputs are then fed into a decoder module that sequentially generates a path through an attention mechanism that handles correlations between the learning and target concepts. Our recommendation policy is optimized by policy gradient. In addition, we also introduce an auxiliary module based on knowledge tracing to enhance the model’s stability by evaluating students’ learning effects on learning concepts. We conduct extensive experiments on two real-world public datasets and one industrial dataset, and the experimental results demonstrate the superiority and effectiveness of SRC. Code now is available at https://gitee.com/mindspore/models/tree/master/research/recommend/SRC.

PDF Details DOI

ICLR Conference 2022 Conference Paper

Path Auxiliary Proposal for MCMC in Discrete Space

  • Haoran Sun
  • Hanjun Dai
  • Wei Xia
  • Arun Ramamurthy

Energy-based Model (EBM) offers a powerful approach for modeling discrete structure, but both inference and learning of EBM are hard as it involves sampling from discrete distributions. Recent work shows Markov Chain Monte Carlo (MCMC) with the informed proposal is a powerful tool for such sampling. However, an informed proposal only allows local updates as it requires evaluating all energy changes in the neighborhood. In this work, we present a path auxiliary algorithm that uses a composition of local moves to efficiently explore large neighborhoods. We also give a fast version of our algorithm that only queries the evaluation of energy function twice for each proposal via linearization of the energy function. Empirically, we show that our path auxiliary algorithms considerably outperform other generic samplers on various discrete models for sampling, inference, and learning. Our method can also be used to train deep EBMs for high-dimensional discrete data.

Details

NeurIPS Conference 2021 Conference Paper

Long Short-Term Transformer for Online Action Detection

  • Mingze Xu
  • Yuanjun Xiong
  • Hao Chen
  • Xinyu Li
  • Wei Xia
  • Zhuowen Tu
  • Stefano Soatto

We present Long Short-term TRansformer (LSTR), a temporal modeling algorithm for online action detection, which employs a long- and short-term memory mechanism to model prolonged sequence data. It consists of an LSTR encoder that dynamically leverages coarse-scale historical information from an extended temporal window (e. g. , 2048 frames spanning of up to 8 minutes), together with an LSTR decoder that focuses on a short time window (e. g. , 32 frames spanning 8 seconds) to model the fine-scale characteristics of the data. Compared to prior work, LSTR provides an effective and efficient method to model long videos with fewer heuristics, which is validated by extensive empirical analysis. LSTR achieves state-of-the-art performance on three standard online action detection benchmarks, THUMOS'14, TVSeries, and HACS Segment. Code has been made available at: https: //xumingze0308. github. io/projects/lstr.

PDF Details

AIIM Journal 2021 Journal Article

Multiple instance convolutional neural network with modality-based attention and contextual multi-instance learning pooling layer for effective differentiation between borderline and malignant epithelial ovarian tumors

  • Junming Jian
  • Wei Xia
  • Rui Zhang
  • Xingyu Zhao
  • Jiayi Zhang
  • Xiaodong Wu
  • Yong'ai Li
  • Jinwei Qiang

Malignant epithelial ovarian tumors (MEOTs) are the most lethal gynecologic malignancies, accounting for 90% of ovarian cancer cases. By contrast, borderline epithelial ovarian tumors (BEOTs) have low malignant potential and are generally associated with a good prognosis. Accurate preoperative differentiation between BEOTs and MEOTs is crucial for determining the appropriate surgical strategies and improving the postoperative quality of life. Multimodal magnetic resonance imaging (MRI) is an essential diagnostic tool. Although state-of-the-art artificial intelligence technologies such as convolutional neural networks can be used for automated diagnoses, their application have been limited owing to their high demand for graphics processing unit memory and hardware resources when dealing with large 3D volumetric data. In this study, we used multimodal MRI with a multiple instance learning (MIL) method to differentiate between BEOT and MEOT. We proposed the use of MAC-Net, a multiple instance convolutional neural network (MICNN) with modality-based attention (MA) and contextual MIL pooling layer (C-MPL). The MA module can learn from the decision-making patterns of clinicians to automatically perceive the importance of different MRI modalities and achieve multimodal MRI feature fusion based on their importance. The C-MPL module uses strong prior knowledge of tumor distribution as an important reference and assesses contextual information between adjacent images, thus achieving a more accurate prediction. The performance of MAC-Net is superior, with an area under the receiver operating characteristic curve of 0. 878, surpassing that of several known MICNN approaches. Therefore, it can be used to assist clinical differentiation between BEOTs and MEOTs.

Details DOI

AAAI Conference 2020 Conference Paper

Generalized Arc Consistency Algorithms for Table Constraints: A Summary of Algorithmic Ideas

  • Roland H. C. Yap
  • Wei Xia
  • Ruiwei Wang

Constraint Programming is a powerful paradigm to model and solve combinatorial problems. While there are many kinds of constraints, the table constraint (also called a CSP) is perhaps the most significant—being the most well-studied and has the ability to encode any other constraints defined on finite variables. Thus, designing efficient filtering algorithms on table constraints has attracted significant research efforts. In turn, there have been great improvements in efficiency over time with the evolution and development of AC and GAC algorithms. In this paper, we survey the existing filtering algorithms for table constraint focusing on historically important ideas and recent successful techniques shown to be effective.

PDF Details

AAAI Conference 2020 Conference Paper

Tensor-SVD Based Graph Learning for Multi-View Subspace Clustering

  • Quanxue Gao
  • Wei Xia
  • Zhizhen Wan
  • Deyan Xie
  • Pu Zhang

Low-rank representation based on tensor-Singular Value Decomposition (t-SVD) has achieved impressive results for multi-view subspace clustering, but it does not well deal with noise and illumination changes embedded in multi-view data. The major reason is that all the singular values have the same contribution in tensor-nuclear norm based on t-SVD, which does not make sense in the existence of noise and illumination change. To improve the robustness and clustering performance, we study the weighted tensor-nuclear norm based on t-SVD and develop an efficient algorithm to optimize the weighted tensor-nuclear norm minimization (WTNNM) problem. We further apply the WTNNM algorithm to multiview subspace clustering by exploiting the high order correlations embedded in different views. Extensive experimental results reveal that our WTNNM method is superior to several state-of-the-art multi-view subspace clustering methods in terms of performance.

PDF Details

AAAI Conference 2018 Conference Paper

Learning Robust Search Strategies Using a Bandit-Based Approach

  • Wei Xia
  • Roland Yap

Effective solving of constraint problems often requires choosing good or specific search heuristics. However, choosing or designing a good search heuristic is non-trivial and is often a manual process. In this paper, rather than manually choosing/designing search heuristics, we propose the use of bandit-based learning techniques to automatically select search heuristics. Our approach is online where the solver learns and selects from a set of heuristics during search. The goal is to obtain automatic search heuristics which give robust performance. Preliminary experiments show that our adaptive technique is more robust than the original search heuristics. It can also outperform the original heuristics.

PDF Details

IJCAI Conference 2016 Conference Paper

Optimizing Simple Tabular Reduction with a Bitwise Representation

  • Ruiwei Wang
  • Wei Xia
  • Roland H. C. Yap
  • Zhanshan Li

Maintaining Generalized Arc Consistency (GAC) during search is considered an efficient way to solve non-binary constraint satisfaction problems. Bit-based representations have been used effectively in Arc Consistency algorithms. We propose STRbit, a GAC algorithm, based on simple tabular reduction (STR) using an efficient bit vector support data structure. STRbit is extended to deal with compression of the underlying constraint with c-tuples. Experimental evaluation show our algorithms are faster than many algorithms (STR2, STR2-C, STR3, STR3-C and MDDc) across a variety of benchmarks except for problems with small tables where complex data structures do not payoff.

PDF Details

IJCAI Conference 2015 Conference Paper

Decomposition of the Factor Encoding for CSPs

  • Chavalit Likitvivatanavong
  • Wei Xia
  • Roland H. C. Yap

Generalized arc consistency (GAC) is one of the most fundamental properties for reducing the search space when solving constraint satisfaction problems (CSPs). Consistencies stronger than GAC have also been shown useful, but the challenge is to develop efficient and simple filtering algorithms. Several CSP transformations are proposed recently so that the GAC algorithms can be applied on the transformed CSP to enforce stronger consistencies. Among them, the factor encoding (FE) is shown to be promising with respect to recent higher-order consistency algorithms. Nonetheless, one potential drawback of the FE is the fact that it enlarges the table relations as it increases constraint arity. We propose a variation of the FE that aims at reducing redundant columns in the constraints of the FE while still preserving full pairwise consistency. Experiments show that the new approach is competitive over a variety of random and structured benchmarks.

PDF Details