Searcharxiv⌕ Search

arXiv subjects

Meng Xiao

Publications and source records attributed to Meng Xiao.

At least 73 records · Page 4Linked to original sources
arXiv2024

Interdisciplinary Fairness in Imbalanced Research Proposal Topic Inference: A Hierarchical Transformer-based Method with Selective Interpolation

The objective of topic inference in research proposals aims to obtain the most suitable disciplinary division from the discipline system defined by a funding agency. The agency will subsequently find appropriate peer review experts from their database based on this division. Automated topic inference can reduce human errors caused by manual topic filling, bridge the knowledge gap between funding agencies and project applicants, and improve system efficiency. Existing methods focus on modeling this as a hierarchical multi-label classification problem, using generative models to iteratively infer the most appropriate topic information. However, these methods overlook the gap in scale between interdisciplinary research proposals and non-interdisciplinary ones, leading to an unjust phenomenon where the automated inference system categorizes interdisciplinary proposals as non-interdisciplinary, causing unfairness during the expert assignment. How can we address this data imbalance issue under a complex discipline system and hence resolve this unfairness? In this paper, we implement a topic label inference system based on a Transformer encoder-decoder architecture. Furthermore, we utilize interpolation techniques to create a series of pseudo-interdisciplinary proposals from non-interdisciplinary ones during training based on non-parametric indicators such as cross-topic probabilities and topic occurrence probabilities. This approach aims to reduce the bias of the system during model training. Finally, we conduct extensive experiments on a real-world dataset to verify the effectiveness of the proposed method. The experimental results demonstrate that our training strategy can significantly mitigate the unfairness generated in the topic inference task.

cs.CL↗
arXiv2024

DP-CRE: Continual Relation Extraction via Decoupled Contrastive Learning and Memory Structure Preservation

Continuous Relation Extraction (CRE) aims to incrementally learn relation knowledge from a non-stationary stream of data. Since the introduction of new relational tasks can overshadow previously learned information, catastrophic forgetting becomes a significant challenge in this domain. Current replay-based training paradigms prioritize all data uniformly and train memory samples through multiple rounds, which would result in overfitting old tasks and pronounced bias towards new tasks because of the imbalances of the replay set. To handle the problem, we introduce the DecouPled CRE (DP-CRE) framework that decouples the process of prior information preservation and new knowledge acquisition. This framework examines alterations in the embedding space as new relation classes emerge, distinctly managing the preservation and acquisition of knowledge. Extensive experiments show that DP-CRE significantly outperforms other CRE baselines across two datasets.

cs.CL↗
arXiv2024

scInterpreter: Training Large Language Models to Interpret scRNA-seq Data for Cell Type Annotation

Despite the inherent limitations of existing Large Language Models in directly reading and interpreting single-cell omics data, they demonstrate significant potential and flexibility as the Foundation Model. This research focuses on how to train and adapt the Large Language Model with the capability to interpret and distinguish cell types in single-cell RNA sequencing data. Our preliminary research results indicate that these foundational models excel in accurately categorizing known cell types, demonstrating the potential of the Large Language Models as effective tools for uncovering new biological insights.

q-bio.GN↗
arXiv2024

Arbitrarily configurable nonlinear topological modes

Topological modes (TMs) are typically localized at boundaries, interfaces and dislocations, and exponentially decay into the bulk of a large enough lattice. Recently, the non-Hermitian skin effect has been leveraged to delocalize the wavefunctions of TMs from the boundary and thus to increase the capacity of TMs dramatically. Here, we explore the capability of nonlinearity in designing and reconfiguring the wavefunctions of TMs. With growing intensity, wavefunctions of these in-gap nonlinear TMs undergo an initial deviation from exponential decay, gradually merge into arbitrarily designable plateaus, then encompass the entire nonlinear domain, and eventually concentrate at the nonlinear boundary. Intriguingly, such extended nonlinear TMs are still robust against defects and disorders, and stable in dynamics under external excitation. Advancing the conceptual understanding of the nonlinear TMs, our results open new avenues for increasing the capacity of TMs and developing compact and reconfigurable topological devices.

quant-ph↗
arXiv2024

Temporal Inductive Path Neural Network for Temporal Knowledge Graph Reasoning

Temporal Knowledge Graph (TKG) is an extension of traditional Knowledge Graph (KG) that incorporates the dimension of time. Reasoning on TKGs is a crucial task that aims to predict future facts based on historical occurrences. The key challenge lies in uncovering structural dependencies within historical subgraphs and temporal patterns. Most existing approaches model TKGs relying on entity modeling, as nodes in the graph play a crucial role in knowledge representation. However, the real-world scenario often involves an extensive number of entities, with new entities emerging over time. This makes it challenging for entity-dependent methods to cope with extensive volumes of entities, and effectively handling newly emerging entities also becomes a significant challenge. Therefore, we propose Temporal Inductive Path Neural Network (TiPNN), which models historical information in an entity-independent perspective. Specifically, TiPNN adopts a unified graph, namely history temporal graph, to comprehensively capture and encapsulate information from history. Subsequently, we utilize the defined query-aware temporal paths on a history temporal graph to model historical path information related to queries for reasoning. Extensive experiments illustrate that the proposed model not only attains significant performance enhancements but also handles inductive settings, while additionally facilitating the provision of reasoning evidence through history temporal graphs.

cs.AI↗
arXiv2024

Braiding topology of symmetry-protected degeneracy points in non-Hermitian systems

Degeneracy points in non-Hermitian systems are of great interest. While a homotopic framework exists for understanding their behavior in the absence of symmetry, it does not apply to symmetry-protected degeneracy points with reduced codimension. In this work, utilizing algebraic topology, we provide a systematic classification of these symmetry-protected degenerate points and investigate the braid conservation rule followed by them. Using a model Hamiltonian and circuit simulation, we discover that, contrary to simple annihilation, pairwise-created symmetry-protected degeneracy points merge into a higher-order degeneracy point, which goes beyond the abelian picture. Our findings empower researchers across diverse fields to uncover new phenomena and applications harnessing symmetry-protected non-Hermitian degeneracy points.

cond-mat.mes-hall↗
arXiv2023

Resolving the Imbalance Issue in Hierarchical Disciplinary Topic Inference via LLM-based Data Augmentation

In addressing the imbalanced issue of data within the realm of Natural Language Processing, text data augmentation methods have emerged as pivotal solutions. This data imbalance is prevalent in the research proposals submitted during the funding application process. Such imbalances, resulting from the varying popularity of disciplines or the emergence of interdisciplinary studies, significantly impede the precision of downstream topic models that deduce the affiliated disciplines of these proposals. At the data level, proposals penned by experts and scientists are inherently complex technological texts, replete with intricate terminologies, which augmenting such specialized text data poses unique challenges. At the system level, this, in turn, compromises the fairness of AI-assisted reviewer assignment systems, which raises a spotlight on solving this issue. This study leverages large language models (Llama V1) as data generators to augment research proposals categorized within intricate disciplinary hierarchies, aiming to rectify data imbalances and enhance the equity of expert assignments. We first sample within the hierarchical structure to find the under-represented class. Then we designed a prompt for keyword-based research proposal generation. Our experiments attests to the efficacy of the generated data, demonstrating that research proposals produced using the prompts can effectively address the aforementioned issues and generate high quality scientific text data, thus help the model overcome the imbalanced issue.

cs.CL↗
arXiv2023

Reinforcement-Enhanced Autoregressive Feature Transformation: Gradient-steered Search in Continuous Space for Postfix Expressions

Feature transformation aims to generate new pattern-discriminative feature space from original features to improve downstream machine learning (ML) task performances. However, the discrete search space for the optimal feature explosively grows on the basis of combinations of features and operations from low-order forms to high-order forms. Existing methods, such as exhaustive search, expansion reduction, evolutionary algorithms, reinforcement learning, and iterative greedy, suffer from large search space. Overly emphasizing efficiency in algorithm design usually sacrifices stability or robustness. To fundamentally fill this gap, we reformulate discrete feature transformation as a continuous space optimization task and develop an embedding-optimization-reconstruction framework. This framework includes four steps: 1) reinforcement-enhanced data preparation, aiming to prepare high-quality transformation-accuracy training data; 2) feature transformation operation sequence embedding, intending to encapsulate the knowledge of prepared training data within a continuous space; 3) gradient-steered optimal embedding search, dedicating to uncover potentially superior embeddings within the learned space; 4) transformation operation sequence reconstruction, striving to reproduce the feature transformation solution to pinpoint the optimal feature space.

cs.LG↗
arXiv2023

Beyond Discrete Selection: Continuous Embedding Space Optimization for Generative Feature Selection

The goal of Feature Selection - comprising filter, wrapper, and embedded approaches - is to find the optimal feature subset for designated downstream tasks. Nevertheless, current feature selection methods are limited by: 1) the selection criteria of these methods are varied for different domains, making them hard to generalize; 2) the selection performance of these approaches drops significantly when processing high-dimensional feature space coupled with small sample size. In light of these challenges, we pose the question: can selected feature subsets be more robust, accurate, and input dimensionality agnostic? In this paper, we reformulate the feature selection problem as a deep differentiable optimization task and propose a new research perspective: conceptualizing discrete feature subsetting as continuous embedding space optimization. We introduce a novel and principled framework that encompasses a sequential encoder, an accuracy evaluator, a sequential decoder, and a gradient ascent optimizer. This comprehensive framework includes four important steps: preparation of features-accuracy training data, deep feature subset embedding, gradient-optimized search, and feature subset reconstruction. Specifically, we utilize reinforcement feature selection learning to generate diverse and high-quality training data and enhance generalization. By optimizing reconstruction and accuracy losses, we embed feature selection knowledge into a continuous space using an encoder-evaluator-decoder model structure. We employ a gradient ascent search algorithm to find better embeddings in the learned embedding space. Furthermore, we reconstruct feature selection solutions using these embeddings and select the feature subset with the highest performance for downstream tasks as the optimal subset.

cs.LG↗
arXiv2023

Observation of Hybrid-Order Topological Pump in a Kekule-Textured Graphene Lattice

Thouless charge pumping protocol provides an effective route for realizing topological particle transport. To date, the first-order and higher-order topological pumps, exhibiting transitions of edge-bulk-edge and corner-bulk-corner states, respectively, are observed in a variety of experimental platforms. Here, we propose a concept of hybrid-order topological pump, which involves a transition of bulk, edge, and corner states simultaneously. More specifically, we consider a Kekulé-textured graphene lattice that features a tunable phase parameter. The finite sample of zigzag boundaries, where the corner configuration is abnormal and inaccessible by repeating unit cells, hosts topological responses at both the edges and corners. The former is protected by a nonzero winding number, while the latter can be explained by a nontrivial vector Chern number. Using our skillful acoustic experiments, we verify those nontrivial boundary landmarks and visualize the consequent hybrid-order topological pump process directly. This work deepens our understanding to higher-order topological phases and broadens the scope of topological pumps.

cond-mat.mes-hall↗
arXiv2023

Spin-dependent gain and loss in photonic quantum spin Hall systems

Topological phases are greatly enriched by including non-Hermiticity. While most works focus on the topology of the eigenvalues and eigenstates, how topologically nontrivial non-Hermitian systems behave in dynamics has only drawn limited attention. Here, we consider a breathing honeycomb lattice known to emulate the quantum spin Hall effect and exhibits higher-order corner modes. We find that non-reciprocal intracell couplings introduce gain in one pseudo-spin subspace while loss with the same magnitude in the other. In addition, non-reciprocal intracell couplings can also suppress the spin mixture of the edge modes at the boundaries and delocalize the higher-order corner mode. Our findings deepen the understanding of non-Hermitian topological phases and bring in the spin degree of freedom in manipulating the dynamics in non-Hermitian systems.

cond-mat.mes-hall↗
arXiv2023

Traceable Group-Wise Self-Optimizing Feature Transformation Learning: A Dual Optimization Perspective

Feature transformation aims to reconstruct an effective representation space by mathematically refining the existing features. It serves as a pivotal approach to combat the curse of dimensionality, enhance model generalization, mitigate data sparsity, and extend the applicability of classical models. Existing research predominantly focuses on domain knowledge-based feature engineering or learning latent representations. However, these methods, while insightful, lack full automation and fail to yield a traceable and optimal representation space. An indispensable question arises: Can we concurrently address these limitations when reconstructing a feature space for a machine-learning task? Our initial work took a pioneering step towards this challenge by introducing a novel self-optimizing framework. This framework leverages the power of three cascading reinforced agents to automatically select candidate features and operations for generating improved feature transformation combinations. Despite the impressive strides made, there was room for enhancing its effectiveness and generalization capability. In this extended journal version, we advance our initial work from two distinct yet interconnected perspectives: 1) We propose a refinement of the original framework, which integrates a graph-based state representation method to capture the feature interactions more effectively and develop different Q-learning strategies to alleviate Q-value overestimation further. 2) We utilize a new optimization technique (actor-critic) to train the entire self-optimizing framework in order to accelerate the model convergence and improve the feature transformation performance. Finally, to validate the improved effectiveness and generalization capability of our framework, we perform extensive experiments and conduct comprehensive analyses.

cs.LG↗
arXiv2023

Hierarchical MixUp Multi-label Classification with Imbalanced Interdisciplinary Research Proposals

Funding agencies are largely relied on a topic matching between domain experts and research proposals to assign proposal reviewers. As proposals are increasingly interdisciplinary, it is challenging to profile the interdisciplinary nature of a proposal, and, thereafter, find expert reviewers with an appropriate set of expertise. An essential step in solving this challenge is to accurately model and classify the interdisciplinary labels of a proposal. Existing methodological and application-related literature, such as textual classification and proposal classification, are insufficient in jointly addressing the three key unique issues introduced by interdisciplinary proposal data: 1) the hierarchical structure of discipline labels of a proposal from coarse-grain to fine-grain, e.g., from information science to AI to fundamentals of AI. 2) the heterogeneous semantics of various main textual parts that play different roles in a proposal; 3) the number of proposals is imbalanced between non-interdisciplinary and interdisciplinary research. Can we simultaneously address the three issues in understanding the proposal's interdisciplinary nature? In response to this question, we propose a hierarchical mixup multiple-label classification framework, which we called H-MixUp. H-MixUp leverages a transformer-based semantic information extractor and a GCN-based interdisciplinary knowledge extractor for the first and second issues. H-MixUp develops a fused training method of Wold-level MixUp, Word-level CutMix, Manifold MixUp, and Document-level MixUp to address the third issue.

cs.CL↗
arXiv2023

The suppression of Finite Size Effect within a Few Lattices

Boundary modes localized on the boundaries of a finite-size lattice experience a finite size effect (FSE) that could result in unwanted couplings, crosstalks and formation of gaps even in topological boundary modes. It is commonly believed that the FSE decays exponentially with the size of the system and thus requires many lattices before eventually becoming negligibly small. Here we identify a special type of FSE of some boundary modes that apparently vanishes at some particular wave vectors along the boundary. Meanwhile, the number of wave vectors where the FSE vanishes equals the number of lattices across the strip. We analytically prove this type of FSE in a simple model and prove this peculiar feature. We also provide a physical system consisting of a plasmonic sphere array where this FSE is present. Our work points to the possibility of almost arbitrarily tunning of the FSE, which facilitates unprecedented manipulation of the coupling strength between modes or channels such as the integration of multiple waveguides and photonic non-abelian braiding.

physics.class-ph↗
arXiv2023

A Universal Mirror-stacking Approach for Constructing Topological Bound States in the Continuum

Bound states in the continuum (BICs) are counter-intuitive localized states with eigenvalues embedded in the continuum of extended states. Recently, nontrivial band topology is exploited to enrich the BIC physics, resulted in topological BICs (TBICs) with extraordinary robustness against perturbations or disorders. Here, we propose a simple but universal mirror-stacking approach to turn nontrivial bound states of any topological monolayer model into TBICs. Physically, the mirror-stacked bilayer Hamiltonian can be decoupled into two independent subspaces of opposite mirror parities, each of which directly inherits the energy spectrum information and band topology of the original monolayer. By tuning the interlayer couplings, the topological bound state of one subspace can move into and out of the continuum of the other subspace continuously without hybridization. As representative examples, we construct one-dimensional first-order and two-dimensional higher-order TBICs, and demonstrate them unambiguously by acoustic experiments. Our findings will expand the research implications of both topological materials and BICs.

cond-mat.mes-hall↗
arXiv2023

Hierarchical Interdisciplinary Topic Detection Model for Research Proposal Classification

The peer merit review of research proposals has been the major mechanism for deciding grant awards. However, research proposals have become increasingly interdisciplinary. It has been a longstanding challenge to assign interdisciplinary proposals to appropriate reviewers, so proposals are fairly evaluated. One of the critical steps in reviewer assignment is to generate accurate interdisciplinary topic labels for proposal-reviewer matching. Existing systems mainly collect topic labels manually generated by principal investigators. However, such human-reported labels can be non-accurate, incomplete, labor intensive, and time costly. What role can AI play in developing a fair and precise proposal reviewer assignment system? In this study, we collaborate with the National Science Foundation of China to address the task of automated interdisciplinary topic path detection. For this purpose, we develop a deep Hierarchical Interdisciplinary Research Proposal Classification Network (HIRPCN). Specifically, we first propose a hierarchical transformer to extract the textual semantic information of proposals. We then design an interdisciplinary graph and leverage GNNs for learning representations of each discipline in order to extract interdisciplinary knowledge. After extracting the semantic and interdisciplinary knowledge, we design a level-wise prediction component to fuse the two types of knowledge representations and detect interdisciplinary topic paths for each proposal. We conduct extensive experiments and expert evaluations on three real-world datasets to demonstrate the effectiveness of our proposed model.

cs.IR↗
arXiv2023

NEEDED: Introducing Hierarchical Transformer to Eye Diseases Diagnosis

With the development of natural language processing techniques(NLP), automatic diagnosis of eye diseases using ophthalmology electronic medical records (OEMR) has become possible. It aims to evaluate the condition of both eyes of a patient respectively, and we formulate it as a particular multi-label classification task in this paper. Although there are a few related studies in other diseases, automatic diagnosis of eye diseases exhibits unique characteristics. First, descriptions of both eyes are mixed up in OEMR documents, with both free text and templated asymptomatic descriptions, resulting in sparsity and clutter of information. Second, OEMR documents contain multiple parts of descriptions and have long document lengths. Third, it is critical to provide explainability to the disease diagnosis model. To overcome those challenges, we present an effective automatic eye disease diagnosis framework, NEEDED. In this framework, a preprocessing module is integrated to improve the density and quality of information. Then, we design a hierarchical transformer structure for learning the contextualized representations of each sentence in the OEMR document. For the diagnosis part, we propose an attention-based predictor that enables traceable diagnosis by obtaining disease-specific information. Experiments on the real dataset and comparison with several baseline models show the advantage and explainability of our framework.

cs.CL↗
arXiv2022

Traceable Automatic Feature Transformation via Cascading Actor-Critic Agents

Feature transformation for AI is an essential task to boost the effectiveness and interpretability of machine learning (ML). Feature transformation aims to transform original data to identify an optimal feature space that enhances the performances of a downstream ML model. Existing studies either combines preprocessing, feature selection, and generation skills to empirically transform data, or automate feature transformation by machine intelligence, such as reinforcement learning. However, existing studies suffer from: 1) high-dimensional non-discriminative feature space; 2) inability to represent complex situational states; 3) inefficiency in integrating local and global feature information. To fill the research gap, we formulate the feature transformation task as an iterative, nested process of feature generation and selection, where feature generation is to generate and add new features based on original features, and feature selection is to remove redundant features to control the size of feature space. Finally, we present extensive experiments and case studies to illustrate 24.7\% improvements in F1 scores compared with SOTAs and robustness in high-dimensional data.

cs.LG↗