arXiv subjects

Xiaoyan Zhu

Publications and source records attributed to Xiaoyan Zhu.

At least 19 recordsLinked to original sources

arXiv2026

Do Papers Tell the Whole Story? A Benchmark and Framework for Uncovering Hidden Implementation Gaps in Bioinformatics

As bioinformatics software is increasingly applied across a broader range of scenarios and the rapid development of large language models (LLMs) further lowers the barriers to software use and development, the composition of the bioinformatics research community is undergoing substantial change. Consequently, a growing number of researchers require a deeper understanding of methodological details and software behavior. In this context, systematically analyzing the relationship between paper descriptions and code implementations is emerging as an important new challenge in the field. To address this challenge, we introduce paper-code consistency analysis as a new research perspective and construct BioCon, the first benchmark dataset for paper-code consistency analysis in bioinformatics. Furthermore, we develop a unified cross-modal analysis framework to systematically investigate this problem from three perspectives: sentence-level detection, cross-modal retrieval, and project-level assessment. Experimental results demonstrate that the proposed framework can effectively model the semantic relationships between scientific publications and software implementations. Further case studies reveal that paper-code inconsistency is not a single phenomenon but arises from multiple underlying causes, among which Author-Perceived Non-Essential Details represents the most prevalent category. These findings suggest that paper-code consistency analysis is not merely a technical problem but also raises broader discussions regarding knowledge dissemination, the boundaries of code disclosure, and community norms. We hope that this work will encourage the bioinformatics community to re-examine the relationship between scientific publications and software implementations while providing a foundation for future research in paper-code consistency analysis.

cs.LG↗

arXiv2026

Ensemble Learning Based Classification Algorithm Recommendation

Selecting an appropriate classification algorithm for a given data set remains a challenging problem in data mining and machine learning. Existing algorithm recommendation models are typically trained with individual learners and rely on only one type of meta-feature, which may limit their ability to capture the diverse characteristics of classification problems. This paper proposes a multi-view ensemble meta-learning framework for classification algorithm recommendation. The framework constructs base recommendation models from different combinations of heterogeneous meta-feature groups and combines them through an accuracy- and diversity-aware ensemble strategy. The main focus of this work is empirical: we evaluate the proposed method on 1,090 benchmark classification problems derived from 84 public data sets, using 13 widely used candidate classification algorithms and five types of meta-features. The experimental results show that the proposed ensemble recommendation method consistently improves ranking loss, average precision, and top-ranked recommendation precision over individual recommendation models. These results suggest that combining complementary meta-feature views is an effective strategy for robust classification algorithm recommendation.

cs.IR↗

arXiv2026

Regret Pre-training: Bridging Prior and Posterior Views for Enhanced Knowledge Grounding

Causal language models factorize sequence probabilities using only preceding context, leaving future information unexploited during training despite its availability in the training data. This paper introduces Regret Pre-training, a self-supervised framework grounded in the Learning Using Privileged Information (LUPI) paradigm. The framework employs a dual-view architecture in which a single model generates both a causal Student distribution and a future-conditioned Teacher distribution. The training objective augments standard language modeling with a regret loss that minimizes the KL divergence from teacher to student, transferring future-aware signals to the causal representations. We investigate two teacher configurations on the OLMoE-1B-7B architecture:LocalRegret, which extends attention by one future token, andGlobalRegret, which conditions on bidirectional context with the target position masked. Experiments on nine downstream tasks following 4 billion tokens of training demonstrate that both configurations consistently outperform the baseline. On average,GlobalRegret andLocalRegret achieve 33.9% and 32.2% accuracy respectively, surpassing the baseline's 30.2%. Most notably,GlobalRegret improves BoolQ performance by 18.1 percentage points (61.0% vs 42.9%). The framework introduces no additional parameters and requires only one extra inference-mode forward pass per training step.

cs.CL↗

arXiv2026

BioDefect: The First Dataset for Defect Detection in Bioinformatics Software

Software defect detection is a critical task in software engineering. However, no prior studies have specifically addressed defect detection in bioinformatics software. Given that the performance of defect detection tasks is primarily influenced by both models and datasets, our experiments controlled for model-related factors and confirmed the limitations of existing datasets in bioinformatics software. To address this issue, we introduce BioDefect, the first dataset specifically designed for defect detection in bioinformatics software, aiming to overcome the limitations of existing datasets in this context. Unlike prior datasets, BioDefect includes complete source code repositories, preserving the actual contextual information of defective code, thereby more accurately reflecting real-world defect scenarios in bioinformatics software. Additionally, BioDefect mitigates issues related to label inconsistency and data leakage, ensuring high data quality and experimental reliability. To evaluate the effectiveness of BioDefect, we conduct a systematic assessment on nine language models (LMs), including DeepSeek-R1. The results demonstrate that BioDefect significantly enhances defect detection performance for bioinformatics software. Compared to existing datasets, BioDefect achieves an average F1-score improvement of 29.61% to 38.04% across all models, highlighting its superior advantages. This study fills a critical research gap in bioinformatics software defect detection, laying a foundation for future studies in this field and offering new insights for improving bioinformatics software quality assurance.

cs.SE↗

arXiv2026

ClaHF: A Human Feedback-inspired Reinforcement Learning Framework for Improving Classification Tasks

Text classification models are typically trained via supervised fine-tuning (SFT). However, SFT essentially performs behavior cloning from instance-wise labels and thus fails to adequately capture relative preference relations among samples, which limits the model's ability to shape decision boundaries and calibrate predictive confidence. In this paper, we propose ClaHF, a human feedback-inspired reinforcement learning (RL) framework for text classification that integrates preference modeling and RL optimization into the classification pipeline without requiring additional human annotations. Unlike prior work that relies solely on instance-wise supervision, ClaHF constructs multiple candidate predictions together with their relative ranking relations, and jointly models the Top-1 preference and the ordering among non-optimal candidates within a reward model (RM). This design converts conventional label supervision into preference signals that are directly applicable to policy optimization. We conduct systematic evaluations on eight classification tasks spanning three categories of scenarios. Results demonstrate that ClaHF consistently improves both classification performance and confidence calibration across diverse language models (LMs). The data and code are available at https://anonymous.4open.science/r/ClaHF.

cs.LG↗

arXiv2026

Investigating the intrinsic anomalous Hall effect in MnPt3 topological semimetal

The cubic Cu$_3$Au-type $X$Pt$_3$ family ($X$ = V, Cr, and Mn) is a topological semimetal characterized by anti-crossing gapped nodal lines near the Fermi level, which give rise to significant Berry curvatures and thus to the anomalous Hall effect (AHE). Among the three members, CrPt$_3$ has been experimentally verified to exhibit a large anomalous Hall conductivity (AHC), while its counterparts MnPt$_3$ and VPt$_3$ remain largely unexplored. Here, a series of MnPt$_3$ thin films with varying thicknesses (20--70 nm) was epitaxially grown on the MgO substrates using magnetron sputtering and was systematically investigated by magnetization, electrical resistivity, and Hall resistivity measurements. MnPt$_3$ films undergo a ferromagnetic transition at a Curie temperature $T_\mathrm{C}$, which increases as the film thickness increases, reaching $\sim$ 344 K for the 70-nm-thick film. All the anomalous Hall transport properties of MnPt$_3$ films, including the resistivity, conductivity, and angle, exhibit a strong correlation with their magnetic properties. The scaling analysis suggests that the intrinsic Berry-curvature mechanism dominates the observed AHE, while the extrinsic contributions are much smaller. The intrinsic AHC increases as the film thickness increases, while the extrinsic AHC is thickness-independent. Such an enhanced intrinsic AHC in the MnPt$_3$ films is most likely attributed to the strain effect, implying that it serves as an effective method to tune the electronic band topology in the $X$Pt$_3$ topological semimetal.

cond-mat.str-el↗

arXiv2026

Large Language Models for Limited Noisy Data: A Gravitational Wave Identification Study

This work investigates whether large language models (LLMs) offer advantages over traditional neural networks for astronomical data processing, in regimes with non-Gaussian, non-stationary noise and limited labeled samples. Gravitational wave observations provide an suitable test case, using only 90 LIGO events, finetuned LLMs achieve 97.4\% accuracy for identifying signals. Further experiments show that, in contrast to traditional networks that rely on large simulated datasets, additional simulated samples do not improve LLM performance, while scaling studies reveal predictable gains with increasing model size and dataset size. These results indicate that LLMs can extract discriminative structure directly from observational data and provide an efficient assessment for gravitational wave identification. The same strategy may extend to other astronomical domains with similar noise properties, such as radio or pulsar observations.

astro-ph.IM↗

arXiv2025

Making Every Head Count: Sparse Attention Without the Speed-Performance Trade-off

The design of Large Language Models (LLMs) has long been hampered by a fundamental conflict within their core attention mechanism: its remarkable expressivity is built upon a computational complexity of O(H N^2) that grows quadratically with the context size (N) and linearly with the number of heads (H). This standard implementation harbors significant computational redundancy, as all heads independently compute attention over the same sequence space. Existing sparse methods, meanwhile, often trade information integrity for computational efficiency. To resolve this efficiency-performance trade-off, we propose SPAttention, whose core contribution is the introduction of a new paradigm we term Principled Structural Sparsity. SPAttention does not merely drop connections but instead reorganizes the computational task by partitioning the total attention workload into balanced, non-overlapping distance bands, assigning each head a unique segment. This approach transforms the multi-head attention mechanism from H independent O(N^2) computations into a single, collaborative O(N^2) computation, fundamentally reducing complexity by a factor of H. The structured inductive bias compels functional specialization among heads, enabling a more efficient allocation of computational resources from redundant modeling to distinct dependencies across the entire sequence span. Our work demonstrates that thoughtfully designed structural sparsity can serve as an effective inductive bias that simultaneously improves both computational efficiency and model performance, opening a new avenue for the architectural design of next-generation, high-performance LLMs.

cs.LG↗

arXiv2025

Mosaic Pruning: A Hierarchical Framework for Generalizable Pruning of Mixture-of-Experts Models

Sparse Mixture-of-Experts (SMoE) architectures have enabled a new frontier in scaling Large Language Models (LLMs), offering superior performance by activating only a fraction of their total parameters during inference. However, their practical deployment is severely hampered by substantial static memory overhead, as all experts must be loaded into memory. Existing post-training pruning methods, while reducing model size, often derive their pruning criteria from a single, general-purpose corpus. This leads to a critical limitation: a catastrophic performance degradation when the pruned model is applied to other domains, necessitating a costly re-pruning for each new domain. To address this generalization gap, we introduce Mosaic Pruning (MoP). The core idea of MoP is to construct a functionally comprehensive set of experts through a structured ``cluster-then-select" process. This process leverages a similarity metric that captures expert performance across different task domains to functionally cluster the experts, and subsequently selects the most representative expert from each cluster based on our proposed Activation Variability Score. Unlike methods that optimize for a single corpus, our proposed Mosaic Pruning ensures that the pruned model retains a functionally complementary set of experts, much like the tiles of a mosaic that together form a complete picture of the original model's capabilities, enabling it to handle diverse downstream tasks.Extensive experiments on various MoE models demonstrate the superiority of our approach. MoP significantly outperforms prior work, achieving a 7.24\% gain on general tasks and 8.92\% on specialized tasks like math reasoning and code generation.

cs.LG↗

arXiv2025

Large unconventional anomalous Hall effect far above room temperature in epitaxial Fe$_3$Ga$_4$ films

Noncoplanar spin textures usually exhibit a finite scalar spin chirality (SSC) that can generate effective magnetic fields and lead to additional contributions to the Hall effect, namely topological or unconventional anomalous Hall effect (UAHE). Unlike topological spin textures (e.g., magnetic skyrmions), materials that exhibit fluctuation-driven SSC and UAHE are rare. So far, their realization has been limited to either low temperatures or high magnetic fields, both of which are unfavorable for practical applications. Identifying new materials that exhibit UAHE in a low magnetic field at room temperature is therefore essential. Here, we report the discovery of a large UAHE far above room temperature in epitaxial Fe$_3$Ga$_4$ films, where the fluctuation-driven SSC stems from the field-induced transverse-conical-spiral phase. Considering their epitaxial nature and the large UAHE stabilized at room temperature in a low magnetic field, Fe$_3$Ga$_4$ films represent an exciting, albeit rare, case of a promising candidate material for spintronic devices.

cond-mat.mtrl-sci↗

arXiv2023

EVA2.0: Investigating Open-Domain Chinese Dialogue Systems with Large-Scale Pre-Training

Large-scale pre-training has shown remarkable performance in building open-domain dialogue systems. However, previous works mainly focus on showing and evaluating the conversational performance of the released dialogue model, ignoring the discussion of some key factors towards a powerful human-like chatbot, especially in Chinese scenarios. In this paper, we conduct extensive experiments to investigate these under-explored factors, including data quality control, model architecture designs, training approaches, and decoding strategies. We propose EVA2.0, a large-scale pre-trained open-domain Chinese dialogue model with 2.8 billion parameters, and will make our models and codes publicly available. Automatic and human evaluations show that EVA2.0 significantly outperforms other open-source counterparts. We also discuss the limitations of this work by presenting some failure cases and pose some future research directions on large-scale Chinese open-domain dialogue systems.

cs.CL↗

arXiv2023

Absence of topological Hall effect in Fe$_x$Rh$_{100-x}$ epitaxial films: revisiting their phase diagram

A series of Fe$_x$Rh$_{100-x}$ ($30 \leq x \leq 57$) films were epitaxially grown using magnetron sputtering, and were systematically studied by magnetization-, electrical resistivity-, and Hall resistivity measurements. After optimizing the growth conditions, phase-pure Fe$_{x}$Rh$_{100-x}$ films were obtained, and their magnetic phase diagram was revisited. The ferromagnetic (FM) to antiferromagnetic (AFM) transition is limited at narrow Fe-contents with $48 \leq x \leq 54$ in the bulk Fe$_x$Rh$_{100-x}$ alloys. By contrast, the FM-AFM transition in the Fe$_x$Rh$_{100-x}$ films is extended to cover a much wider $x$ range between 33 % and 53 %, whose critical temperature slightly decreases as increasing the Fe-content. The resistivity jump and magnetization drop at the FM-AFM transition are much more significant in the Fe$_x$Rh$_{100-x}$ films with $\sim$50 % Fe-content than in the Fe-deficient films, the latter have a large amount of paramagnetic phase. The magnetoresistivity (MR) is rather weak and positive in the AFM state, while it becomes negative when the FM phase shows up, and a giant MR appears in the mixed FM- and AFM states. The Hall resistivity is dominated by the ordinary Hall effect in the AFM state, while in the mixed state or high-temperature FM state, the anomalous Hall effect takes over. The absence of topological Hall resistivity in Fe$_{x}$Rh$_{100-x}$ films with various Fe-contents implies that the previously observed topological Hall effect is most likely extrinsic. We propose that the anomalous Hall effect caused by the FM iron moments at the interfaces nicely explains the hump-like anomaly in the Hall resistivity. Our systematic investigations may offer valuable insights into the spintronics based on iron-rhodium alloys.

cond-mat.mtrl-sci↗

arXiv2023

DecompEval: Evaluating Generated Texts as Unsupervised Decomposed Question Answering

Existing evaluation metrics for natural language generation (NLG) tasks face the challenges on generalization ability and interpretability. Specifically, most of the well-performed metrics are required to train on evaluation datasets of specific NLG tasks and evaluation dimensions, which may cause over-fitting to task-specific datasets. Furthermore, existing metrics only provide an evaluation score for each dimension without revealing the evidence to interpret how this score is obtained. To deal with these challenges, we propose a simple yet effective metric called DecompEval. This metric formulates NLG evaluation as an instruction-style question answering task and utilizes instruction-tuned pre-trained language models (PLMs) without training on evaluation datasets, aiming to enhance the generalization ability. To make the evaluation process more interpretable, we decompose our devised instruction-style question about the quality of generated texts into the subquestions that measure the quality of each sentence. The subquestions with their answers generated by PLMs are then recomposed as evidence to obtain the evaluation result. Experimental results show that DecompEval achieves state-of-the-art performance in untrained metrics for evaluating text summarization and dialogue generation, which also exhibits strong dimension-level / task-level generalization ability and interpretability.

cs.CL↗

arXiv2023

SoK: Comparing Different Membership Inference Attacks with a Comprehensive Benchmark

Membership inference (MI) attacks threaten user privacy through determining if a given data example has been used to train a target model. However, it has been increasingly recognized that the "comparing different MI attacks" methodology used in the existing works has serious limitations. Due to these limitations, we found (through the experiments in this work) that some comparison results reported in the literature are quite misleading. In this paper, we seek to develop a comprehensive benchmark for comparing different MI attacks, called MIBench, which consists not only the evaluation metrics, but also the evaluation scenarios. And we design the evaluation scenarios from four perspectives: the distance distribution of data samples in the target dataset, the distance between data samples of the target dataset, the differential distance between two datasets (i.e., the target dataset and a generated dataset with only nonmembers), and the ratio of the samples that are made no inferences by an MI attack. The evaluation metrics consist of ten typical evaluation metrics. We have identified three principles for the proposed "comparing different MI attacks" methodology, and we have designed and implemented the MIBench benchmark with 84 evaluation scenarios for each dataset. In total, we have used our benchmark to fairly and systematically compare 15 state-of-the-art MI attack algorithms across 588 evaluation scenarios, and these evaluation scenarios cover 7 widely used datasets and 7 representative types of models. All codes and evaluations of MIBench are publicly available at https://github.com/MIBench/MIBench.github.io/blob/main/README.md.

cs.CR↗

arXiv2022

CTRLEval: An Unsupervised Reference-Free Metric for Evaluating Controlled Text Generation

Existing reference-free metrics have obvious limitations for evaluating controlled text generation models. Unsupervised metrics can only provide a task-agnostic evaluation result which correlates weakly with human judgments, whereas supervised ones may overfit task-specific data with poor generalization ability to other datasets. In this paper, we propose an unsupervised reference-free metric called CTRLEval, which evaluates controlled text generation from different aspects by formulating each aspect into multiple text infilling tasks. On top of these tasks, the metric assembles the generation probabilities from a pre-trained language model without any model training. Experimental results show that our metric has higher correlations with human judgments than other baselines, while obtaining better generalization of evaluating generated texts from different models and with different qualities.

cs.CL↗

arXiv2022

KPT: Keyword-guided Pre-training for Grounded Dialog Generation

Incorporating external knowledge into the response generation process is essential to building more helpful and reliable dialog agents. However, collecting knowledge-grounded conversations is often costly, calling for a better pre-trained model for grounded dialog generation that generalizes well w.r.t. different types of knowledge. In this work, we propose KPT (Keyword-guided Pre-Training), a novel self-supervised pre-training method for grounded dialog generation without relying on extra knowledge annotation. Specifically, we use a pre-trained language model to extract the most uncertain tokens in the dialog as keywords. With these keywords, we construct two kinds of knowledge and pre-train a knowledge-grounded response generation model, aiming at handling two different scenarios: (1) the knowledge should be faithfully grounded; (2) it can be selectively used. For the former, the grounding knowledge consists of keywords extracted from the response. For the latter, the grounding knowledge is additionally augmented with keywords extracted from other utterances in the same dialog. Since the knowledge is extracted from the dialog itself, KPT can be easily performed on a large volume and variety of dialogue data. We considered three data sources (open-domain, task-oriented, conversational QA) with a total of 2.5M dialogues. We conduct extensive experiments on various few-shot knowledge-grounded generation tasks, including grounding on dialog acts, knowledge graphs, persona descriptions, and Wikipedia passages. Our comprehensive experiments and analyses demonstrate that KPT consistently outperforms state-of-the-art methods on these tasks with diverse grounding knowledge.

cs.CL↗

arXiv2022

Learning Instructions with Unlabeled Data for Zero-Shot Cross-Task Generalization

Training language models to learn from human instructions for zero-shot cross-task generalization has attracted much attention in NLP communities. Recently, instruction tuning (IT), which fine-tunes a pre-trained language model on a massive collection of tasks described via human-craft instructions, has been shown effective in instruction learning for unseen tasks. However, IT relies on a large amount of human-annotated samples, which restricts its generalization. Unlike labeled data, unlabeled data are often massive and cheap to obtain. In this work, we study how IT can be improved with unlabeled data. We first empirically explore the IT performance trends versus the number of labeled data, instructions, and training tasks. We find it critical to enlarge the number of training instructions, and the instructions can be underutilized due to the scarcity of labeled data. Then, we propose Unlabeled Data Augmented Instruction Tuning (UDIT) to take better advantage of the instructions during IT by constructing pseudo-labeled data from unlabeled plain texts. We conduct extensive experiments to show UDIT's effectiveness in various scenarios of tasks and datasets. We also comprehensively analyze the key factors of UDIT to investigate how to better improve IT with unlabeled data. The code is publicly available at https://github.com/thu-coai/UDIT.

cs.CL↗

arXiv2022

Curriculum-Based Self-Training Makes Better Few-Shot Learners for Data-to-Text Generation

Despite the success of text-to-text pre-trained models in various natural language generation (NLG) tasks, the generation performance is largely restricted by the number of labeled data in downstream tasks, particularly in data-to-text generation tasks. Existing works mostly utilize abundant unlabeled structured data to conduct unsupervised pre-training for task adaption, which fail to model the complex relationship between source structured data and target texts. Thus, we introduce self-training as a better few-shot learner than task-adaptive pre-training, which explicitly captures this relationship via pseudo-labeled data generated by the pre-trained model. To alleviate the side-effect of low-quality pseudo-labeled data during self-training, we propose a novel method called Curriculum-Based Self-Training (CBST) to effectively leverage unlabeled data in a rearranged order determined by the difficulty of text generation. Experimental results show that our method can outperform fine-tuning and task-adaptive pre-training methods, and achieve state-of-the-art performance in the few-shot setting of data-to-text generation.

cs.CL↗