arXiv subjects

Huijun Liu

Publications and source records attributed to Huijun Liu.

At least 19 recordsLinked to original sources

arXiv2025

Identifying Knowledge Editing Types in Large Language Models

Knowledge editing has emerged as an efficient technique for updating the knowledge of large language models (LLMs), attracting increasing attention in recent years. However, there is a lack of effective measures to prevent the malicious misuse of this technique, which could lead to harmful edits in LLMs. These malicious modifications could cause LLMs to generate toxic content, misleading users into inappropriate actions. In front of this risk, we introduce a new task, $\textbf{K}$nowledge $\textbf{E}$diting $\textbf{T}$ype $\textbf{I}$dentification (KETI), aimed at identifying different types of edits in LLMs, thereby providing timely alerts to users when encountering illicit edits. As part of this task, we propose KETIBench, which includes five types of harmful edits covering the most popular toxic types, as well as one benign factual edit. We develop five classical classification models and three BERT-based models as baseline identifiers for both open-source and closed-source LLMs. Our experimental results, across 92 trials involving four models and three knowledge editing methods, demonstrate that all eight baseline identifiers achieve decent identification performance, highlighting the feasibility of identifying malicious edits in LLMs. Additional analyses reveal that the performance of the identifiers is independent of the reliability of the knowledge editing methods and exhibits cross-domain generalization, enabling the identification of edits from unknown sources. All data and code are available in https://github.com/xpq-tech/KETI.

cs.CL↗

arXiv2025

SWEA: Updating Factual Knowledge in Large Language Models via Subject Word Embedding Altering

The general capabilities of large language models (LLMs) make them the infrastructure for various AI applications, but updating their inner knowledge requires significant resources. Recent model editing is a promising technique for efficiently updating a small amount of knowledge of LLMs and has attracted much attention. In particular, local editing methods, which directly update model parameters, are proven suitable for updating small amounts of knowledge. Local editing methods update weights by computing least squares closed-form solutions and identify edited knowledge by vector-level matching in inference, which achieve promising results. However, these methods still require a lot of time and resources to complete the computation. Moreover, vector-level matching lacks reliability, and such updates disrupt the original organization of the model's parameters. To address these issues, we propose a detachable and expandable Subject Word Embedding Altering (SWEA) framework, which finds the editing embeddings through token-level matching and adds them to the subject word embeddings in Transformer input. To get these editing embeddings, we propose optimizing then suppressing fusion method, which first optimizes learnable embedding vectors for the editing target and then suppresses the Knowledge Embedding Dimensions (KEDs) to obtain final editing embeddings. We thus propose SWEA$\oplus$OS method for editing factual knowledge in LLMs. We demonstrate the overall state-of-the-art (SOTA) performance of SWEA$\oplus$OS on the CounterFact and zsRE datasets. To further validate the reasoning ability of SWEA$\oplus$OS in editing knowledge, we evaluate it on the more complex RippleEdits benchmark. The results demonstrate that SWEA$\oplus$OS possesses SOTA reasoning ability.

cs.CL↗

arXiv2024

Generalization-Enhanced Code Vulnerability Detection via Multi-Task Instruction Fine-Tuning

Code Pre-trained Models (CodePTMs) based vulnerability detection have achieved promising results over recent years. However, these models struggle to generalize as they typically learn superficial mapping from source code to labels instead of understanding the root causes of code vulnerabilities, resulting in poor performance in real-world scenarios beyond the training instances. To tackle this challenge, we introduce VulLLM, a novel framework that integrates multi-task learning with Large Language Models (LLMs) to effectively mine deep-seated vulnerability features. Specifically, we construct two auxiliary tasks beyond the vulnerability detection task. First, we utilize the vulnerability patches to construct a vulnerability localization task. Second, based on the vulnerability features extracted from patches, we leverage GPT-4 to construct a vulnerability interpretation task. VulLLM innovatively augments vulnerability classification by leveraging generative LLMs to understand complex vulnerability patterns, thus compelling the model to capture the root causes of vulnerabilities rather than overfitting to spurious features of a single task. The experiments conducted on six large datasets demonstrate that VulLLM surpasses seven state-of-the-art models in terms of effectiveness, generalization, and robustness.

cs.CR↗

arXiv2023

Deep Domain Adaptation for Pavement Crack Detection

Deep learning-based pavement cracks detection methods often require large-scale labels with detailed crack location information to learn accurate predictions. In practice, however, crack locations are very difficult to be manually annotated due to various visual patterns of pavement crack. In this paper, we propose a Deep Domain Adaptation-based Crack Detection Network (DDACDN), which learns domain invariant features by taking advantage of the source domain knowledge to predict the multi-category crack location information in the target domain, where only image-level labels are available. Specifically, DDACDN first extracts crack features from both the source and target domain by a two-branch weights-shared backbone network. And in an effort to achieve the cross-domain adaptation, an intermediate domain is constructed by aggregating the three-scale features from the feature space of each domain to adapt the crack features from the source domain to the target domain. Finally, the network involves the knowledge of both domains and is trained to recognize and localize pavement cracks. To facilitate accurate training and validation for domain adaptation, we use two challenging pavement crack datasets CQU-BPDD and RDD2020. Furthermore, we construct a new large-scale Bituminous Pavement Multi-label Disease Dataset named CQU-BPMDD, which contains 38994 high-resolution pavement disease images to further evaluate the robustness of our model. Extensive experiments demonstrate that DDACDN outperforms state-of-the-art pavement crack detection methods in predicting the crack location on the target domain.

cs.CV↗

arXiv2022

Span-based joint entity and relation extraction augmented with sequence tagging mechanism

Span-based joint extraction simultaneously conducts named entity recognition (NER) and relation extraction (RE) in text span form. However, since previous span-based models rely on span-level classifications, they cannot benefit from token-level label information, which has been proven advantageous for the task. In this paper, we propose a Sequence Tagging augmented Span-based Network (STSN), a span-based joint model that can make use of token-level label information. In STSN, we construct a core neural architecture by deep stacking multiple attention layers, each of which consists of three basic attention units. On the one hand, the core architecture enables our model to learn token-level label information via the sequence tagging mechanism and then uses the information in the span-based joint extraction; on the other hand, it establishes a bi-directional information interaction between NER and RE. Experimental results on three benchmark datasets show that STSN consistently outperforms the strongest baselines in terms of F1, creating new state-of-the-art results.

cs.CL↗

arXiv2022

A Two-Phase Paradigm for Joint Entity-Relation Extraction

An exhaustive study has been conducted to investigate span-based models for the joint entity and relation extraction task. However, these models sample a large number of negative entities and negative relations during the model training, which are essential but result in grossly imbalanced data distributions and in turn cause suboptimal model performance. In order to address the above issues, we propose a two-phase paradigm for the span-based joint entity and relation extraction, which involves classifying the entities and relations in the first phase, and predicting the types of these entities and relations in the second phase. The two-phase paradigm enables our model to significantly reduce the data distribution gap, including the gap between negative entities and other entities, as well as the gap between negative relations and other relations. In addition, we make the first attempt at combining entity type and entity distance as global features, which has proven effective, especially for the relation extraction. Experimental results on several datasets demonstrate that the spanbased joint extraction model augmented with the two-phase paradigm and the global features consistently outperforms previous state-of-the-art span-based models for the joint extraction task, establishing a new standard benchmark. Qualitative and quantitative analyses further validate the effectiveness the proposed paradigm and the global features.

cs.CL↗

arXiv2022

Few-shot Named Entity Recognition with Entity-level Prototypical Network Enhanced by Dispersedly Distributed Prototypes

Few-shot named entity recognition (NER) enables us to build a NER system for a new domain using very few labeled examples. However, existing prototypical networks for this task suffer from roughly estimated label dependency and closely distributed prototypes, thus often causing misclassifications. To address the above issues, we propose EP-Net, an Entity-level Prototypical Network enhanced by dispersedly distributed prototypes. EP-Net builds entity-level prototypes and considers text spans to be candidate entities, so it no longer requires the label dependency. In addition, EP-Net trains the prototypes from scratch to distribute them dispersedly and aligns spans to prototypes in the embedding space using a space projection. Experimental results on two evaluation tasks and the Few-NERD settings demonstrate that EP-Net consistently outperforms the previous strong models in terms of overall performance. Extensive analyses further validate the effectiveness of EP-Net.

cs.CL↗

arXiv2022

A Context-Aware Approach for Textual Adversarial Attack through Probability Difference Guided Beam Search

Textual adversarial attacks expose the vulnerabilities of text classifiers and can be used to improve their robustness. Existing context-aware methods solely consider the gold label probability and use the greedy search when searching an attack path, often limiting the attack efficiency. To tackle these issues, we propose PDBS, a context-aware textual adversarial attack model using Probability Difference guided Beam Search. The probability difference is an overall consideration of all class label probabilities, and PDBS uses it to guide the selection of attack paths. In addition, PDBS uses the beam search to find a successful attack path, thus avoiding suffering from limited search space. Extensive experiments and human evaluation demonstrate that PDBS outperforms previous best models in a series of evaluation metrics, especially bringing up to a +19.5% attack success rate. Ablation studies and qualitative analyses further confirm the efficiency of PDBS.

cs.CL↗

arXiv2022

Boosting Span-based Joint Entity and Relation Extraction via Squence Tagging Mechanism

Span-based joint extraction simultaneously conducts named entity recognition (NER) and relation extraction (RE) in text span form. Recent studies have shown that token labels can convey crucial task-specific information and enrich token semantics. However, as far as we know, due to completely abstain from sequence tagging mechanism, all prior span-based work fails to use token label in-formation. To solve this problem, we pro-pose Sequence Tagging enhanced Span-based Network (STSN), a span-based joint extrac-tion network that is enhanced by token BIO label information derived from sequence tag-ging based NER. By stacking multiple atten-tion layers in depth, we design a deep neu-ral architecture to build STSN, and each atten-tion layer consists of three basic attention units. The deep neural architecture first learns seman-tic representations for token labels and span-based joint extraction, and then constructs in-formation interactions between them, which also realizes bidirectional information interac-tions between span-based NER and RE. Fur-thermore, we extend the BIO tagging scheme to make STSN can extract overlapping en-tity. Experiments on three benchmark datasets show that our model consistently outperforms previous optimal models by a large margin, creating new state-of-the-art results.

cs.CL↗

arXiv2022

Win-Win Cooperation: Bundling Sequence and Span Models for Named Entity Recognition

For Named Entity Recognition (NER), sequence labeling-based and span-based paradigms are quite different. Previous research has demonstrated that the two paradigms have clear complementary advantages, but few models have attempted to leverage these advantages in a single NER model as far as we know. In our previous work, we proposed a paradigm known as Bundling Learning (BL) to address the above problem. The BL paradigm bundles the two NER paradigms, enabling NER models to jointly tune their parameters by weighted summing each paradigm's training loss. However, three critical issues remain unresolved: When does BL work? Why does BL work? Can BL enhance the existing state-of-the-art (SOTA) NER models? To address the first two issues, we implement three NER models, involving a sequence labeling-based model--SeqNER, a span-based NER model--SpanNER, and BL-NER that bundles SeqNER and SpanNER together. We draw two conclusions regarding the two issues based on the experimental results on eleven NER datasets from five domains. We then apply BL to five existing SOTA NER models to investigate the third issue, consisting of three sequence labeling-based models and two span-based models. Experimental results indicate that BL consistently enhances their performance, suggesting that it is possible to construct a new SOTA NER system by incorporating BL into the current SOTA system. Moreover, we find that BL reduces both entity boundary and type prediction errors. In addition, we compare two commonly used labeling tagging methods as well as three types of span semantic representations.

cs.CL↗

arXiv2022

Topic-Grained Text Representation-based Model for Document Retrieval

Document retrieval enables users to find their required documents accurately and quickly. To satisfy the requirement of retrieval efficiency, prevalent deep neural methods adopt a representation-based matching paradigm, which saves online matching time by pre-storing document representations offline. However, the above paradigm consumes vast local storage space, especially when storing the document as word-grained representations. To tackle this, we present TGTR, a Topic-Grained Text Representation-based Model for document retrieval. Following the representation-based matching paradigm, TGTR stores the document representations offline to ensure retrieval efficiency, whereas it significantly reduces the storage requirements by using novel topicgrained representations rather than traditional word-grained. Experimental results demonstrate that compared to word-grained baselines, TGTR is consistently competitive with them on TREC CAR and MS MARCO in terms of retrieval accuracy, but it requires less than 1/10 of the storage space required by them. Moreover, TGTR overwhelmingly surpasses global-grained baselines in terms of retrieval accuracy.

cs.IR↗

arXiv2020

Screening potential topological insulators in half-Heusler compounds via compressed-sensing

Ternary half-Heusler compounds with widely tunable electronic structures, present a new platform to discover topological insulators. Due to time-consuming computations and synthesis procedures, the identification of new topological insulators is however a rough task. Here, we adopt a compressed-sensing approach to rapidly screen potential topological insulators in half-Heusler family, which is realized via a two-dimensional descriptor that only depends on the fundamental properties of the constituent atoms. Beyond the finite training data, the proposed descriptor is employed to screen many new half-Heusler compounds, including those with integer and fractional stoichiometry, and a larger number of possible topological insulators are predicted.

cond-mat.mtrl-sci↗

arXiv2020

Artificial Intelligence for High-Throughput Discovery of Topological Insulators: the Example of Alloyed Tetradymites

Significant advances have been made in predicting new topological materials using high-throughput empirical descriptors or symmetry-based indicators. To date, these approaches have been applied to materials in existing databases, and are severely limited to systems with well-defined symmetries, leaving a much larger materials space unexplored. Using tetradymites as a prototypical class of examples, we uncover a novel two-dimensional descriptor by applying an artificial intelligence (AI) based approach for fast and reliable identification of the topological characters of a drastically expanded range of materials, without prior determination of their specific symmetries and detailed band structures. By leveraging this descriptor that contains only the atomic number and electronegativity of the constituent species, we have readily scanned a huge number of alloys in the tetradymite family. Strikingly, nearly half of which are identified to be topological insulators, revealing a much larger territory of the topological materials world. The present work also attests the increasingly important role of such AI-based approaches in modern materials discovery.

cond-mat.mtrl-sci↗

arXiv2019

High thermoelectric performance of two-dimensional (PbTe)2 layer

The electronic, phonon and thermoelectric transport properties of (PbTe)2 layer are systematically investigated by using first-principles pseudopotential method and Boltzmann transport equation. Our calculations demonstrate that there is a valley degeneracy of six for the top valence band, which leads to larger carrier concentration and thus higher electrical conductivity without obvious reduction in the Seebeck coefficient. Moreover, the intrinsic van der Waals interactions between neighboring Pb layers induce additional phonon scattering and thus ultrasmall lattice thermal conductivity. As a consequence, a maximum p-type ZT value of 2.9 can be achieved at 1000 K. Moreover, we find almost identical n- and p-type ZT in the temperature range from 300 K to 800 K.

cond-mat.mtrl-sci↗

arXiv2019

Designing graphene/hexagonal boron nitride superlattice monolayer with high thermoelectric performance

We design a hybrid graphene/hexagonal boron nitride superlattice monolayer and investigate its thermoelectric properties using density functional theory and Boltzmann transport equations with the relaxation time accurately treated by electron-phonon coupling calculations. Compared with that of pristine graphene, the lattice thermal conductivity of the superlattice structure is more than two orders of magnitude lower due to the enhanced three-phonon scattering process originated from the mixed-bond characteristics. Besides, the coexistence of light and heavy bands around the Fermi level leads to an ultrahigh power factor along the zigzag direction, where the highest ZT value of ~2.5 can be achieved for the n-type system at 1100 K. Moreover, it is noted that the carrier transport near the valance band minimum is almost entirely contributed by the graphene part of the superlattice. As a consequence, the thermoelectric performance of p-type system can be enhanced to be comparable with that of n-type one by appropriate substitution of nitrogen atom with phosphorus, which can suppress the lattice thermal conductivity but nearly have no influence on the hole transport.

cond-mat.mtrl-sci↗

arXiv2019

Convolutional Neural Networks to Enhance Coded Speech

Enhancing coded speech suffering from far-end acoustic background noise, quantization noise, and potentially transmission errors, is a challenging task. In this work we propose two postprocessing approaches applying convolutional neural networks (CNNs) either in the time domain or the cepstral domain to enhance the coded speech without any modification of the codecs. The time domain approach follows an end-to-end fashion, while the cepstral domain approach uses analysis-synthesis with cepstral domain features. The proposed postprocessors in both domains are evaluated for various narrowband and wideband speech codecs in a wide range of conditions. The proposed postprocessor improves speech quality (PESQ) by up to 0.25 MOS-LQO points for G.711, 0.30 points for G.726, 0.82 points for G.722, and 0.26 points for adaptive multirate wideband codec (AMR-WB). In a subjective CCR listening test, the proposed postprocessor on G.711-coded speech exceeds the speech quality of an ITU-T-standardized postfilter by 0.36 CMOS points, and obtains a clear preference of 1.77 CMOS points compared to legacy G.711, even better than uncoded speech with statistical significance. The source code for the cepstral domain approach to enhance G.711-coded speech is made available.

eess.AS↗

arXiv2017

Large scale calculations of thermoelectric transport coefficients: a case study of γ-graphyne with point defects

Defects such as vacancies and impurities could have profound effects on the transport properties of thermoelectric materials. However, it is usually quite difficult to directly calculate the thermoelectric properties of defect-containing systems via first-principles method since very large supercell is required. In this work, based on the linear response theory and the kernel polynomial method, we present an efficient approach that can help to calculate the thermoelectric transport coefficients of a large system containing millions of atoms at arbitrary chemical potential and temperature. As a prototype example, we consider dilute vacancies and hydrogen impurities in a large scale γ-graphyne sheet and discuss their effects on the thermoelectric transport properties.

physics.comp-ph↗

arXiv2017

Effects of biaxial strain on the electronic structures and band topologies of group-V elemental monolayers

Using first-principles calculations, we systematically investigate the electronic structures and band topologies of four kinds of group-V elemental (P, As, Sb and Bi) monolayers with buckled honeycomb structure. It is found that all these monolayers can change from semiconducting to semimetallic under compressive strain. If a tensile strain is however applied, the P, As and Sb monolayers undergo phase transition from topologically trivial to non-trivial regime, whereas the topological insulating nature of Bi monolayer remains unchanged. With tunability of the band gaps and band topologies, it can be expected that these elemental monolayers could be promising candidates for future optoelectronic and spintronic applications.

cond-mat.mes-hall↗