Searcharxiv⌕ Search

arXiv subjects

Zhiping Zhou

Publications and source records attributed to Zhiping Zhou.

17 recordsLinked to original sources
arXiv2026

Decoupling is a Necessity: Transformation-Agnostic Decompiled Code Recovery under Optimization and Obfuscation

Reverse engineering is essential for software security analysis and vulnerability detection. Decompilation, the process of lifting binaries to high-level pseudocode, is central to this task. However, production binaries are hostile environments: aggressive compiler optimizations and adversarial obfuscation jointly mangle control structures, obscure variable intents, and disguise high-level program logic. Consequently, existing LLM-based decompilation tools frequently suffer from structural collapse and semantic hallucinations. We present ReSource, the first multi-phase LLM framework designed for transformation-agnostic source recovery. To tackle these intertwined distortions, ReSource conceptualizes the binary-to-source discrepancies into three orthogonal tiers, namely lexical, syntactic, and semantic, and decouples the recovery process accordingly. First, to ground the LLM and prevent logic drift, it retrieves empirical priors from a curated Semantic Distortion Database. Second, to resolve control-flow flattening, it integrates a lightweight predictor to reconstruct the source-level structural skeleton. Finally, a contextual lexical deduction stage refines identifiers to restore human readability. Evaluated on a massive benchmark of over 80,000 decompiled-source function pairs across three optimization levels and four obfuscation techniques, ReSource achieves an 83% Top-5 source retrieval accuracy and an average similarity score of 0.66. By maintaining robust semantic identifiability where state-of-the-art baselines (DeGPT, LLM4Decompile, and FidelityGPT) severely overfit or degrade, ReSource provides a scalable and reliable foundation for downstream security analysis.

cs.SE↗
arXiv2026

TruthLens: Object Hallucination Detection via Self-Evaluating Truthfulness Scores in LVLMs

Despite the remarkable progress of large vision language models (LVLMs), object hallucination remains a fundamental challenge that hinders their trustworthy deployment. A key finding motivates our work: real and hallucinated object tokens are clearly separable in hidden representations, yet this separability is largely lost at the language-modeling (LM) head. We propose TruthLens, a self-evaluation framework that teaches the LM head to expose a per-object truthfulness signal without any auxiliary model or additional inference cost. Concretely, a rarely-used special token is repurposed as a reference token. For each object-token position, we extract the log-probability assigned to this special token by the LM head, and define its difference from a predefined constant as the truthfulness score. The model is then fine-tuned with an MSE objective that drives scores toward 1 for real objects and 0 for hallucinated ones, while a divergence constraint preserves the original generation capability. Despite being trained on only a limited set of object categories, TruthLens generalizes effectively to benchmarks with substantially larger label spaces. Extensive experiments across multiple LVLMs demonstrate state-of-the-art performance; notably, on Qwen2.5-VL-7B, TruthLens outperforms the previous best method on MS-COCO by over 17\% in AUROC. Our code is available at https://github.com/wyqstan/TruthLens.

cs.CV↗
arXiv2026

Monomial bases and canonical bases for quantum affine algebras

We construct a monomial basis of a quantum affine algebra of simply-laced type, associated to the PBW basis of Beck-Nakajima. We show that there exists a simple algorithm of computing canonical basis in terms of the monomial basis. We dsicuss the relations of the canonical basis obtained from this PBW basis with Lusztig's canonical basis constructed by using the geometry of quivers.

math.QA↗
arXiv2025

FidelityGPT: Correcting Decompilation Distortions with Retrieval Augmented Generation

Decompilation converts machine code into human-readable form, enabling analysis and debugging without source code. However, fidelity issues often degrade the readability and semantic accuracy of decompiled output. Existing methods, such as variable renaming or structural simplification, provide partial improvements but lack robust detection and correction, particularly for complex closed-source binaries. We present FidelityGPT, a framework that enhances decompiled code accuracy and readability by systematically detecting and correcting semantic distortions. FidelityGPT introduces distortion-aware prompt templates tailored to closed-source settings and integrates Retrieval-Augmented Generation (RAG) with a dynamic semantic intensity algorithm to locate distorted lines and retrieve semantically similar code from a database. A variable dependency algorithm further mitigates long-context limitations by analyzing redundant variables and integrating their dependencies into the prompt context. Evaluated on 620 function pairs from a binary similarity benchmark, FidelityGPT achieved an average detection accuracy of 89% and a precision of 83%. Compared to the state-of-the-art DeGPT (Fix Rate 83%, Corrected Fix Rate 37%), FidelityGPT attained 94% FR and 64% CFR, demonstrating significant gains in accuracy and readability. These results highlight its potential to advance LLM-based decompilation and reverse engineering.

cs.SE↗
arXiv2025

It Only Gets Worse: Revisiting DL-Based Vulnerability Detectors from a Practical Perspective

With the growing threat of software vulnerabilities, deep learning (DL)-based detectors have gained popularity for vulnerability detection. However, doubts remain regarding their consistency within declared CWE ranges, real-world effectiveness, and applicability across scenarios. These issues may lead to unreliable detection, high false positives/negatives, and poor adaptability to emerging vulnerabilities. A comprehensive analysis is needed to uncover critical factors affecting detection and guide improvements in model design and deployment. In this paper, we present VulTegra, a novel evaluation framework that conducts a multidimensional comparison of scratch-trained and pre-trained-based DL models for vulnerability detection. VulTegra reveals that state-of-the-art (SOTA) detectors still suffer from low consistency, limited real-world capabilities, and scalability challenges. Contrary to common belief, pre-trained models are not consistently better than scratch-trained models but exhibit distinct strengths in specific contexts.Importantly, our study exposes the limitations of relying solely on CWE-based classification and identifies key factors that significantly affect model performance. Experimental results show that adjusting just one such factor consistently improves recall across all seven evaluated detectors, with six also achieving better F1 scores. Our findings provide deeper insights into model behavior and emphasize the need to consider both vulnerability types and inherent code features for effective detection.

cs.SE↗
arXiv2025

Algorithm for computing canonical bases and foldings of quantum groups

Let ${\mathbf U}_q^-$ be the negative half of a quantum group of finite type. Let $P$ be the transition matrix between the canonical basis and a PBW basis of ${\mathbf U}_q^-$. In the case ${\mathbf U}_q^-$ is symmetric, Antor gave a simple algorithm of computing $P$ by making use of monomial bases. By the folding theory, ${\mathbf U}_q^-$ (symmetric, with a certain automorphism) is related to a quantum group $\underline{\mathbf U}_q^-$ of non-symmetric type. In this paper, we extend the results of Antor to the non-symmetric case, and discuss the relationship between the algorithms for ${\mathbf U}_q^-$ and for $\underline{\mathbf U}_q^-$.

math.QA↗
arXiv2025

Elementary construction of canonical bases, foldings, and piecewise linear bijections

Let ${\mathbf U}_q^-$ be the negative half of a quantum group of finite type. We construct the canonical basis of ${\mathbf U}_q^-$ by applying the folding theory of quantum groups, and piecewise linear parametrization of canonical basis. Our construction is elementary, in the sense that we don't appeal to Lusztig's geometric theory of canonical bases, nor to Kashiwara's theory of crystal bases.

math.QA↗
arXiv2024

Class Incremental Learning with Task-Specific Batch Normalization and Out-of-Distribution Detection

This study focuses on incremental learning for image classification, exploring how to reduce catastrophic forgetting of all learned knowledge when access to old data is restricted. The challenge lies in balancing plasticity (learning new knowledge) and stability (retaining old knowledge). Based on whether the task identifier (task-ID) is available during testing, incremental learning is divided into task incremental learning (TIL) and class incremental learning (CIL). The TIL paradigm often uses multiple classifier heads, selecting the corresponding head based on the task-ID. Since the CIL paradigm cannot access task-ID, methods originally developed for TIL require explicit task-ID prediction to bridge this gap and enable their adaptation to the CIL paradigm. {In this study, a novel continual learning framework extends the TIL method for CIL by introducing out-of-distribution detection for task-ID prediction. Our framework utilizes task-specific Batch Normalization (BN) and task-specific classification heads to effectively adjust feature map distributions for each task, enhancing plasticity. With far fewer parameters than convolutional kernels, task-specific BN helps minimize parameter growth, preserving stability. Based on multiple task-specific classification heads, we introduce an ``unknow'' class for each head. During training, data from other tasks are mapped to this unknown class. During inference, the task-ID is predicted by selecting the classification head with the lowest probability assigned to the unknown class. Our method achieves state-of-the-art performance on two medical image datasets and two natural image datasets. The source code is available at https://github.com/z1968357787/mbn_ood_git_main.

cs.LG↗
arXiv2023

Enhancing Large Language Models for Secure Code Generation: A Dataset-driven Study on Vulnerability Mitigation

Large language models (LLMs) have brought significant advancements to code generation, benefiting both novice and experienced developers. However, their training using unsanitized data from open-source repositories, like GitHub, introduces the risk of inadvertently propagating security vulnerabilities. To effectively mitigate this concern, this paper presents a comprehensive study focused on evaluating and enhancing code LLMs from a software security perspective. We introduce SecuCoGen\footnote{SecuCoGen has been uploaded as supplemental material and will be made publicly available after publication.}, a meticulously curated dataset targeting 21 critical vulnerability types. SecuCoGen comprises 180 samples and serves as the foundation for conducting experiments on three crucial code-related tasks: code generation, code repair and vulnerability classification, with a strong emphasis on security. Our experimental results reveal that existing models often overlook security concerns during code generation, leading to the generation of vulnerable code. To address this, we propose effective approaches to mitigate the security vulnerabilities and enhance the overall robustness of code generated by LLMs. Moreover, our study identifies weaknesses in existing models' ability to repair vulnerable code, even when provided with vulnerability information. Additionally, certain vulnerability types pose challenges for the models, hindering their performance in vulnerability classification. Based on these findings, we believe our study will have a positive impact on the software engineering community, inspiring the development of improved methods for training and utilizing LLMs, thereby leading to safer and more trustworthy model deployment.

cs.SE↗
arXiv2022

Foldings of KLR algebras

Let ${\mathbf U}^-_q$ be the negative half of the quantum group associated to a Kac-Moody algebra ${\mathfrak g}$, and $\underline{\mathbf U}^-_q$ the quantum group obtained by a folding of ${\mathfrak g}$. Let ${\mathbf A} = {\mathbf Z}[q,q^{-1}]$. McNamara showed that $\underline{\mathbf U}^-_q$ is categorified over a certain extenion ring $\widetilde{\mathbf A}$ of ${\mathbf A}$, by uing the folding theory of KLR algebras. He posed a question whether $\widetilde{\mathbf A}$ coincides with ${\mathbf A}$ or not. In this paper, we give an affirmative answer for this problem.

math.QA↗
arXiv2022

Diagram automorphisms and canonical bases for quantized enveloping algebras

Let ${\mathbf U}^-_q$ be the negative part of the quantized enveloping algebra associated to a Kac-Moody algebra ${\mathfrak g}$ of symmetric type, and $\underline{\mathbf U}^-_q$ the algebra corresponding to the orbit algebra ${\mathfrak g}^σ$ obtained from an admissible diagram automorphism $σ$ on ${\mathfrak g}$. Lusztig consructed the canonical basis ${\mathbf B}$ of ${\mathbf U}_q^-$ and the canonical signed basis $\underline{\widetilde{\mathbf B}}$ of $\underline{\mathbf U}_q^-$ by making use of the geometric theory of quivers. He proved that there is a natural bijection $\widetilde{\mathbf B}^σ \to \widetilde{\underline{\mathbf B}}$. In this paper, assuming the existence of the canonical basis ${\mathbf B}$ of ${\mathbf U}_q^-$, we construct the canonical signed basis $\widetilde{\underline{\mathbf B}}$ of $\underline{\mathbf U}_q^-$, and a natural bijection $\widetilde{\mathbf B}^σ \to \widetilde{\underline{\mathbf B}}$ by an elementary method.

math.QA↗
arXiv2022

Diagram automorphisms and canonical bases for quantum affine algebras, II

Let ${\mathbf U}_q^-$ be the negative part of the quantum enveloping algebra, and $σ$ the algebra automorphism on ${\mathbf U}_q^-$ induced from a diagram automorphism. Let $\underline{\mathbf U}_q^-$ be the quantum algebra obtained from $σ$, and $\widetilde{\mathbf B}$ (resp. $\widetilde{\underline{\mathbf B}}$) the canonical signed basis of ${\mathbf U}_q^-$ (resp. $\underline{\mathbf U}_q^-$). Assume that ${\mathbf U}_q^-$ is simply-laced of finite or affine type. In our previous papers [SZ1, 2], we have proved by an elementary method, that there exists a natural bijection $\widetilde{\mathbf B}^σ \simeq \widetilde{\underline{\mathbf B}}$ in the case where $σ$ is admissible. In this paper, we show that such a bijection exists even if $σ$ is not admissible, possibly except some small rank cases.

math.QA↗
arXiv2022

Improving Adversarial Waveform Generation based Singing Voice Conversion with Harmonic Signals

Adversarial waveform generation has been a popular approach as the backend of singing voice conversion (SVC) to generate high-quality singing audio. However, the instability of GAN also leads to other problems, such as pitch jitters and U/V errors. It affects the smoothness and continuity of harmonics, hence degrades the conversion quality seriously. This paper proposes to feed harmonic signals to the SVC model in advance to enhance audio generation. We extract the sine excitation from the pitch, and filter it with a linear time-varying (LTV) filter estimated by a neural network. Both these two harmonic signals are adopted as the inputs to generate the singing waveform. In our experiments, two mainstream models, MelGAN and ParallelWaveGAN, are investigated to validate the effectiveness of the proposed approach. We conduct a MOS test on clean and noisy test sets. The result shows that both signals significantly improve SVC in fidelity and timbre similarity. Besides, the case analysis further validates that this method enhances the smoothness and continuity of harmonics in the generated audio, and the filtered excitation better matches the target audio.

cs.SD↗
arXiv2019

Diagram automorphisms and canonical bases for quantum affine algebras

Let ${\mathbf U}^-_q$ be the negative part of the quantum enveloping algebra associated to a simply laced Kac-Moody Lie algebra ${\mathfrak g}$, and $\underline{\mathbf U}^-_q$ the algebra corresponding to the fixed point subalgebra of ${\mathfrak g}$ obtained from a diagram automorphism $σ$ on ${\mathfrak g}$. Let ${\mathbf B}^σ$ be the set of $σ$-fixed elements in the canonical basis of ${\mathbf U}_q^-$, and $\underline{\mathbf B}$ the canonical basis of $\underline{\mathbf U}_q^-$. Lusztig proved that there exists a canonical bijection ${\mathbf B}^σ \simeq \underline{\mathbf B}$ based on his geometric construction of canonical bases. In this paper, we prove (the signed bases version of) this fact, in the case where ${\mathfrak g}$ is finite or affine type, in an elementary way, in the sense that we don't appeal to the geometric theory of canonical bases nor Kashiwara's theory of crystal bases. We also discuss the correspondence between PBW-bases, by using a new type of PBW-bases of ${\mathbf U}_q^-$ obtained by Muthiah-Tingley, which is a generalization of PBW-bases constructed by Beck-Nakajima.

math.QA↗
arXiv2019

Diagram automorphisms and quantum groups

Let $U^-_q = U^-_q(\mathfrak g)$ be the negative part of the quantum group associated to a finite dimensional simple Lie algebra $\mathfrak g$, and $σ: \mathfrak g \to \mathfrak g$ be the automorphism obtained from the diagram automorphism. Let $\mathfrak g^σ$ be the fixed point subalgebra of $\mathfrak g$, and put $\underline U^-_q = U^-_q(\mathfrak g^σ)$. Let $B$ be the canonical basis of $U_q^-$ and $\underline B$ the canonical basis of $\underline U_q^-$. $σ$ induces a natural action on $B$, and we denote by $B^σ$ the set of $σ$-fixed elements in $B$. Lusztig proved that there exists a canonical bijection $B^σ \simeq \underline B$ by using geometric considerations. In this paper, we construct such a bijection in an elementary way. We also consider such a bijection in the case of certain affine quantum groups, by making use of PBW-bases constructed by Beck and Nakajima.

math.QA↗
arXiv2019

Miniature Multi-Level Optical Memristive Switch Using Phase Change Material

The optical memristive switches are electrically activated optical switches that can memorize the current state. They can be used as optical latching switches in which the switching state is changed only by applying an electrical Write/Erase pulse and maintained without external power supply. We demonstrate an optical memristive switch based on a silicon MMI structure covered with nanoscale-size Ge2Sb2Te5 (GST) material on top. The phase change of GST is triggered by resistive heating of the silicon layer beneath GST with an electrical pulse. Experimental results reveal that the optical transmissivity can be tuned in a controllable and repeatable manner with the maximum transmission contrast exceeding 20 dB. Partial crystallization of GST is obtained by controlling the width and amplitude of the electric pulses. Crucially, we also demonstrate that both Erase and Write operations, to and from any intermediate level, are possible with accurate control of the electrical pulses. Our work marks a significant step forward towards realizing photonic memristive switches without static power consumption, which are highly demanded in high-density large-scale integrated photonics.

physics.optics↗