Author name cluster

Qian Yang

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

23 papers

1 author row

AAAI Conference 2025 Conference Paper

CodeHalu: Investigating Code Hallucinations in LLMs via Execution-based Verification

Yuchen Tian
Weixiang Yan
Qian Yang
Xuandong Zhao
Qian Chen
Wen Wang
Ziyang Luo
Lei Ma

Large Language Models (LLMs) have made significant progress in code generation, offering developers groundbreaking automated programming support. However, LLMs often generate code that is syntactically correct and even semantically plausible, but may not execute as expected or fulfill specified requirements. This phenomenon of hallucinations in the code domain has not been systematically explored. To advance the community's understanding and research on this issue, we introduce the concept of code hallucinations and propose a classification method for code hallucination based on execution verification. We categorize code hallucinations into four main types: mapping, naming, resource, and logic hallucinations, with each category further divided into different subcategories to understand and address the unique challenges faced by LLMs in code generation with finer granularity. Additionally, we present a dynamic detection algorithm called CodeHalu designed to detect and quantify code hallucinations. We also introduce the CodeHaluEval benchmark, which includes 8,883 samples from 699 tasks, to systematically and quantitatively evaluate code hallucinations. By evaluating 17 popular LLMs using this benchmark, we reveal significant differences in their accuracy and reliability in code generation, offering detailed insights for further improving the code generation capabilities of LLMs.

PDF Details DOI

TIST Journal 2025 Journal Article

Cross-platform Prediction of Depression Treatment Outcome Using Location Sensory Data on Smartphones

Soumyashree Sahoo
Md. Zakir Hossain
Chinmaey Shende
Parit Patel
Yushuo Niu
Reynaldo Morillo
Xinyu Wang
Shweta Ware

ABSTRACT Currently, depression treatment relies on closely monitoring patients’ response to treatment and adjusting the treatment as needed. Using self-reported or physician-administrated questionnaires to monitor treatment response is, however, subjective, costly and suffers from recall bias. In this paper, we explore using location sensory data collected passively on smartphones to predict treatment outcome. To address heterogeneous data collection on Android and iOS phones, the two predominant smartphone platforms, we explore using domain adaptation techniques to map their data to a common feature space, and then use the data jointly to train machine learning models. We further explore integrating contrastive learning with domain adaptation to augment data and learn feature embeddings. These learned embeddings are then used to train machine learning models to predict depression treatment outcomes. Our evaluation shows that using the embeddings learned by jointly integrating contrastive learning and domain adaptation leads to the best prediction accuracy. In addition, our results show that using location features and baseline self-reported questionnaire score can lead to F1 score up to 0.76. This accuracy is comparable to that obtained using periodic self-reported questionnaires, indicating that using location data is a promising direction for predicting depression treatment outcome. Last, when all location and questionnaire data are used together, the F1 score further increases to 0.79.

Details DOI

JBHI Journal 2024 Journal Article

Synthesizing Feature-Aligned and Category-Aware Electronic Medical Records for Intracranial Aneurysm Rupture Prediction

Qian Yang
Caizi Li
Chubin Ou
Kang Li
Xiangyun Liao
Chuanzhi Duan
Lequan Yu
Weixin Si

Rupture prediction is crucial for precise treatment and follow-up management of patients with intracranial aneurysms (IAs). Considerable machine learning (ML) methods have been proposed to improve rupture prediction by leveraging electronic medical records (EMRs), however, data scarcity and category imbalance strongly influence performance. Thus, we propose a novel data synthesis method i. e. , Transformer-based conditional GAN (TransCGAN), to synthesize highly authentic and category-aware EMRs to address above challenges. Specifically, we first align feature-wise context relationship and distribution between synthetic and original data to enhance synthetic data quality. To achieve this, we first integrate the Transformer structure into GAN to match the contextual relationship by processing the long-range dependencies among clinical factors and introduce a statistical loss to maintain distributional consistency by constraining the mean and variance of the synthesis features. Additionally, a conditional module is designed to assign the category of the synthesis data, thereby addressing the challenge of category imbalance. Subsequently, the synthetic data are merged with the original data to form a large-scale and category-balanced training dataset for IAs rupture prediction. Experimental results show that using TransCGAN's synthetic data enhances classifier performance, achieving AUC of 0. 89 and outperforming state-of-the-art resampling methods by 5 $\%$ –33 $\%$ in F1 score.

Details DOI

EAAI Journal 2024 Journal Article

Time series classification models based on nonlinear spiking neural P systems

Xin Xiong
Min Wu
Juan He
Hong Peng
Jun Wang
Xianzhong Long
Qian Yang

Reservoir computing (RC) is a novel class of recurrent neural networks (RNN) models. Nonlinear spiking neural P (NSNP) systems are neural-like computing models with nonlinear spiking mechanisms. By introducing NSNP systems as the reservoir, we propose a new RC model for time series classification task, termed TSC-NSNP model. However, due to the high-dimensional nature of the reservoir state space, the TSC-NSNP model, like existing RC models, will encounter some challenges. To address the challenges. we utilize the reservoir model space representation and dimensionality reduction method to propose two improved models, termed TSC-DR-NSNP model and TSC-RMS-NSNP model. The three RC models can be easily realized and learnt in the RC framework. The proposed three RC models are evaluated on 21 benchmark time series classification data sets, and are compared with 20 classification models. The comparisons demonstrate the effectiveness of the presented three RC models for time series classification tasks.

Details DOI

AAAI Conference 2023 Conference Paper

Med-EASi: Finely Annotated Dataset and Models for Controllable Simplification of Medical Texts

Chandrayee Basu
Rosni Vasu
Michihiro Yasunaga
Qian Yang

Automatic medical text simplification can assist providers with patient-friendly communication and make medical texts more accessible, thereby improving health literacy. But curating a quality corpus for this task requires the supervision of medical experts. In this work, we present Med-EASi (Medical dataset for Elaborative and Abstractive Simplification), a uniquely crowdsourced and finely annotated dataset for supervised simplification of short medical texts. Its expert-layman-AI collaborative annotations facilitate controllability over text simplification by marking four kinds of textual transformations: elaboration, replacement, deletion, and insertion. To learn medical text simplification, we fine-tune T5-large with four different styles of input-output combinations, leading to two control-free and two controllable versions of the model. We add two types of controllability into text simplification, by using a multi-angle training approach: position-aware, which uses in-place annotated inputs and outputs, and position-agnostic, where the model only knows the contents to be edited, but not their positions. Our results show that our fine-grained annotations improve learning compared to the unannotated baseline. Furthermore, our position-aware control enhances the model's ability to generate better simplification than the position-agnostic version. The data and code are available at https://github.com/Chandrayee/CTRL-SIMP.

PDF Details DOI

TCS Journal 2023 Journal Article

Nonlinear spiking neural P systems with multiple channels

Qian Yang
Xin Xiong
Hong Peng
Jun Wang
Xiaoxiao Song

This paper investigates a new variant of spiking neural P systems (SN P systems), called nonlinear spiking neural P systems with multiple channels (in short, NSN P-MC systems). In this variant, we consider two features: (i) each neuron can use its multiple channels to connect one or more different successor neurons; (ii) nonlinear spiking rules are introduced to control the spiking of neurons. The computational power of NSN P-MC systems is discussed. Turing universality of NSN P-MC systems as number generating/accepting devices is proven. In addition, a small universal NSN P-MC system with 54 neurons is constructed to compute any Turing computable function, and a small universal NSN P-MC system with 63 neurons is constructed to generate number.

Details DOI

TCS Journal 2022 Journal Article

Computational completeness of spiking neural P systems with inhibitory rules for generating string languages

Nan Zhou
Hong Peng
Jun Wang
Qian Yang
Xiaohui Luo

Spiking neural P systems with inhibitory rules (in short, SNP-IR systems) are a distributed parallel computing model, abstracted by the spiking and inhibitory mechanisms of biological neurons. Computational completeness of SNP-IR systems as number generating/accepting and function computing devices has been studied recently. However, computational completeness of SNP-IR systems as language generating devices still has not been investigated. We discuss the relationship of languages generated by SNP-IR systems with regular languages. Moreover, we prove that SNP-IR systems can generate recursively enumerable languages by means of a projection of inverse-morphic image.

Details DOI

NeurIPS Conference 2022 Conference Paper

Dict-TTS: Learning to Pronounce with Prior Dictionary Knowledge for Text-to-Speech

Ziyue Jiang
Zhe Su
Zhou Zhao
Qian Yang
Yi Ren
Jinglin Liu
振辉叶

Polyphone disambiguation aims to capture accurate pronunciation knowledge from natural text sequences for reliable Text-to-speech (TTS) systems. However, previous approaches require substantial annotated training data and additional efforts from language experts, making it difficult to extend high-quality neural TTS systems to out-of-domain daily conversations and countless languages worldwide. This paper tackles the polyphone disambiguation problem from a concise and novel perspective: we propose Dict-TTS, a semantic-aware generative text-to-speech model with an online website dictionary (the existing prior information in the natural language). Specifically, we design a semantics-to-pronunciation attention (S2PA) module to match the semantic patterns between the input text sequence and the prior semantics in the dictionary and obtain the corresponding pronunciations; The S2PA module can be easily trained with the end-to-end TTS model without any annotated phoneme labels. Experimental results in three languages show that our model outperforms several strong baseline models in terms of pronunciation accuracy and improves the prosody modeling of TTS systems. Further extensive analyses demonstrate that each design in Dict-TTS is effective. The code is available at https: //github. com/Zain-Jiang/Dict-TTS.

PDF Details

TCS Journal 2022 Journal Article

Dynamic threshold P systems with delay on synapses for shortest path problems

Silu Yang
Dong Li
Hong Peng
Wenjie Zhou
Xiaohui Luo
Qian Yang
Jun Wang
Xiaoxiao Song

Dynamic threshold neural P (DTNP) systems are recently developed neural-like membrane computing models. By introducing the delay mechanism on synapses, we propose a variant, called dynamic threshold neural P systems with delay on synapses or termed as DTNP-DS systems. DTNP-DS systems possess three recognizable features: spiking mechanism, dynamic threshold mechanism, and delay mechanism on synapses. Based on the three mechanisms, we develop a novel algorithm based on DTNP-DS systems for solving shortest path problems. In principle, the three mechanisms provide a parallel search way for the shortest path in a given network, and they can effectively control the speed and direction of the search. Simulation results demonstrate that DTNP-DS systems can provide an effective solution for the shortest path problems.

Details DOI

YNICL Journal 2022 Journal Article

Fast and equilibrium CEST imaging of brain tumor patients at 3T

Yin Wu
Zhou Liu
Qian Yang
Liyan Zou
Fan Zhang
Long Qian
Xin Liu
Hairong Zheng

Chemical exchange saturation transfer (CEST) MRI, versatile for detecting endogenous mobile proteins and tissue pH, has proved valuable in tumor imaging. However, CEST MRI scans are often performed under non-equilibrium conditions, which confound tissue characterization. This study proposed a quasi-steady-state (QUASS) CEST MRI algorithm to standardize fast and accurate tumor imaging at 3 T. The CEST signal evolution was modeled by longitudinal relaxation rate during relaxation delay (Td) and spinlock relaxation during RF saturation time (Ts), from which the QUASS CEST effect is derived. Numerical simulation and human MR imaging experiments (7 healthy volunteers and 19 tumor patients) were conducted at 3 T to compare the CEST measurements obtained under two representative experimental conditions. In addition, amide proton transfer (APT), combined magnetization transfer (MT) and nuclear overhauser enhancement (NOE) effects, and direct water saturation were isolated using a 3-pool Lorentzian fitting in white matter and gray matter of healthy volunteers and for patients in the contralateral normal-appearing white matter and tumor regions. Finally, the student's t-test was performed between conventional and QUASS CEST measurements. The routine APT and combined MT & NOE measures significantly varied with Ts and Td (P .05), indicating the accuracy and robustness of QUASS CEST MRI for tumor imaging. To summarize, the QUASS CEST reconstruction algorithm enables fast and accurate tumor CEST imaging at 3 T, promising to expedite and standardize clinical CEST MRI.

Details DOI

I&C Journal 2021 Journal Article

Computational completeness of sequential spiking neural P systems with inhibitory rules

Tingting Bao
Nan Zhou
Hong Peng
Qian Yang
Jun Wang

Spiking neural P systems with inhibitory rules (in short, IR-SN P systems) are a kind of bio-inspired computing systems, which are abstracted by the inhibitory synaptic mechanism of biological neurons. IR-SN P systems work in synchronous mode. This paper investigates their sequential version, sequential IR-SN P systems (in short, IR-SSN P systems). In sequential mode, not only the rules in each neuron are applied sequentially, but also the neurons fire in a sequential manner. The maximum-spike-number strategy is considered in sequential mode, and two sub-modes are further distinguished: max-sequentiality strategy and max-pseudo-sequentiality strategy. Computational completeness of IR-SSN P systems as number generating/accepting devices and function computing devices are discussed. It is proven that IR-SSN P systems are Turing universal number generating/accepting devices. Moreover, a small universal IR-SSN P system for computing functions is established in max-sequential strategy.

Details DOI

TCS Journal 2021 Journal Article

Computational power of dynamic threshold neural P systems for generating string languages

Yue Huang
Wenmei Yi
Hong Peng
Jun Wang
Xiaohui Luo
Qian Yang

Inspired from spiking and dynamic mechanisms of neurons, dynamic threshold neural P systems (DTNP systems) have been developed and their computational completeness as number-generating/accepting devices and function computing devices has been investigated. However, a universality result of DTNP systems as language generators has not been established so far. This paper discusses computational power of DTNP systems as language generators. We first discuss the relationship between the languages generated by DTNP systems and finite languages, and then prove that regular languages can be generated by finite DTNP systems. Moreover, we prove that recursively enumerable languages can be characterized by projections of inverse-morphic images of the languages generated by DTNP systems.

Details DOI

TCS Journal 2021 Journal Article

Computational power of sequential dendrite P systems

Tingting Bao
Qian Yang
Hong Peng
Xiaohui Luo
Jun Wang
Xiaoxiao Song

Dendrite P (DeP) systems are a new variant of neural-like P systems, abstracted by the information processing and feedback mechanisms of dendrites. In the variant, a global block is assumed to synchronize all of neurons, hence, DeP systems work in synchronous mode. This paper investigates sequential version of the variant, that is, sequential dendrite P (SDeP) systems. Based on maximum number of spikes in neurons, two sequential modes are distinguished: max-sequentiality and max-pseudo-sequentiality strategies. SDeP systems have two interesting and recognizable features: (i) it behaves as a firing-storing process; (ii) cooperative firing mechanism. The computational completeness of SDeP systems is discussed. We prove that SDeP systems can be used as Turing universal number generating/accepting devices for max-sequentiality and max-pseudo-sequentiality strategies. We also establish a small universal function computing device of SDeP systems with 91 neurons in max-sequentiality strategy.

Details DOI

I&C Journal 2021 Journal Article

Nonlinear neural P systems for generating string languages

Nan Zhou
Qian Yang
Hong Peng
Jun Wang
Xiaohui Luo

Nonlinear spiking neural P (NSNP) system is a distributed parallel computing model inspired from the mechanisms of spiking neurons. Computational completeness of NSNP systems as number generating/accepting devices and function computing devices has been already discussed. However, universality result of NSNP systems as language generating devices has not been established so far. This paper investigates computational power of NSNP systems as language generating devices. The relationships of languages generated by NSNP systems with regular languages are investigated. Moreover, we prove that recursively enumerable languages can be characterized by the projection of inverse-morphic images of languages generated by NSNP systems.

Details DOI

AAAI Conference 2020 Conference Paper

Graph-Driven Generative Models for Heterogeneous Multi-Task Learning

Wenlin Wang
Hongteng Xu
Zhe Gan
Bai Li
Guoyin Wang
Liqun Chen
Qian Yang
Wenqi Wang

We propose a novel graph-driven generative model, that uniﬁes multiple heterogeneous learning tasks into the same framework. The proposed model is based on the fact that heterogeneous learning tasks, which correspond to different generative processes, often rely on data with a shared graph structure. Accordingly, our model combines a graph convolutional network (GCN) with multiple variational autoencoders, thus embedding the nodes of the graph (i. e. , samples for the tasks) in a uniform manner, while specializing their organization and usage to different tasks. With a focus on healthcare applications (tasks), including clinical topic modeling, procedure recommendation and admission-type prediction, we demonstrate that our method successfully leverages information across different tasks, boosting performance in all tasks and outperforming existing state-of-the-art approaches.

PDF Details

TCS Journal 2020 Journal Article

Spiking neural P systems with structural plasticity and anti-spikes

Qian Yang
Bo Li
Yue Huang
Hong Peng
Jun Wang

Spiking neural P systems (in short, SNP systems) are a class of distributed parallel computing devices, abstracted from the way neurons communicate by means of spikes. This paper discusses spiking neural P systems with structural plasticity and anti-spikes (in short, SNP-SPA systems), a new variant of SNP systems with two interesting features: structural plasticity and anti-spike. By means of plasticity rules in neurons, SNP-SPA systems can provide a dynamic directed graph structure. Turing universality of SNP-SPA systems is discussed. It is proven that SNP-SPA systems as number generating/accepting devices are Turing universal, and a small example with 56 neurons that computes a universal function is constructed. The introduction of anti-spikes allows to reduce the modules in the proof of universality.

Details DOI

NeurIPS Conference 2019 Conference Paper

Improving Textual Network Learning with Variational Homophilic Embeddings

Wenlin Wang
Chenyang Tao
Zhe Gan
Guoyin Wang
Liqun Chen
Xinyuan Zhang
Ruiyi Zhang
Qian Yang

The performance of many network learning applications crucially hinges on the success of network embedding algorithms, which aim to encode rich network information into low-dimensional vertex-based vector representations. This paper considers a novel variational formulation of network embeddings, with special focus on textual networks. Different from most existing methods that optimize a discriminative objective, we introduce Variational Homophilic Embedding (VHE), a fully generative model that learns network embeddings by modeling the semantic (textual) information with a variational autoencoder, while accounting for the structural (topology) information through a novel homophilic prior design. Homophilic vertex embeddings encourage similar embedding vectors for related (connected) vertices. The VHE encourages better generalization for downstream tasks, robustness to incomplete observations, and the ability to generalize to unseen vertices. Extensive experiments on real-world networks, for multiple tasks, demonstrate that the proposed method achieves consistently superior performance relative to competing state-of-the-art approaches.

PDF Details

TCS Journal 2019 Journal Article

Numerical P systems with Boolean condition

Liucheng Liu
Wenmei Yi
Qian Yang
Hong Peng
Jun Wang

Numerical P systems (NP systems, in short) are a class of distributed and parallel computation systems abstracted from the structure of biological cells. By introducing some control mechanisms to manipulate the application of programs, a number of NP systems have been proposed, including enzymatic NP systems and NP systems with thresholds/production thresholds. In order to control the application of programs more effectively, this paper introduces a control condition with Boolean form to propose a variant of NP systems, NP systems with Boolean condition (for short, BNP systems). Compared to the existing variants, BNP systems can provide flexible and more powerful program control ability. The computational power of the new variant is investigated. In particular, we prove that BNP systems as number generating/accepting devices and function computing devices, respectively, under three working modes (all-parallel, one-parallel and sequential), are Turing universal.

Details DOI

NeurIPS Conference 2019 Conference Paper

Ouroboros: On Accelerating Training of Transformer-Based Language Models

Qian Yang
Zhouyuan Huo
Wenlin Wang
Lawrence Carin

Language models are essential for natural language processing (NLP) tasks, such as machine translation and text summarization. Remarkable performance has been demonstrated recently across many NLP domains via a Transformer-based language model with over a billion parameters, verifying the benefits of model size. Model parallelism is required if a model is too large to fit in a single computing device. Current methods for model parallelism either suffer from backward locking in backpropagation or are not applicable to language models. We propose the first model-parallel algorithm that speeds the training of Transformer-based language models. We also prove that our proposed algorithm is guaranteed to converge to critical points for non-convex problems. Extensive experiments on Transformer and Transformer-XL language models demonstrate that the proposed algorithm obtains a much faster speedup beyond data parallelism, with comparable or better accuracy. Code to reproduce experiments is to be found at \url{https: //github. com/LaraQianYang/Ouroboros}.

PDF Details

IJCAI Conference 2017 Conference Paper

Joint Training for Pivot-based Neural Machine Translation

Yong Cheng
Qian Yang
Yang Liu
Maosong Sun
Wei Xu

While recent neural machine translation approaches have delivered state-of-the-art performance for resource-rich language pairs, they suffer from the data scarcity problem for resource-scarce language pairs. Although this problem can be alleviated by exploiting a pivot language to bridge the source and target languages, the source-to-pivot and pivot-to-target translation models are usually independently trained. In this work, we introduce a joint training algorithm for pivot-based neural machine translation. We propose three methods to connect the two models and enable them to interact with each other during training. Experiments on Europarl and WMT corpora show that joint training of source-to-pivot and pivot-to-target models leads to significant improvements over independent training across various languages.

PDF Details

AAAI Conference 2016 Conference Paper

PEAK: Pyramid Evaluation via Automated Knowledge Extraction

Qian Yang
Rebecca Passonneau
Gerard de Melo

Evaluating the selection of content in a summary is important both for human-written summaries, which can be a useful pedagogical tool for reading and writing skills, and machinegenerated summaries, which are increasingly being deployed in information management. The pyramid method assesses a summary by aggregating content units from the summaries of a wise crowd (a form of crowdsourcing). It has proven highly reliable but has largely depended on manual annotation. We propose PEAK, the ﬁrst method to automatically assess summary content using the pyramid method that also generates the pyramid content models. PEAK relies on open information extraction and graph algorithms. The resulting scores correlate well with manually derived pyramid scores on both human and machine summaries, opening up the possibility of wide-spread use in numerous applications.

PDF Details

AAAI Conference 2011 Conference Paper

Integrating Rules and Description Logics by Circumscription

Qian Yang
Jia-Huai You
Zhiyong Feng

We present a new approach to characterizing the semantics for the integration of rules and ﬁrst-order logic in general, and description logics in particular, based on a circumscription characterization of answer set programming, introduced earlier by Lin and Zhou. We show that both Rosati’s semantics based on NM-models and Lukasiewicz’s answer set semantics can be characterized by circumscription, and the difference between the two can be seen as a matter of circumscription policies. This approach leads to a number of new insights. First, we rebut a criticism on Lukasiewicz’s semantics for its inability to reason for negative consequences. Second, our approach leads to a spectrum of possible semantics based on different circumscription policies, and shows a clear picture of how they are related. Finally, we show that the idea of this paper can be applied to ﬁrst-order general stable models.

PDF Details

EAAI Journal 1998 Journal Article

Combined use of computational intelligence and materials data for on-line monitoring and control of MBE experiments

Zhuo Meng
Qian Yang
P.C. Yip
K.G. Eyink
W.T. Taferner
B. Igelnik

This paper describes the combined use of computational intelligence procedures and materials data for monitoring and controlling the growth of thin films using molecular beam epitaxy (MBE). Given ellipsometry data (Ψ and Δ) at a specific wavelength, a genetic algorithm-like method is used to solve an inverse problem, and estimate values of the complex refractive index and the deposition rate. Using a set of such values at different wavelengths, and combining the use of multiwavelength spectroscopic materials data and computational intelligence procedures, it is then possible to provide an optimal estimate of the composition of the material being deposited. Control of the film growth is then accomplished through adjustments of cell temperatures. This procedure is described in this paper, and examples of monitoring and control results are reported for the system of Al x Ga1−x As film on GaAs substrate.

Details DOI