Author name cluster

Ting Zhang

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.

15 papers

2 author rows

EAAI Journal 2025 Journal Article

Explainable remaining useful life uncertainty prediction method for rolling bearing

Ting Zhang
Honglei Wang

Rolling bearing remaining useful life prediction is the core technology of equipment maintenance. Although deep learning-based prediction methods have made significant breakthroughs, the problems of insufficient model explainability and prediction result uncertainty quantification have seriously constrained the credibility of maintenance decisions. Therefore, this research combines prediction uncertainty quantification with model explanation to propose an explainable uncertainty prediction method. The method includes multi-dimensional feature extraction, remaining useful life uncertainty prediction, and Shapley additive explanations interpreter. For feature extraction, multi-dimensional feature vectors are constructed as network inputs by extracting time-domain features and frequency-domain features. Then, the remaining useful life prediction interval for the rolling bearing is compressed by the proposed gated temporal quantile network. Finally, the prediction model is explained using the Shapley additive explanations interpreter. The multi-case validation results based on the Xi'an Jiaotong University and Changxing Sumyoung Technology Co. , Ltd. (XJTU-SY) and Intelligent Maintenance Systems (IMS) rolling bearing full life cycle datasets show that the proposed model has an average interval coverage of 93. 96 % and an average interval width of 9. 92 %, which indicates that the model maintains high accuracy and robustness in different cases. The nonlinear mapping relationship between the prediction results and the features is clarified by analyzing the Shapley values. Finally, the Shapley values are used to rank the importance of the features to locate the position that may cause rolling bearing performance degradation, which provides credible decision support for the development of the predictive maintenance strategy.

Details DOI

NeurIPS Conference 2025 Conference Paper

Glance2Gaze: Efficient Vision-Language Models from Glance Fusion to Gaze Compression

Juan Chen
Honglin Liu
Yingying Ao
Ting Zhang
Yan Huang
Xudong Liu
Biao Li
Jintao Fang

Vision-language models heavily rely on visual representations, yet ensuring its efficiency remains a critical challenge. Most existing approaches focus on reducing visual tokens either at the visual encoder phase or during the LLM decoder stage. Inspired by human visual cognition, where an initial global glance precedes focused attention on semantically salient regions, we introduce Glance2Gaze, a cognitively inspired framework that mimics the human two-stage attention process. The framework consists of two key components: the Glance Fusion module, which integrates multi-layer vision transformer features with text-aware attention to generate a semantically enriched global representation, and the Gaze Compression module, which utilizes a novel query-guided mechanism to selectively compress visual tokens based on their semantic relevance. Experimental results on widely adopted benchmarks demonstrate that Glance2Gaze outperforms existing methods, achieving superior performance with equal or lower computational cost. Furthermore, it generalizes well to high-resolution and video scenarios, showcasing robust and scalable efficiency improvements in VLMs.

PDF Details

AAAI Conference 2025 Conference Paper

RingFormer: A Ring-Enhanced Graph Transformer for Organic Solar Cell Property Prediction

Zhihao Ding
Ting Zhang
Yiran Li
Jieming Shi
Chen Jason Zhang

Organic Solar Cells (OSCs) are a promising technology for sustainable energy production. However, the identification of molecules with desired OSC properties typically involves laborious experimental research. To accelerate progress in the field, it is crucial to develop machine learning models capable of accurately predicting the properties of OSC molecules. While graph representation learning has demonstrated success in molecular property prediction, it remains underexplored for OSC-specific tasks. Existing methods fail to capture the unique structural features of OSC molecules, particularly the intricate ring systems that critically influence OSC properties, leading to suboptimal performance. To fill the gap, we present RingFormer, a novel graph transformer framework specially designed to capture both atom and ring level structural patterns in OSC molecules. RingFormer constructs a hierarchical graph that integrates atomic and ring structures and employs a combination of local message passing and global attention mechanisms to generate expressive graph representations for accurate OSC property prediction. We evaluate RingFormer's effectiveness on five curated OSC molecule datasets through extensive experiments. The results demonstrate that RingFormer consistently outperforms existing methods, achieving a 22.77% relative improvement over the nearest competitor on the CEPDB dataset.

PDF Details DOI

YNICL Journal 2024 Journal Article

Alterations in neural circuit dynamics between the limbic network and prefrontal/default mode network in patients with generalized anxiety disorder

Xiaonan Pang
Siyu Fan
Yulin Zhang
Ting Zhang
Qiangqiang Hou
Yue Wu
Ye Zhang
Yanghua Tian

BACKGROUND: Widespread functional alterations have been implicated in patients with generalized anxiety disorder (GAD). However, most studies have primarily focused on static brain network features in patients with GAD. The current research focused on exploring the dynamics within functional brain networks among individuals diagnosed with GAD. METHODS: Seventy-five participants were divided into patients with GAD and healthy controls (HCs), and resting-state functional magnetic resonance imaging data were collected. The severity of symptoms was measured using the Hamilton Anxiety Scale and the Patient Health Questionnaire. Co-activation pattern (CAP) analysis, centered on the bed nucleus of the stria terminalis, was applied to explore network dynamics. The capability of these dynamic characteristics to distinguish between patients with GAD and HCs was evaluated using a support vector machine. RESULTS: Patients with GAD exhibited disruptions in the limbic-prefrontal and limbic-default-mode network circuits. Particularly noteworthy was the marked reduction in dynamic indicators such as occurrence, EntriesFromBaseline, ExitsToBaseline, in-degree, out-degree, and resilience. Moreover, these decreased dynamic features effectively distinguished the GAD group from the HC in this study. CONCLUSIONS: The current findings revealed the underlying brain networks associated with compromised emotion regulation in individuals with GAD. The dynamic reduction in connectivity between the limbic-default mode network and limbic-prefrontal networks could potentially act as a biomarker and therapeutic target for GAD in the future.

Details DOI

NeurIPS Conference 2024 Conference Paper

DeTeCtive: Detecting AI-generated Text via Multi-Level Contrastive Learning

Xun Guo
Shan Zhang
Yongxin He
Ting Zhang
Wanquan Feng
Haibin Huang
Chongyang Ma

Current techniques for detecting AI-generated text are largely confined to manual feature crafting and supervised binary classification paradigms. These methodologies typically lead to performance bottlenecks and unsatisfactory generalizability. Consequently, these methods are often inapplicable for out-of-distribution (OOD) data and newly emerged large language models (LLMs). In this paper, we revisit the task of AI-generated text detection. We argue that the key to accomplishing this task lies in distinguishing writing styles of different authors, rather than simply classifying the text into human-written or AI-generated text. To this end, we propose DeTeCtive, a multi-task auxiliary, multi-level contrastive learning framework. DeTeCtive is designed to facilitate the learning of distinct writing styles, combined with a dense information retrieval pipeline for AI-generated text detection. Our method is compatible with a range of text encoders. Extensive experiments demonstrate that our method enhances the ability of various text encoders in detecting AI-generated text across multiple benchmarks and achieves state-of-the-art results. Notably, in OOD zero-shot evaluation, our method outperforms existing approaches by a large margin. Moreover, we find our method boasts a Training-Free Incremental Adaptation (TFIA) capability towards OOD data, further enhancing its efficacy in OOD detection scenarios. We will open-source our code and models in hopes that our work will spark new thoughts in the field of AI-generated text detection, ensuring safe application of LLMs and enhancing compliance.

PDF Details DOI

IJCAI Conference 2023 Conference Paper

Mimicking the Thinking Process for Emotion Recognition in Conversation with Prompts and Paraphrasing

Ting Zhang
Zhuang Chen
Ming Zhong
Tieyun Qian

Emotion recognition in conversation, which aims to predict the emotion for all utterances, has attracted considerable research attention in recent years. It is a challenging task since the recognition of the emotion in one utterance involves many complex factors, such as the conversational context, the speaker's background, and the subtle difference between emotion labels. In this paper, we propose a novel framework which mimics the thinking process when modeling these factors. Specifically, we first comprehend the conversational context with a history-oriented prompt to selectively gather information from predecessors of the target utterance. We then model the speaker's background with an experience-oriented prompt to retrieve the similar utterances from all conversations. We finally differentiate the subtle label semantics with a paraphrasing mechanism to elicit the intrinsic label related knowledge. We conducted extensive experiments on three benchmarks. The empirical results demonstrate the superiority of our proposed framework over the state-of-the-art baselines.

PDF Details DOI

NeurIPS Conference 2023 Conference Paper

Model-enhanced Vector Index

Hailin Zhang
Yujing Wang
Qi Chen
Ruiheng Chang
Ting Zhang
Ziming Miao
Yingyan Hou
Yang Ding

Embedding-based retrieval methods construct vector indices to search for document representations that are most similar to the query representations. They are widely used in document retrieval due to low latency and decent recall performance. Recent research indicates that deep retrieval solutions offer better model quality, but are hindered by unacceptable serving latency and the inability to support document updates. In this paper, we aim to enhance the vector index with end-to-end deep generative models, leveraging the differentiable advantages of deep retrieval models while maintaining desirable serving efficiency. We propose Model-enhanced Vector Index (MEVI), a differentiable model-enhanced index empowered by a twin-tower representation model. MEVI leverages a Residual Quantization (RQ) codebook to bridge the sequence-to-sequence deep retrieval and embedding-based models. To substantially reduce the inference time, instead of decoding the unique document ids in long sequential steps, we first generate some semantic virtual cluster ids of candidate documents in a small number of steps, and then leverage the well-adapted embedding vectors to further perform a fine-grained search for the relevant documents in the candidate virtual clusters. We empirically show that our model achieves better performance on the commonly used academic benchmarks MSMARCO Passage and Natural Questions, with comparable serving latency to dense retrieval solutions.

PDF Details

AAAI Conference 2023 Conference Paper

PeCo: Perceptual Codebook for BERT Pre-training of Vision Transformers

Xiaoyi Dong
Jianmin Bao
Ting Zhang
DongDong Chen
Weiming Zhang
Lu Yuan
Dong Chen
Fang Wen

This paper explores a better prediction target for BERT pre-training of vision transformers. We observe that current prediction targets disagree with human perception judgment. This contradiction motivates us to learn a perceptual prediction target. We argue that perceptually similar images should stay close to each other in the prediction target space. We surprisingly find one simple yet effective idea: enforcing perceptual similarity during the dVAE training. Moreover, we adopt a self-supervised transformer model for deep feature extraction and show that it works well for calculating perceptual similarity. We demonstrate that such learned visual tokens indeed exhibit better semantic meanings, and help pre-training achieve superior transfer performance in various downstream tasks. For example, we achieve 84.5% Top-1 accuracy on ImageNet-1K with ViT-B backbone, outperforming the competitive method BEiT by +1.3% under the same pre-training epochs. Our approach also gets significant improvement on object detection and segmentation on COCO and semantic segmentation on ADE20K. Equipped with a larger backbone ViT-H, we achieve the state-of-the-art ImageNet accuracy (88.3%) among methods using only ImageNet-1K data.

PDF Details DOI

EAAI Journal 2023 Journal Article

Ultimate bearing capacity prediction method and sensitivity analysis of PBL

Yixin Chen
Yanke Huang
Hao Liu
Yongsheng Liu
Ting Zhang

The ultimate bearing capacity of Perfobond leiste (PBL) is one of the key parameters to evaluate the bearing capacity and reliability of steel–concrete structures, and it is very important to predict the ultimate bearing capacity of PBL more accurately. Based on the improved cuckoo search algorithm (CS), two prediction model optimized by back propagation neural network (BPNN) algorithm and extreme learning machine (ELM) were proposed. The local search ability of CS algorithm was improved by the triangular mutation operator and the distance-based distributed discovery probability, and the global search ability was improved by multi-step selection strategy. The weight, threshold, number of input parameters and number of nodes in the hidden layer of BPNN and ELM were optimized by the triangular multi-step cuckoo search (TMCS). The comprehensive sensitivity analysis (CSA) method and Morris sensitivity analysis (MSA) method were used to analyze the sensitivity of six key parameters of PBL, such as thickness of perforated steel plate, diameter of perforated holes, number of perforated holes, diameter of through reinforcement, yield strength of through reinforcement and compressive strength of concrete. The experimental data of push-out tests in published literatures were selected as samples, and the results show that the proposed TMCS-ELM algorithm and TMCS-BPNN algorithm can accurately predict the ultimate bearing capacity of PBL, and the average errors are 4. 17% and 2. 16% respectively. The sensitivity analysis results show that the compressive strength of concrete has the highest influence on the bearing capacity of PBL, followed by the yield strength of through reinforcement.

Details DOI

AAAI Conference 2018 Conference Paper

Decoupled Convolutions for CNNs

Guotian Xie
Ting Zhang
Kuiyuan Yang
Jianhuang Lai
Jingdong Wang

In this paper, we are interested in designing small CNNs by decoupling the convolution along the spatial and channel domains. Most existing decoupling techniques focus on approximating the ﬁlter matrix through decomposition. In contrast, we provide a two-step interpretation of the standard convolution from the ﬁlter at a single location to all locations, which is exactly equivalent to the standard convolution. Motivated by the observations in our decoupling view, we propose an effective approach to relax the sparsity of the ﬁlter in spatial aggregation by learning a spatial conﬁguration, and reduce the redundancy by reducing the number of intermediate channels. Our approach achieves comparable classiﬁcation performance with the standard uncoupled convolution, but with a smaller model size over CIFAR-100, CIFAR-10 and ImageNet.

PDF Details

YNIMG Journal 2017 Journal Article

Neural correlates of believing

Xiaochun Han
Ting Zhang
Shiyu Wang
Shihui Han

Beliefs provide a fundamental cognitive basis for human behavior. But how the brain believes remains a mystery. We investigated the neural underpinnings of believing by scanning healthy adults using functional magnetic resonance imaging when they made yes/no responses to the questions whether they believe or think that a trait adjective describes themselves or a celebrity. We found that, relative to thinking, believing was characterized with better memory of self-related adjectives. Moreover, believing (vs. thinking) was associated with stronger activations in the left anterior insula/inferior frontal cortex, stronger functional connectivity between the medial prefrontal cortex and left occipital cortex during judgments of one's own personality traits, and stronger intrinsic connectivity between the left occipital cortex and the left anterior insula/inferior frontal cortex. Our findings shed new light on the neurocognitive processes that characterize believing as a mental process in healthy adults.

Details DOI

LPAR Conference 2013 Conference Paper

Partial Backtracking in CDCL Solvers

Chuan Jiang
Ting Zhang

Abstract Backtracking is a basic technique of search-based satisfiability (SAT) solvers. In order to backtrack, a SAT solver uses conflict analysis to compute a backtracking level and discards all the variable assignments made between the conflicting level and the backtracking level. We observed that, due to the branching heuristics, the solver may repeat lots of previous decisions and propagations later. In this paper, we present a new backtracking strategy, which we refer to as partial backtracking. We implemented this strategy in our solver Nigma. Using this strategy, Nigma amends the variable assignments instead of discarding them completely so that it does not backtrack as many levels as the classic strategy. Our experiments show that Nigma solves 5% more instances than the version without partial backtracking.

Details

GandALF Workshop 2012 Workshop Paper

Can Nondeterminism Help Complementation?

Yang Cai
Ting Zhang

Complementation and determinization are two fundamental notions in automata theory. The close relationship between the two has been well observed in the literature. In the case of nondeterministic finite automata on finite words (NFA), complementation and determinization have the same state complexity, namely Theta(2^n) where n is the state size. The same similarity between determinization and complementation was found for Buchi automata, where both operations were shown to have 2^Θ(n lg n) state complexity. An intriguing question is whether there exists a type of omega-automata whose determinization is considerably harder than its complementation. In this paper, we show that for all common types of omega-automata, the determinization problem has the same state complexity as the corresponding complementation problem at the granularity of 2^Θ(. ).

PDF Details DOI

CSL Conference 2011 Conference Paper

Tight Upper Bounds for Streett and Parity Complementation

Yang Cai 0001
Ting Zhang

Complementation of finite automata on infinite words is not only a fundamental problem in automata theory, but also serves as a cornerstone for solving numerous decision problems in mathematical logic, model-checking, program analysis and verification. For Streett complementation, a significant gap exists between the current lower bound 2^{Omega(n*log(n*k))} and upper bound 2^{O(n*k*log(n*k))}, where n is the state size, k is the number of Streett pairs, and k can be as large as 2^{n}. Determining the complexity of Streett complementation has been an open question since the late 80's. In this paper we show a complementation construction with upper bound 2^{O(n*log(n)+n*k*log(k))} for k=O(n) and 2^{O(n^{2}*log(n))} for k=Omega(n), which matches well the lower bound obtained in the paper arXiv: 1102. 2963. We also obtain a tight upper bound 2^{O(n*log(n))} for parity complementation.

Details

I&C Journal 2006 Journal Article

Decision procedures for term algebras with integer constraints

Ting Zhang
Henny B. Sipma
Zohar Manna

Term algebras can model recursive data structures which are widely used in programming languages. To verify programs we must be able to reason about these structures. However, as programming languages often involve multiple data domains, in program verification decision procedures for a single theory are usually not applicable. An important class of mixed constraints consists of combinations of data structures with integer constraints on the size of data structures. Such constraints can express memory safety properties such as absence of memory overflow and out-of-bound array access, which are crucial for program correctness. In this paper we extend the theory of term algebras with the length function which maps a term to its size, resulting in a combined theory of term algebras and Presburger arithmetic. This arithmetic extension provides a natural but tight coupling between the two theories, and hence the general purpose combination methods like Nelson-Oppen combination are not applicable. We present decision procedures for quantifier-free theories in structures with an infinite constant domain and with a finite constant domain. We also present a quantifier elimination procedure for the extended first-order theory that can remove a block of existential quantifiers in one step.

Details DOI