arXiv subjects

Zhen Tian

Publications and source records attributed to Zhen Tian.

At least 73 records · Page 4Linked to original sources

arXiv2024

Exploring Context Window of Large Language Models via Decomposed Positional Vectors

Transformer-based large language models (LLMs) typically have a limited context window, resulting in significant performance degradation when processing text beyond the length of the context window. Extensive studies have been proposed to extend the context window and achieve length extrapolation of LLMs, but there is still a lack of in-depth interpretation of these approaches. In this study, we explore the positional information within and beyond the context window for deciphering the underlying mechanism of LLMs. By using a mean-based decomposition method, we disentangle positional vectors from hidden states of LLMs and analyze their formation and effect on attention. Furthermore, when texts exceed the context window, we analyze the change of positional vectors in two settings, i.e., direct extrapolation and context window extension. Based on our findings, we design two training-free context window extension methods, positional vector replacement and attention window extension. Experimental results show that our methods can effectively extend the context window length.

cs.CL↗

arXiv2024

Patient-Specific CBCT Synthesis for Real-time Tumor Tracking in Surface-guided Radiotherapy

We present a new imaging system to support real-time tumor tracking for surface-guided radiotherapy (SGRT). SGRT uses optical surface imaging (OSI) to acquire real-time surface topography images of the patient on the treatment couch. However, OSI cannot visualize internal anatomy. This study proposes an Advanced Surface Imaging (A-SI) framework to address this issue. In the proposed A-SI framework, a high-speed surface imaging camera consistently captures surface images during radiation delivery, and a CBCT imager captures single-angle X-ray projections at low frequency. The A-SI then utilizes a generative model to generate real-time volumetric images with full anatomy, referred to as Optical Surface-Derived cone beam computed tomography (OSD-CBCT), based on the real-time high-frequent surface images and the low-frequency collected single-angle X-ray projections. The generated OSD-CBCT can provide accurate tumor motion for precise radiation delivery. The A-SI framework uses a patient-specific generative model: physics-integrated consistency-refinement denoising diffusion probabilistic model (PC-DDPM). This model leverages patient-specific anatomical structures and respiratory motion patterns derived from four-dimensional CT (4DCT) during treatment planning. It then employs a geometric transformation module (GTM) to extract volumetric anatomy information from the single-angle X-ray projection. A simulation study with 22 lung cancer patients evaluated the A-SI framework supported by PC-DDPM. The results showed that the framework produced real-time OSD-CBCT with high reconstruction fidelity and precise tumor localization. This study demonstrates the potential of A-SI to enable real-time tumor tracking with minimal imaging dose, advancing SGRT for motion-associated cancers and interventional procedures.

physics.med-ph↗

arXiv2024

Forgetting Through Transforming: Enabling Federated Unlearning via Class-Aware Representation Transformation

Federated Unlearning (FU) enables clients to selectively remove the influence of specific data from a trained federated learning model, addressing privacy concerns and regulatory requirements. However, existing FU methods often struggle to balance effective erasure with model utility preservation, especially for class-level unlearning in non-IID settings. We propose Federated Unlearning via Class-aware Representation Transformation (FUCRT), a novel method that achieves unlearning through class-aware representation transformation. FUCRT employs two key components: (1) a transformation class selection strategy to identify optimal forgetting directions, and (2) a transformation alignment technique using dual class-aware contrastive learning to ensure consistent transformations across clients. Extensive experiments on four datasets demonstrate FUCRT's superior performance in terms of erasure guarantee, model utility preservation, and efficiency. FUCRT achieves complete (100\%) erasure of unlearning classes while maintaining or improving performance on remaining classes, outperforming state-of-the-art baselines across both IID and Non-IID settings. Analysis of the representation space reveals FUCRT's ability to effectively merge unlearning class representations with the transformation class from remaining classes, closely mimicking the model retrained from scratch.

cs.LG↗

arXiv2024

Enhanced Visual SLAM for Collision-free Driving with Lightweight Autonomous Cars

The paper presents a vision-based obstacle avoidance strategy for lightweight self-driving cars that can be run on a CPU-only device using a single RGB-D camera. The method consists of two steps: visual perception and path planning. The visual perception part uses ORBSLAM3 enhanced with optical flow to estimate the car's poses and extract rich texture information from the scene. In the path planning phase, we employ a method combining a control Lyapunov function and control barrier function in the form of quadratic program (CLF-CBF-QP) together with an obstacle shape reconstruction process (SRP) to plan safe and stable trajectories. To validate the performance and robustness of the proposed method, simulation experiments were conducted with a car in various complex indoor environments using the Gazebo simulation environment. Our method can effectively avoid obstacles in the scenes. The proposed algorithm outperforms benchmark algorithms in achieving more stable and shorter trajectories across multiple simulated scenes.

cs.RO↗

arXiv2024

CP-Prompt: Composition-Based Cross-modal Prompting for Domain-Incremental Continual Learning

The key challenge of cross-modal domain-incremental learning (DIL) is to enable the learning model to continuously learn from novel data with different feature distributions under the same task without forgetting old ones. However, existing top-performing methods still cause high forgetting rates, by lacking intra-domain knowledge extraction and inter-domain common prompting strategy. In this paper, we propose a simple yet effective framework, CP-Prompt, by training limited parameters to instruct a pre-trained model to learn new domains and avoid forgetting existing feature distributions. CP-Prompt captures intra-domain knowledge by compositionally inserting personalized prompts on multi-head self-attention layers and then learns the inter-domain knowledge with a common prompting strategy. CP-Prompt shows superiority compared with state-of-the-art baselines among three widely evaluated DIL tasks. The source code is available at https://github.com/dannis97500/CP_Prompt.

cs.CL↗

arXiv2024

Contribution Evaluation of Heterogeneous Participants in Federated Learning via Prototypical Representations

Contribution evaluation in federated learning (FL) has become a pivotal research area due to its applicability across various domains, such as detecting low-quality datasets, enhancing model robustness, and designing incentive mechanisms. Existing contribution evaluation methods, which primarily rely on data volume, model similarity, and auxiliary test datasets, have shown success in diverse scenarios. However, their effectiveness often diminishes due to the heterogeneity of data distributions, presenting a significant challenge to their applicability. In response, this paper explores contribution evaluation in FL from an entirely new perspective of representation. In this work, we propose a new method for the contribution evaluation of heterogeneous participants in federated learning (FLCE), which introduces a novel indicator \emph{class contribution momentum} to conduct refined contribution evaluation. Our core idea is the construction and application of the class contribution momentum indicator from individual, relative, and holistic perspectives, thereby achieving an effective and efficient contribution evaluation of heterogeneous participants without relying on an auxiliary test dataset. Extensive experimental results demonstrate the superiority of our method in terms of fidelity, effectiveness, efficiency, and heterogeneity across various scenarios.

cs.LG↗

arXiv2024

Adaptive Proton Therapy Using CBCT-Guided Digital Twins

This study aims to develop a digital twin (DT) framework to enhance adaptive proton stereotactic body radiation therapy (SBRT) for prostate cancer. Prostate SBRT has emerged as a leading option for external beam radiotherapy due to its effectiveness and reduced treatment duration. However, interfractional anatomy variations can impact treatment outcomes. This study seeks to address these uncertainties using DT concept, with the goal of improving treatment quality, potentially revolutionizing prostate radiotherapy to offer personalized treatment solutions. Our study presented a pioneering approach that leverages DT technology to enhance adaptive proton SBRT. The framework improves treatment plans by utilizing patient-specific CTV setup uncertainty, which is usually smaller than conventional clinical setups. This research contributes to the ongoing efforts to enhance the efficiency and efficacy of prostate radiotherapy, with ultimate goals of improving patient outcomes and life quality.

physics.med-ph↗

arXiv2024

EulerFormer: Sequential User Behavior Modeling with Complex Vector Attention

To capture user preference, transformer models have been widely applied to model sequential user behavior data. The core of transformer architecture lies in the self-attention mechanism, which computes the pairwise attention scores in a sequence. Due to the permutation-equivariant nature, positional encoding is used to enhance the attention between token representations. In this setting, the pairwise attention scores can be derived by both semantic difference and positional difference. However, prior studies often model the two kinds of difference measurements in different ways, which potentially limits the expressive capacity of sequence modeling. To address this issue, this paper proposes a novel transformer variant with complex vector attention, named EulerFormer, which provides a unified theoretical framework to formulate both semantic difference and positional difference. The EulerFormer involves two key technical improvements. First, it employs a new transformation function for efficiently transforming the sequence tokens into polar-form complex vectors using Euler's formula, enabling the unified modeling of both semantic and positional information in a complex rotation form.Secondly, it develops a differential rotation mechanism, where the semantic rotation angles can be controlled by an adaptation function, enabling the adaptive integration of the semantic and positional information according to the semantic contexts.Furthermore, a phase contrastive learning task is proposed to improve the isotropy of contextual representations in EulerFormer. Our theoretical framework possesses a high degree of completeness and generality. It is more robust to semantic variations and possesses moresuperior theoretical properties in principle. Extensive experiments conducted on four public datasets demonstrate the effectiveness and efficiency of our approach.

cs.IR↗

arXiv2024

Sequence-level Semantic Representation Fusion for Recommender Systems

With the rapid development of recommender systems, there is increasing side information that can be employed to improve the recommendation performance. Specially, we focus on the utilization of the associated \emph{textual data} of items (eg product title) and study how text features can be effectively fused with ID features in sequential recommendation. However, there exists distinct data characteristics for the two kinds of item features, making a direct fusion method (eg adding text and ID embeddings as item representation) become less effective. To address this issue, we propose a novel {\ul \emph{Te}}xt-I{\ul \emph{D}} semantic fusion approach for sequential {\ul \emph{Rec}}ommendation, namely \textbf{\our}. The core idea of our approach is to conduct a sequence-level semantic fusion approach by better integrating global contexts. The key strategy lies in that we transform the text embeddings and ID embeddings by Fourier Transform from \emph{time domain} to \emph{frequency domain}. In the frequency domain, the global sequential characteristics of the original sequences are inherently aggregated into the transformed representations, so that we can employ simple multiplicative operations to effectively fuse the two kinds of item features. Our fusion approach can be proved to have the same effects of contextual convolution, so as to achieving sequence-level semantic fusion. In order to further improve the fusion performance, we propose to enhance the discriminability of the text embeddings from the text encoder, by adaptively injecting positional information via a mixture-of-experts~(MoE) modulation method. Our implementation is available at this repository: \textcolor{magenta}{\url{https://github.com/RUCAIBox/TedRec}}.

cs.IR↗

arXiv2023

UFIN: Universal Feature Interaction Network for Multi-Domain Click-Through Rate Prediction

Click-Through Rate (CTR) prediction, which aims to estimate the probability of a user clicking on an item, is a key task in online advertising. Numerous existing CTR models concentrate on modeling the feature interactions within a solitary domain, thereby rendering them inadequate for fulfilling the requisites of multi-domain recommendations in real industrial scenarios. Some recent approaches propose intricate architectures to enhance knowledge sharing and augment model training across multiple domains. However, these approaches encounter difficulties when being transferred to new recommendation domains, owing to their reliance on the modeling of ID features (e.g., item id). To address the above issue, we propose the Universal Feature Interaction Network (UFIN) approach for CTR prediction. UFIN exploits textual data to learn universal feature interactions that can be effectively transferred across diverse domains. For learning universal feature representations, we regard the text and feature as two different modalities and propose an encoder-decoder network founded on a Large Language Model (LLM) to enforce the transfer of data from the text modality to the feature modality. Building upon the above foundation, we further develop a mixtureof-experts (MoE) enhanced adaptive feature interaction model to learn transferable collaborative patterns across multiple domains. Furthermore, we propose a multi-domain knowledge distillation framework to enhance feature interaction learning. Based on the above methods, UFIN can effectively bridge the semantic gap to learn common knowledge across various domains, surpassing the constraints of ID-based models. Extensive experiments conducted on eight datasets show the effectiveness of UFIN, in both multidomain and cross-platform settings. Our code is available at https://github.com/RUCAIBox/UFIN.

cs.IR↗

arXiv2023

EulerNet: Adaptive Feature Interaction Learning via Euler's Formula for CTR Prediction

Learning effective high-order feature interactions is very crucial in the CTR prediction task. However, it is very time-consuming to calculate high-order feature interactions with massive features in online e-commerce platforms. Most existing methods manually design a maximal order and further filter out the useless interactions from them. Although they reduce the high computational costs caused by the exponential growth of high-order feature combinations, they still suffer from the degradation of model capability due to the suboptimal learning of the restricted feature orders. The solution to maintain the model capability and meanwhile keep it efficient is a technical challenge, which has not been adequately addressed. To address this issue, we propose an adaptive feature interaction learning model, named as EulerNet, in which the feature interactions are learned in a complex vector space by conducting space mapping according to Euler's formula. EulerNet converts the exponential powers of feature interactions into simple linear combinations of the modulus and phase of the complex features, making it possible to adaptively learn the high-order feature interactions in an efficient way. Furthermore, EulerNet incorporates the implicit and explicit feature interactions into a unified architecture, which achieves the mutual enhancement and largely boosts the model capabilities. Such a network can be fully learned from data, with no need of pre-designed form or order for feature interactions. Extensive experiments conducted on three public datasets have demonstrated the effectiveness and efficiency of our approach. Our code is available at: https://github.com/RUCAIBox/EulerNet.

cs.IR↗

arXiv2023

One-step Iterative Estimation of Effective Atomic Number and Electron Density for Dual Energy CT

Dual-energy computed tomography (DECT) is a promising technology that has shown a number of clinical advantages over conventional X-ray CT, such as improved material identification, artifact suppression, etc. For proton therapy treatment planning, besides material-selective images, maps of effective atomic number (Z) and relative electron density to that of water ($ρ_e$) can also be achieved and further employed to improve stopping power ratio accuracy and reduce range uncertainty. In this work, we propose a one-step iterative estimation method, which employs multi-domain gradient $L_0$-norm minimization, for Z and $ρ_e$ maps reconstruction. The algorithm was implemented on GPU to accelerate the predictive procedure and to support potential real-time adaptive treatment planning. The performance of the proposed method is demonstrated via both phantom and patient studies.

physics.med-ph↗

arXiv2022

Directed Acyclic Graph Factorization Machines for CTR Prediction via Knowledge Distillation

With the growth of high-dimensional sparse data in web-scale recommender systems, the computational cost to learn high-order feature interaction in CTR prediction task largely increases, which limits the use of high-order interaction models in real industrial applications. Some recent knowledge distillation based methods transfer knowledge from complex teacher models to shallow student models for accelerating the online model inference. However, they suffer from the degradation of model accuracy in knowledge distillation process. It is challenging to balance the efficiency and effectiveness of the shallow student models. To address this problem, we propose a Directed Acyclic Graph Factorization Machine (KD-DAGFM) to learn the high-order feature interactions from existing complex interaction models for CTR prediction via Knowledge Distillation. The proposed lightweight student model DAGFM can learn arbitrary explicit feature interactions from teacher networks, which achieves approximately lossless performance and is proved by a dynamic programming algorithm. Besides, an improved general model KD-DAGFM+ is shown to be effective in distilling both explicit and implicit feature interactions from any complex teacher model. Extensive experiments are conducted on four real-world datasets, including a large-scale industrial dataset from WeChat platform with billions of feature dimensions. KD-DAGFM achieves the best performance with less than 21.5% FLOPs of the state-of-the-art method on both online and offline experiments, showing the superiority of DAGFM to deal with the industrial scale data in CTR prediction task. Our implementation code is available at: https://github.com/RUCAIBox/DAGFM.

cs.IR↗

arXiv2022

Recent Advances in RecBole: Extensions with more Practical Considerations

RecBole has recently attracted increasing attention from the research community. As the increase of the number of users, we have received a number of suggestions and update requests. This motivates us to make some significant improvements on our library, so as to meet the user requirements and contribute to the research community. In order to show the recent update in RecBole, we write this technical report to introduce our latest improvements on RecBole. In general, we focus on the flexibility and efficiency of RecBole in the past few months. More specifically, we have four development targets: (1) more flexible data processing, (2) more efficient model training, (3) more reproducible configurations, and (4) more comprehensive user documentation. Readers can download the above updates at: https://github.com/RUCAIBox/RecBole.

cs.IR↗

arXiv2022

A Retrospective Study on the Investigation of Potential Clinical Benefits of Online Adaptive Proton Therapy for Head and Neck Cancer

Online adaptive proton therapy (APT) is an ideal solution theoretically, which however is challenging to proton clinics. Although multiple groups have been endeavoring to develop online APT technology, there is a concern in the radiotherapy community about the necessity of online APT because of its unknown impact on treatment outcomes. Hence, we have performed a retrospective study to investigate the potential clinical effects of online APT for HN cancer patients in relative to the current offline APT via simulations. To mimic an online APT treatment course, we have recalculated and evaluated the actual dose of the current treatment plan on patient actual treatment anatomy captured by cone beam CT for each fraction. The cumulative dose of simulated online APT courses was compared to actual offline APT courses and the initially designed treatment plan dose. For patients 1 and 2, the simulated online ART course maintained a relatively higher CTV dose coverages than the offline ART course, particularly for CTV-Low, which led to an improvement of 2.66% and 4.52% in TCP of CTV-Low. For patients 3 and 4, with clinically comparable CTV dose coverages, the simulated online ART course achieved better OAR sparing than the offline ART course. The mean doses of right parotid and oral cavity were decreased from 29.52 Gy relative biological effectiveness (RBE) and 41.89 Gy RBE to 22.16 Gy RBE and 34.61 Gy RBE for patient 3, leading to a reduce of 1.67% and 3.40% in NTCP for the two organs. Compared to the current clinical practice, the retrospective study indicated that online APT tended to spare more normal tissues by achieving the clinical goal with merely half of the positional uncertainty margin. Future studies are needed to help identify the patients with large potential benefits prior to treatment to conserve scarce clinical resources.

physics.med-ph↗

arXiv2022

Landmark Tracking in Liver US images Using Cascade Convolutional Neural Networks with Long Short-Term Memory

This study proposed a deep learning-based tracking method for ultrasound (US) image-guided radiation therapy. The proposed cascade deep learning model is composed of an attention network, a mask region-based convolutional neural network (mask R-CNN), and a long short-term memory (LSTM) network. The attention network learns a mapping from a US image to a suspected area of landmark motion in order to reduce the search region. The mask R-CNN then produces multiple region-of-interest (ROI) proposals in the reduced region and identifies the proposed landmark via three network heads: bounding box regression, proposal classification, and landmark segmentation. The LSTM network models the temporal relationship among the successive image frames for bounding box regression and proposal classification. To consolidate the final proposal, a selection method is designed according to the similarities between sequential frames. The proposed method was tested on the liver US tracking datasets used in the Medical Image Computing and Computer Assisted Interventions (MICCAI) 2015 challenges, where the landmarks were annotated by three experienced observers to obtain their mean positions. Five-fold cross-validation on the 24 given US sequences with ground truths shows that the mean tracking error for all landmarks is 0.65+/-0.56 mm, and the errors of all landmarks are within 2 mm. We further tested the proposed model on 69 landmarks from the testing dataset that has a similar image pattern to the training pattern, resulting in a mean tracking error of 0.94+/-0.83 mm. Our experimental results have demonstrated the feasibility and accuracy of our proposed method in tracking liver anatomic landmarks using US images, providing a potential solution for real-time liver tracking for active motion management during radiation therapy.

cs.CV↗

arXiv2022

RecBole 2.0: Towards a More Up-to-Date Recommendation Library

In order to support the study of recent advances in recommender systems, this paper presents an extended recommendation library consisting of eight packages for up-to-date topics and architectures. First of all, from a data perspective, we consider three important topics related to data issues (i.e., sparsity, bias and distribution shift), and develop five packages accordingly: meta-learning, data augmentation, debiasing, fairness and cross-domain recommendation. Furthermore, from a model perspective, we develop two benchmarking packages for Transformer-based and graph neural network (GNN)-based models, respectively. All the packages (consisting of 65 new models) are developed based on a popular recommendation framework RecBole, ensuring that both the implementation and interface are unified. For each package, we provide complete implementations from data loading, experimental setup, evaluation and algorithm implementation. This library provides a valuable resource to facilitate the up-to-date research in recommender systems. The project is released at the link: https://github.com/RUCAIBox/RecBole2.0.

cs.IR↗

arXiv2021

Automatic Inverse Treatment Planning for Gamma Knife Radiosurgery via Deep Reinforcement Learning

Purpose: Several inverse planning algorithms have been developed for Gamma Knife (GK) radiosurgery to determine a large number of plan parameters via solving an optimization problem, which typically consists of multiple objectives. The priorities among these objectives need to be repetitively adjusted to achieve a clinically good plan for each patient. This study aimed to achieve automatic and intelligent priority-tuning, by developing a deep reinforcement learning (DRL) based method to model the tuning behaviors of human planners. Methods: We built a priority-tuning policy network using deep convolutional neural networks. Its input was a vector composed of the plan metrics that were used in our institution for GK plan evaluation. The network can determine which tuning action to take, based on the observed quality of the intermediate plan. We trained the network using an end-to-end DRL framework to approximate the optimal action-value function. A scoring function was designed to measure the plan quality. Results: Vestibular schwannoma was chosen as the test bed in this study. The number of training, validation and testing cases were 5, 5, and 16, respectively. For these three datasets, the average plan scores with initial priorities were 3.63 $\pm$ 1.34, 3.83 $\pm$ 0.86 and 4.20 $\pm$ 0.78, respectively, while can be improved to 5.28 $\pm$ 0.23, 4.97 $\pm$ 0.44 and 5.22 $\pm$ 0.26 through manual priority tuning by human expert planners. Our network achieved competitive results with 5.42 $\pm$ 0.11, 5.10 $\pm$ 0. 42, 5.28 $\pm$ 0.20, respectively. Conclusions: Our network can generate GK plans of comparable or slightly higher quality comparing with the plans generated by human planners via manual priority tuning. The network can potentially be incorporated into the clinical workflow to improve GK planning efficiency.

physics.med-ph↗