Author name cluster

Li Li

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.

47 papers

2 author rows

TMLR Journal 2026 Journal Article

A Survey on Federated Fine-Tuning of Large Language Models

Yebo Wu
Chunlin Tian
Jingguang Li
He Sun
KaHou Tam
Zhanting Zhou
Haicheng Liao
Jing Xiong

Large Language Models (LLMs) have demonstrated impressive success across various tasks. Integrating LLMs with Federated Learning (FL), a paradigm known as FedLLM, offers a promising avenue for collaborative model adaptation while preserving data privacy. This survey provides a systematic and comprehensive review of FedLLM. We begin by tracing the historical development of both LLMs and FL, summarizing relevant prior research to set the context. Subsequently, we delve into an in-depth analysis of the fundamental challenges inherent in deploying FedLLM. Addressing these challenges often requires efficient adaptation strategies; therefore, we conduct an extensive examination of existing Parameter-Efficient Fine-tuning (PEFT) methods and explore their applicability within the FL framework. To rigorously evaluate the performance of FedLLM, we undertake a thorough review of existing fine-tuning datasets and evaluation benchmarks. Furthermore, we discuss FedLLM's diverse real-world applications across multiple domains. Finally, we identify critical open challenges and outline promising research directions to foster future advancements in FedLLM. This survey aims to serve as a foundational resource for researchers and practitioners, offering valuable insights into the rapidly evolving landscape of federated fine-tuning for LLMs. It also establishes a roadmap for future innovations in privacy-preserving AI. We actively maintain a GitHub repo to track cutting-edge advancements in this field.

AAAI Conference 2026 Conference Paper

ARDiff: Anisotropic Residual Diffusion for Heterogeneous Graph Learning

Yong Chen
Li Li
Nannan Zong
Zhihui Liu
Song-Zhi Su

Learning representations on graphs is foundational for many downstream tasks, and its synergy with diffusion models has emerged as a promising direction. However, diffusion-based methods for heterogeneous graphs remain underexplored, confronting two principal challenges: (1) The presence of noise and structural heterogeneity in graphs makes it challenging to accurately capture semantic transitions among diverse relation types. (2) The isotropic Gaussian noise used in forward diffusion fails to reflect graphs' inherent semantics and structural anisotropy. To address these, we propose ARDiff, a novel framework that integrates residual diffusion with anisotropic noise for heterogeneous graph learning. Specifically, we propose a semantic residual diffusion mechanism that progressively refines node embeddings by orchestrating transitions from low-semantic (high-noise) to high-semantic (low-noise) relational contexts, thus enabling step-wise distillation of task-relevant information. In addition, to address the limitations of conventional diffusion, we introduce an anisotropic diffusion strategy: in the forward process, noise injection is oriented by structural and semantic priors; in the denoising step, a conditional diffusion mechanism is guided by a random walk encoding, enhancing both topological consistency and semantic alignment. Extensive evaluation on heterogeneous graph datasets demonstrates that ARDiff significantly surpasses current leading methods in link prediction and node classification, setting a new paradigm and benchmark in heterogeneous graph representation learning.

PDF Details DOI

TMLR Journal 2026 Journal Article

Don't Let It Hallucinate: Premise Verification via Retrieval-Augmented Logical Reasoning

Yuehan Qin
Li Li
Yi Nian
Xinyan Velocity Yu
Yue Zhao
Xuezhe Ma

Large language models (LLMs) have shown substantial capacity for generating fluent, contextually appropriate responses. However, they can produce hallucinated outputs, especially when a user query includes one or more false premises—claims that contradict established facts. Such premises can mislead LLMs into offering fabricated or misleading details. Existing approaches include pretraining, fine-tuning, and inference-time techniques that often rely on access to logits or address hallucinations after they occur. These methods tend to be computationally expensive, require extensive training data, or lack proactive mechanisms to prevent hallucination before generation, limiting their efficiency in real-time applications. We propose a retrieval-based framework that identifies and addresses false premises before generation. Our method first transforms a user’s query into a logical representation, then applies retrieval-augmented generation (RAG) to assess the validity of each premise using factual sources. Finally, we incorporate the verification results into the LLM’s prompt to maintain factual consistency in the final output. Experiments show that this approach effectively reduces hallucinations, improves factual accuracy, and does not require access to model logits or large-scale fine-tuning.

AAAI Conference 2026 Conference Paper

Mitigating Hallucinations in Large Language Models via Causal Reasoning

Yuangang Li
Yiqing Shen
Yi Nian
Jiechao Gao
Ziyi Wang
Chenxiao Yu
Li Li
Jie Wang

Large language models (LLMs) exhibit logically inconsistent hallucinations that appear coherent yet violate reasoning principles, with recent research suggesting an inverse relationship between causal reasoning capabilities and such hallucinations. However, existing reasoning approaches in LLMs, such as Chain-of-Thought (CoT) and its graph-based variants, operate at the linguistic token level rather than modeling the underlying causal relationships between variables, lacking the ability to represent conditional independencies or satisfy causal identification assumptions. To bridge this gap, we introduce causal-DAG construction and reasoning (CDCR-SFT), a supervised fine-tuning framework that trains LLMs to explicitly construct variable-level directed acyclic graph (DAG) and then perform reasoning over it. Moreover, we present a dataset comprising 25,368 samples (CausalDR), where each sample includes an input question, explicit causal DAG, graph-based reasoning trace, and validated answer. Experiments on four LLMs across eight tasks show that CDCR-SFT improves the causal reasoning capability with the state-of-the-art 95.33% accuracy on CLADDER (surpassing human performance of 94.8% for the first time) and reduces the hallucination on HaluEval with 10% improvements. It demonstrates that explicit causal structure modeling in LLMs can effectively mitigate logical inconsistencies in LLM outputs.

PDF Details DOI

AAAI Conference 2026 Conference Paper

Neural Video Compression with Reference Hierarchy

Chuanbo Tang
Zhuoyuan Li
Li Li
Dong Liu
Feng Wu

Efficient reference structures are essential in video compression, enabling the exploitation of temporal dependencies across frames to reduce redundancy. In this paper, we delve into the inter-frame reference management mechanism in neural video codecs (NVCs). Previous schemes have inherited the reference propagation mechanism with the guidance of predefined reference structure, but the reference modeling across diverse reference sources remains underexplored. Moreover, the mismatch between the reference structure used for motion estimation and motion compensation limits the effectiveness of inter-frame prediction. To address the above limitations, we propose the unified reference hierarchy that integrates a learned hierarchical reference structure into the existing inherent reference propagation mechanism. Specifically, we first propose the hierarchical reference structure (HRS) to manage the multiple temporal contexts in the propagated reference feature, where a hierarchy-aware reference modulation module is integrated to select the most relevant reference features across different quality levels under the guidance of the reference balance loss. In addition, we propose the HRS-guided feature-wise inter-frame prediction that learns the low-rank approximation of the selected reference feature for ensuring the consistency and improving the inter-frame prediction performance. We conduct experiments on a state-of-the-art NVC, DCVC-DC. Experimental results show that our codec achieves an average 26% bitrate saving over H.266/VVC, and a 28.2% bitrate reduction compared to DCVC-DC without increasing the decoding complexity.

PDF Details DOI

AAAI Conference 2026 Conference Paper

Transform-Free Feature Coding via Entropy-Constrained Vector Quantization

Qiaoxi Chen
Changsheng Gao
Li Li
Dong Liu

Feature coding has recently emerged as a key technique for efficient transmission of intermediate representations in distributed AI systems. Existing approaches largely follow a transform-based pipeline inherited from image and video coding, where the transform module is used to remove spatial structural redundancies in visual signals. However, our analysis indicates that such redundancies have already been largely removed during feature extraction, which reduces the necessity of the transform module. Building on this insight, we propose a new transform-free pipeline that directly encodes the extracted features via a vector quantization module and an entropy model. The proposed transform‑free framework jointly learns the quantization codebook and entropy model, enabling end‑to‑end optimization tailored to the inherent feature characteristics. Furthermore, the proposed method inherently avoids the computational complexity of the transform module. Experiments on features from diverse architectures and tasks demonstrate that our method achieves superior rate-distortion performance compared to transform-based baselines, while significantly reducing the encoding and decoding complexity.

PDF Details DOI

NeurIPS Conference 2025 Conference Paper

DiffLiG: Diffusion-enhanced Liquid Graph with Attention Propagation for Grid-to-Station Precipitation Correction

Yuxiang Li
Yang Zhang
Li Li
Mengxuan Chen
Meng Jin
Fang Wang
Haohuan Fu
Juepeng Zheng

Modern precipitation forecasting systems, including reanalysis datasets, numerical models, and AI-based approaches, typically produce coarse-resolution gridded outputs. The process of converting these outputs to station-level predictions often introduces substantial spatial biases relative to station-level observations, especially in complex terrains or under extreme conditions. These biases stem from two core challenges: (i) $\textbf{station-level heterogeneity}$, with site-specific temporal and spatial dynamics; and (ii) $\textbf{oversmoothing}$, which blurs fine-scale variability in graph-based models. To address these issues, we propose $\textbf{DiffLiG}$ ($\underline{Diff}$usion-enhanced $\underline{Li}$quid $\underline{G}$raph with Attention Propagation), a graph neural network designed for precise spatial correction from gridded forecasts to station observations. DiffLiG integrates a GeoLiquidNet that adapts temporal encoding via site-aware OU dynamics, a graph neural network with a dynamic edge modulator that learns spatially adaptive connectivity, and a Probabilistic Diffusion Selector that generates and refines ensemble forecasts to mitigate oversmoothing. Experiments across multiple datasets show that DiffLiG consistently outperforms other methods, delivering more accurate and robust corrections across diverse geographic and climatic settings. Moreover, it achieves notable gains on other key meteorological variables, underscoring its generalizability and practical utility.

YNICL Journal 2025 Journal Article

Effects of parietal iTBS on resting-state effective connectivity within the frontoparietal network in patients with schizophrenia: An fMRI study

Li Li
Lina Wang
Han Wu
Bing Li
Weigang Pan
Wenqing Jin
Wen Wang
Yanping Ren

BACKGROUND: Although intermittent theta burst stimulation (iTBS) has shown effectiveness in addressing working memory (WM) deficits in individuals with schizophrenia (SZ), the current body of evidence is limited and the specific mechanisms involved remain unclear. Therefore, this pilot fMRI study aimed to examine the efficacy of parietal iTBS in ameliorating WM impairments and explore its influence on the resting-state effective connectivity within the frontoparietal network in patients with SZ. METHOD: A total of 48 patients diagnosed with SZ were randomly assigned to an active or sham iTBS group and underwent 20 sessions of active or sham iTBS over 4 weeks. Subsequently, all patients underwent cognitive tests, clinical symptom assessments, and resting-state functional MRI (rs-fMRI) scans. The effective connectivity between the frontal and parietal brain regions during the rs-fMRI scans was analyzed using a spectral dynamic causal modeling approach. Additionally, this trial was registered at the Chinese Clinical Trial Registry in November 2022 (registry number: ChiCTR2200057286). RESULTS: iTBS treatment improved the positive symptoms, negative symptoms, general psychopathology, and WM deficits. Following the iTBS intervention, the active group demonstrated a significant increase in connectivity strengths from the right MFG to the right SPL (p = 0.031) and from the left SPL to the left MFG (p = 0.010) compared to the pre-treatment levels. Additionally, compared to the sham group, the active group displayed a significantly higher connectivity strength from the right MFG to the right SPL (p = 0.042) after iTBS treatment. CONCLUSION: All these findings suggest that iTBS targeting the parietal region may influence the resting-state effective connectivity within the frontoparietal network, thereby offering promising therapeutic implications for alleviating the cognitive deficits in SZ.

AAAI Conference 2025 Conference Paper

Few-Shot Domain Adaptation for Learned Image Compression

Tianyu Zhang
Haotian Zhang
Yuqi Li
Li Li
Dong Liu

Learned image compression (LIC) has achieved state-of-the-art rate-distortion performance, deemed promising for next-generation image compression techniques. However, pre-trained LIC models usually suffer from significant performance degradation when applied to out-of-training-domain images, implying their poor generalization capabilities. To tackle this problem, we propose a few-shot domain adaptation method for LIC by integrating plug-and-play adapters into pre-trained models. Drawing inspiration from the analogy between latent channels and frequency components, we examine domain gaps in LIC and observe that out-of-training-domain images disrupt pre-trained channel-wise decomposition. Consequently, we introduce a method for channel-wise re-allocation using convolution-based adapters and low-rank adapters, which are lightweight and compatible to mainstream LIC schemes. Extensive experiments across multiple domains and multiple representative LIC schemes demonstrate that our method significantly enhances pre-trained models, achieving comparable performance to H.266/VVC intra coding with merely 25 target-domain samples. Additionally, our method matches the performance of full-model finetune while transmitting fewer than 2% of the parameters.

PDF Details DOI

IS Journal 2025 Journal Article

Intelligent and Autonomous Systems in Government

Prithviraj Dasgupta
Damian Marriott
Li Li
Priya Narayanan

Artificial intelligence (AI)-driven autonomous and intelligent systems are increasingly shaping human life, with government-led AI projects playing a crucial role in both enhancing societal well-being and influencing AI policy. Implementing AI at national or regional scales presents some unique challenges including ensuring widespread access across diverse populations, guaranteeing fairness and accountability, and effectively communicating the impact of these technologies to the public. Our special issue presents six articles highlighting real-world experiences from ongoing and recently concluded government projects. The articles describe research that leverage autonomy and intelligence for various initiatives including safeguarding citizens and infrastructure from drone-based aerial threats, inspecting civilian infrastructure, cyber-security, conversational AI and responsible use of AI. We envisage that these articles will guide researchers with insights and best practices for ethically and effectively deploying AI in diverse government projects worldwide.

EAAI Journal 2025 Journal Article

Resilient kernel-based unsupervised multi-view feature selection via compact binary hashing

Rongyao Hu
Mengmeng Zhan
Jiangzhang Gan
Li Li
Fei Ye
Tong Liu

Multi-view feature selection across diverse views identifying a compact subset of the most informative feature across various data views without relying on labeled information. While most of the solutions are limited to linear multi-view data or utilize weakly-supervised single-label learning to assist in feature selection, leading to the loss of valuable semantic information, especially when dealing with complex real-world multi-view datasets. To overcome these limitations, we introduce a novel Resilient Kernel-based Unsupervised Multi-view Feature Selection via compact Binary Hashing (RKUMBH), which aims to search a robust and consistent graph representation across views, leveraging binary hashing codes to guide feature selection. Specifically, we first standardize the dimensionality of multi-view data by using non-linear kernel mapping. Then, we explore consistent graph structures across different views by fusing individual similarity graph of each view under a self-representation guidance. Moreover, the low-rank constraints are used to preserve the primary structures and patterns embedding within the data, and an unsupervised hashing feature selection framework is conducted to generate reliable hashing codes across views. Additionally, we design a customized iterative optimization method to solve the unified model. Extensive experiments on six public multi-view datasets demonstrate that our proposed method obtains state-of-the-art results compared to existing works for both clustering and feature selection tasks.

NeurIPS Conference 2025 Conference Paper

SeasonBench-EA: A Multi-Source Benchmark for Seasonal Prediction and Numerical Model Post-Processing in East Asia

Mengxuan Chen
Li Li
Zou Ziheng
Fang Wang
Jinxiao Zhang
Runmin Dong
Juepeng Zheng
Haohuan Fu

Seasonal-scale climate prediction plays a critical role in supporting agricultural planning, disaster prevention, and long-term decision making. In particular, reliable forecasts issued 1-6 months in advance are essential for early warning of flood and drought risks associated with precipitation during the East Asian summer monsoon season. However, while the use of machine learning techniques has advanced rapidly in weather and subseasonal-to-seasonal forecasting, partly driven by the availability of benchmark datasets, their application to seasonal-scale prediction remains limited. Existing seasonal prediction primarily relies on ensemble forecasts from numerical models, which, while physically grounded, are subject to biases and uncertainties at long lead times. Motivated by these challenges, we propose SeasonBench-EA, a benchmark dataset for seasonal prediction in East Asia region. It features multi-resolution, multi-source data with both regional and global coverage, integrating ERA5 reanalysis data and ensemble forecasts from multiple leading forecast centers. Beyond key atmospheric fields, the dataset also includes boundary-related variables, such as ocean state, soil and solar radiation, that are essential for capturing seasonal-scale atmospheric variability. Two tasks are defined and evaluated: 1) machine learning-based seasonal prediction using ERA5 reanalysis, and 2) post-processing of seasonal forecasts from numerical model ensembles. A suite of deterministic and probabilistic metrics is provided for tasks evaluation, along with a hindcast assessment focused on precipitation during the East Asian summer monsoon, aligned with model evaluation protocols used in operations. By offering a unified data and evaluation framework, SeasonBench-EA aims to promote the development and application of data-driven methods for seasonal prediction, a challenging yet highly impactful task with board implications for society and public well-being. Our benchmark is available at https: //github. com/SauryChen/SeasonBench-EA.

NeurIPS Conference 2025 Conference Paper

SPFL: Sequential updates with Parallel aggregation for Enhanced Federated Learning under Category and Domain Shifts

Haoyuan Liang
Shilei Cao
Li Li
Zhiyu Ye
Haohuan Fu
Juepeng Zheng

Federated learning (FL) has recently emerged as the primary approach to overcoming data silos, enabling collaborative model training without sharing sensitive or proprietary data. Parallel federated learning (PFL) aggregates models trained independently on each client’s local data, which can lead to suboptimal convergence due to limited data exposure. In contrast, Sequential Federated Learning (SFL) allows models to traverse client datasets sequentially, enhancing data utilization. However, SFL effectiveness is limited in real-world non-IID scenarios characterized by category shift (inconsistent class distributions) and domain shift (distribution discrepancies). These shifts cause two critical issues: update order sensitivity, where model performance varies significantly with the sequence of client updates, and catastrophic forgetting, where the model forgets previously learned features when trained on new client data. We propose SPFL, a novel updating method that can be integrated into existing FL methods, integrating sequential updates with parallel aggregation to enhance data utilization and ease update order sensitivity. At the same time, we give the convergence analysis of SPFL under strong convex, general convex, and non-convex conditions, proving that this update scheme is significantly better than PFL and SFL. Additionally, we introduce the Global-Local Alignment Module to mitigate catastrophic forgetting by aligning the predictions of the global model with those of the local and previous models during training. Our extensive experiments demonstrate that integrating SPFL into existing PFL methods significantly improves performance under category and domain shifts.

AAAI Conference 2025 Conference Paper

Unsupervised Kernel-based Multi-view Feature Selection with Robust Self-representation and Binary Hashing

Rongyao Hu
Jiangzhang Gan
Mengmeng Zhan
Li Li
Mengling Wei

Unsupervised multi-view feature selection involves selecting a subset of crucial features across diverse views to diminish feature dimensionality without leveraging label information. While numerous studies have delved into this area, current solutions predominantly rely on linear multi-view data or employ weakly supervised learning to aid in feature selection. These approaches may risk losing semantic information when applied to real-world multi-view datasets. In this study, we introduce a novel model, Unsupervised Kernel-based Multi-view Feature selection with Robust self-representation and Binary hashing (UKMFS), which aims to identify robust consistent graph representation across views and leverage binary hashing codes to guide feature selection. Specifically, we first explore the underlying geometry by unifying the dimension of multi-view data with non-linear kernel mapping. Then, we search the consistent graph across views by fusing unique graph representations of each view in a self-representation manner. Additionally, we impose low-rank constraints on the graph of each view to mitigate noise and unimportant parts for preserving the main structures and patterns. Furthermore, we design an unsupervised hashing feature selection model to exploit reliable binary labels across views and weighted matrices from each view. Finally, an effective optimization method is customised to solve the formulated problem iteratively. Comprehensive experiments on public multi-view datasets indicate that our proposed method achieves state-of-the-art performance compared with the representative comparison methods regarding the clustering and the feature selection task.

PDF Details DOI

NeurIPS Conference 2025 Conference Paper

VETA-DiT: Variance-Equalized and Temporally Adaptive Quantization for Efficient 4-bit Diffusion Transformers

Qinkai XU
Yijin Liu
Lin Yang
Li Li
Yuxiang Fu

Diffusion Transformers (DiTs) have recently demonstrated remarkable performance in visual generation tasks, surpassing traditional U-Net-based diffusion models by significantly improving image and video generation quality and scalability. However, the large model size and iterative denoising process introduce substantial computational and memory overhead, limiting their deployment in real-world applications. Post-training quantization (PTQ) is a promising solution that compresses models and accelerates inference by converting weights and activations to low-bit representations. Despite its potential, PTQ faces significant challenges when applied to DiTs, often resulting in severe degradation of generative quality. To address these issues, we propose VETA-DiT (**V**ariance-**E**qualized and **T**emporal **A**daptation for **Di**ffusion **T**ransformers), a dedicated quantization framework for DiTs. Our method first analyzes the sources of quantization error from the perspective of inter-channel variance and introduces a Karhunen–Loève Transform enhanced alignment to equalize variance across channels, facilitating effective quantization under low bit-widths. Furthermore, to handle the temporal variation of activation distributions inherent in the iterative denoising steps of DiTs, we design an incoherence-aware adaptive method that identifies and properly calibrates timesteps with high quantization difficulty. We validate VETA-DiT on extensive image and video generation tasks, preserving acceptable visual quality under the more aggressive W4A4 configuration. Specifically, VETA-DiT reduces FID by 33. 65 on the DiT-XL/2 model and by 45. 76 on the PixArt-$\Sigma$ model compared to the baseline under W4A4, demonstrating its strong quantization capability and generative performance. Code is available at: https: //github. com/xululi0223/VETA-DiT.

IROS Conference 2024 Conference Paper

Bayesian Deep Predictive Coding for Snake-like Robotic Control in Unknown Terrains

William Ziming Qu
Jessica Ziyu Qu
Li Li
Jie Yang
Yuanyuan Jia

Effectively modeling the spatio-temporal interactions both internally and externally is a challenge in controlling multi-linked snake robots. This paper presents an effective method based on deep predictive coding: SnakeFormer, to address the aforementioned issue. The main contributions include: 1) Deriving a variational free energy function with two innovative regularization terms through Bayesian probabilistic analysis, offering a novel perspective to simulate the interactions between agent and the environment; 2) Introducing an interaction-attention model within a Transformer structure for predicting dynamics, and collaboratively addressing path planning and obstacle avoidance tasks. 3) By incorporating serpenoid embedding and optimizing self-attention computations, the gait stability and motion efficiency are improved. Preliminary experiments and comparative analysis with baseline models fully validate the effectiveness and generalizability of the method.

IJCAI Conference 2024 Conference Paper

FedGCS: A Generative Framework for Efficient Client Selection in Federated Learning via Gradient-based Optimization

Zhiyuan Ning
Chunlin Tian
Meng Xiao
Wei Fan
Pengyang Wang
Li Li
Pengfei Wang
Yuanchun Zhou

Federated Learning faces significant challenges in statistical and system heterogeneity, along with high energy consumption, necessitating efficient client selection strategies. Traditional approaches, including heuristic and learning-based methods, fall short of addressing these complexities holistically. In response, we propose FedGCS, a novel generative client selection framework that innovatively recasts the client selection process as a generative task. Drawing inspiration from the methodologies used in large language models, FedGCS efficiently encodes abundant decision-making knowledge within a continuous representation space, enabling efficient gradient-based optimization to search for optimal client selection that will be finally output via generation. The framework comprises four steps: (1) automatic collection of diverse “selection-score” pair data using classical client selection methods; (2) training an encoder-evaluator-decoder framework on this data to construct a continuous representation space; (3) employing gradient-based optimization in this space for optimal client selection; (4) generating the final optimal client selection via using beam search for the well-trained decoder. FedGCS outperforms traditional methods by being more comprehensive, generalizable, and efficient, simultaneously optimizing for model performance, latency, and energy consumption. The effectiveness of FedGCS is proven through extensive experimental analyses.

PDF Details DOI

NeurIPS Conference 2024 Conference Paper

HydraLoRA: An Asymmetric LoRA Architecture for Efficient Fine-Tuning

Chunlin Tian
Zhan Shi
Zhijiang Guo
Li Li
Chengzhong Xu

Adapting Large Language Models (LLMs) to new tasks through fine-tuning has been made more efficient by the introduction of Parameter-Efficient Fine-Tuning (PEFT) techniques, such as LoRA. However, these methods often underperform compared to full fine-tuning, particularly in scenarios involving complex datasets. This issue becomes even more pronounced in complex domains, highlighting the need for improved PEFT approaches that can achieve better performance. Through a series of experiments, we have uncovered two critical insights that shed light on the training and parameter inefficiency of LoRA. Building on these insights, we have developed HydraLoRA, a LoRA framework with an asymmetric structure that eliminates the need for domain expertise. Our experiments demonstrate that HydraLoRA outperforms other PEFT approaches, even those that rely on domain knowledge during the training and inference phases. Our anonymous codes are submitted with the paper and will be publicly available. Code is available: https: //github. com/Clin0212/HydraLoRA.

PDF Details DOI

AAAI Conference 2024 Conference Paper

Offline and Online Optical Flow Enhancement for Deep Video Compression

Chuanbo Tang
Xihua Sheng
Zhuoyuan Li
Haotian Zhang
Li Li
Dong Liu

Video compression relies heavily on exploiting the temporal redundancy between video frames, which is usually achieved by estimating and using the motion information. The motion information is represented as optical flows in most of the existing deep video compression networks. Indeed, these networks often adopt pre-trained optical flow estimation networks for motion estimation. The optical flows, however, may be less suitable for video compression due to the following two factors. First, the optical flow estimation networks were trained to perform inter-frame prediction as accurately as possible, but the optical flows themselves may cost too many bits to encode. Second, the optical flow estimation networks were trained on synthetic data, and may not generalize well enough to real-world videos. We address the twofold limitations by enhancing the optical flows in two stages: offline and online. In the offline stage, we fine-tune a trained optical flow estimation network with the motion information provided by a traditional (non-deep) video compression scheme, e.g. H.266/VVC, as we believe the motion information of H.266/VVC achieves a better rate-distortion trade-off. In the online stage, we further optimize the latent features of the optical flows with a gradient descent-based algorithm for the video to be compressed, so as to enhance the adaptivity of the optical flows. We conduct experiments on two state-of-the-art deep video compression schemes, DCVC and DCVC-DC. Experimental results demonstrate that the proposed offline and online enhancement together achieves on average 13.4% bitrate saving for DCVC and 4.1% bitrate saving for DCVC-DC on the tested videos, without increasing the model or computational complexity of the decoder side.

PDF Details DOI

EAAI Journal 2024 Journal Article

Online hashing with partially known labels for cross-modal retrieval

Zhenqiu Shu
Li Li
Zhengtao Yu

Online hashing technology has attracted widespread attention in the era of big data due to its ability to quickly and effectively process massive streaming media data. However, annotating samples in practical applications is expensive and labor-intensive. Therefore, we usually fail to use sufficient labeled samples to train the hashing model, thereby limiting the application of supervised learning methods. To address this problem, in this paper, we propose a novel hashing method, called online hashing with partially known labels (OHPKL), for cross-modal retrieval. Specifically, the proposed OHPKL method utilizes the semantic similarity of partially known labels to train classifiers of different modalities. Meanwhile, the l 2, 1 norm constraints are applied to the learned classifiers to mitigate the impact of redundancy and noises. Then the learned classifiers are used to predict the labels of unlabeled samples. Moreover, our OHPKL method not only leverages semantic similarity among newly arrived data, but also establishes semantic connections between new and existing data by employing chunk similarity. Extensive experiments on several benchmark datasets demonstrate the superiority of our OHPKL approach in cross-modal retrieval scenarios with partially known labels. The source code for this work is available at https: //github. com/szq0816/OHPKL.

AAAI Conference 2024 Conference Paper

Panoptic Scene Graph Generation with Semantics-Prototype Learning

Li Li
Wei Ji
Yiming Wu
Mengze Li
You Qin
Lina Wei
Roger Zimmermann

Panoptic Scene Graph Generation (PSG) parses objects and predicts their relationships (predicate) to connect human language and visual scenes. However, different language preferences of annotators and semantic overlaps between predicates lead to biased predicate annotations in the dataset, i.e. different predicates for the same object pairs. Biased predicate annotations make PSG models struggle in constructing a clear decision plane among predicates, which greatly hinders the real application of PSG models. To address the intrinsic bias above, we propose a novel framework named ADTrans to adaptively transfer biased predicate annotations to informative and unified ones. To promise consistency and accuracy during the transfer process, we propose to observe the invariance degree of representations in each predicate class, and learn unbiased prototypes of predicates with different intensities. Meanwhile, we continuously measure the distribution changes between each presentation and its prototype, and constantly screen potentially biased data. Finally, with the unbiased predicate-prototype representation embedding space, biased annotations are easily identified. Experiments show that ADTrans significantly improves the performance of benchmark models, achieving a new state-of-the-art performance, and shows great generalization and effectiveness on multiple datasets. Our code is released at https://github.com/lili0415/PSG-biased-annotation.

PDF Details DOI

JBHI Journal 2024 Journal Article

Spatio-Temporal Classification of Lung Ventilation Patterns Using 3D EIT Images: A General Approach for Individualized Lung Function Evaluation

Shuzhe Chen
Li Li
Zhichao Lin
Ke Zhang
Ying Gong
Lu Wang
Xu Wu
Maokun Li

The Pulmonary Function Test (PFT) is a widely utilized and rigorous classification test for evaluating lung function, serving as a comprehensive diagnostic tool for lung conditions. Meanwhile, Electrical Impedance Tomography (EIT) is a rapidly advancing clinical technique that visualizes conductivity distribution induced by ventilation. EIT provides additional spatial and temporal information on lung ventilation beyond traditional PFT. However, relying solely on conventional isolated interpretations of PFT results and EIT images overlooks the continuous dynamic aspects of lung ventilation. This study aims to classify lung ventilation patterns by extracting spatial and temporal features from the 3D EIT image series. The study uses a Variational Autoencoder (VAE) with a MultiRes block to compress the spatial distribution in a 3D image into a one-dimensional vector. These vectors are then stacked to create a feature map for the exhibition of temporal features. A simple convolutional neural network is used for classification. Data from 137 subjects were utilized for the training phase. Initially, the model underwent validation through a leave-one-out cross-validation process. During this validation, the model achieved an accuracy and sensitivity of 0. 96 and 1. 00, respectively, with an f1-score of 0. 98 when identifying the normal subjects. To assess pipeline reliability and feasibility, we tested it on 9 newly recruited subjects, with accurate ventilation mode predictions for 8 out of 9. In addition, we included 2D EIT results for comparison and conducted ablation experiments to validate the effectiveness of the VAE. The study demonstrates the potential of using image series for lung ventilation mode classification, providing a feasible method for patient prescreening and presenting an alternative form of PFT.

NeurIPS Conference 2023 Conference Paper

CLIP4HOI: Towards Adapting CLIP for Practical Zero-Shot HOI Detection

Yunyao Mao
Jiajun Deng
Wengang Zhou
Li Li
Yao Fang
Houqiang Li

Zero-shot Human-Object Interaction (HOI) detection aims to identify both seen and unseen HOI categories. A strong zero-shot HOI detector is supposed to be not only capable of discriminating novel interactions but also robust to positional distribution discrepancy between seen and unseen categories when locating human-object pairs. However, top-performing zero-shot HOI detectors rely on seen and predefined unseen categories to distill knowledge from CLIP and jointly locate human-object pairs without considering the potential positional distribution discrepancy, leading to impaired transferability. In this paper, we introduce CLIP4HOI, a novel framework for zero-shot HOI detection. CLIP4HOI is developed on the vision-language model CLIP and ameliorates the above issues in the following two aspects. First, to avoid the model from overfitting to the joint positional distribution of seen human-object pairs, we seek to tackle the problem of zero-shot HOI detection in a disentangled two-stage paradigm. To be specific, humans and objects are independently identified and all feasible human-object pairs are processed by Human-Object interactor for pairwise proposal generation. Second, to facilitate better transferability, the CLIP model is elaborately adapted into a fine-grained HOI classifier for proposal discrimination, avoiding data-sensitive knowledge distillation. Finally, experiments on prevalent benchmarks show that our CLIP4HOI outperforms previous approaches on both rare and unseen categories, and sets a series of state-of-the-art records under a variety of zero-shot settings.

TMLR Journal 2023 Journal Article

DSpar: An Embarrassingly Simple Strategy for Efficient GNN training and inference via Degree-based Sparsification

Zirui Liu
Kaixiong Zhou
Zhimeng Jiang
Li Li
Rui Chen
Soo-Hyun Choi
Xia Hu

Running Graph Neural Networks (GNNs) on large graphs suffers from notoriously inefficiency. This is attributed to the sparse graph-based operations, which is hard to be accelerated by community hardware, e.g., GPUs and CPUs. One potential solution is to ``sketch'' the original graph by removing unimportant edges, then both the training and inference process are executed on the sparsified graph with improved efficiency. Traditional graph sparsification work calculates the edge importance score, i.e., effective resistance, from graph topology with theoretical guarantee. However, estimating effective resistance is even more expensive than training GNNs itself. Later, learning-based sparsification methods propose to learn the edge importance from data, but with significant overhead due to the extra learning process. Thus, both of them introduce significant ahead-of-training overhead. In this paper, we experimentally and theoretically prove that effective resistance can be approximated using only the node degree information and achieve similar node presentations on graph with/without sparsification. Based on this finding, we propose DSpar, to sparsify the graph once before training based on only the node degree information with negligible ahead-of-training overhead. In practice, for the training phase, DSpar achieves up to $5.9\times$ faster than baseline with almost no accuracy drop. For the inference phase, DSpar reduces up to $90\%$ latency.

IROS Conference 2023 Conference Paper

Reinforcement Learning Based Multi-Layer Bayesian Control for Snake Robots in Cluttered Scenes

Jessica Ziyu Qu
William Ziming Qu
Li Li
Yuanyuan Jia

The majority of current research on reinforcement learning (RL) for snake robot control do not sufficiently account for the spatial and temporal dependencies within the robot or its interaction with its environment during movement. To address this issue, we propose an RL based multi-layer Bayesian method for autonomous snake robot control, which handles challenging scenarios and improves navigation efficiency. There are three major contributions: 1) An innovative hierarchical Bayesian framework unifies gait control, locomotion control, and stimulus reaction; 2) The dynamics of environment is modeled by density propagation and exploited by an LSTM-based agent to improve the learning process; 3) A stimulus reaction model is derived by combining spatial correlation among robot modules and temporal dependency along time sequence. Comparison experiments with a simulated snake robot show that performance of the proposed approach with challenging obstacles is superior to state-of-the-art baseline.

IS Journal 2023 Journal Article

Simulation Driven AI: From Artificial to Actual and Vice Versa

Li Li
Yilun Lin
Yutong Wang
Fei-Yue Wang

In this perspective, we discuss the important role of simulations in building state-of-the-art artificial intelligence (AI) systems. We first explain why simulations become vital in building complex AI systems. Then, we study some challenges and candidate solutions related to simulation-based AI systems. Finally, we discuss future research directions in this field.

NeurIPS Conference 2023 Conference Paper

VPGTrans: Transfer Visual Prompt Generator across LLMs

Ao Zhang
Hao Fei
Yuan Yao
Wei Ji
Li Li
Zhiyuan Liu
Tat-Seng Chua

Since developing a new multimodal LLM (MLLM) by pre-training on tremendous image-text pairs from scratch can be exceedingly resource-consuming, connecting an existing LLM with a comparatively lightweight visual prompt generator (VPG) becomes a feasible paradigm. However, further tuning the VPG component of the MLLM still incurs significant computational costs, such as thousands of GPU hours and millions of training data points. An alternative solution is transferring an existing VPG from one MLLM to the target MLLM. In this work, we investigate VPG transferability across LLMs for the first time, aiming to reduce the cost of VPG training. Specifically, we explore VPG transfer across different LLM sizes (e. g. , small-to-large) and types. We identify key factors to maximize transfer efficiency, based on which we develop a simple yet highly effective two-stage transfer framework, called VPGTrans. Notably, it enables VPG transfer from BLIP-2 OPT 2. 7B to BLIP-2 OPT 6. 7B with less than 10% of the GPU hours using only 10. 7% of the training data compared to training a VPG for OPT 6. 7B from scratch. Furthermore, we provide a series of intriguing findings and discuss potential explanations behind them. Finally, we showcase the practical value of our VPGTrans approach, by customizing two novel MLLMs, including VL-LLaMA and VL-Vicuna, with recently released LLaMA and Vicuna LLMs.

IJCAI Conference 2022 Conference Paper

Table2Graph: Transforming Tabular Data to Unified Weighted Graph

Kaixiong Zhou
Zirui Liu
Rui Chen
Li Li
Soo-Hyun Choi
Xia Hu

Learning useful interactions between input features is crucial for tabular data modeling. Recent efforts start to explicitly model the feature interactions with graph, where each feature is treated as an individual node. However, the existing graph construction methods either heuristically formulate a fixed feature-interaction graph based on specific domain knowledge, or simply apply attention function to compute the pairwise feature similarities for each sample. While the fixed graph may be sub-optimal to downstream tasks, the sample-wise graph construction is time-consuming during model training and inference. To tackle these issues, we propose a framework named Table2Graph to transform the feature interaction modeling to learning a unified graph. Represented as a probability adjacency matrix, the unified graph learns to model the key feature interactions shared by the diverse samples in the tabular data. To well optimize the unified graph, we employ the reinforcement learning policy to capture the key feature interactions stably. A sparsity constraint is also proposed to regularize the learned graph from being overly-sparse/smooth. The experimental results in a variety of real-world applications demonstrate the effectiveness and efficiency of our Table2Graph, in terms of the prediction accuracy and feature interaction detection.

PDF Details DOI

NeurIPS Conference 2021 Conference Paper

Dirichlet Energy Constrained Learning for Deep Graph Neural Networks

Kaixiong Zhou
Xiao Huang
Daochen Zha
Rui Chen
Li Li
Soo-Hyun Choi
Xia Hu

Graph neural networks (GNNs) integrate deep architectures and topological structure modeling in an effective way. However, the performance of existing GNNs would decrease significantly when they stack many layers, because of the over-smoothing issue. Node embeddings tend to converge to similar vectors when GNNs keep recursively aggregating the representations of neighbors. To enable deep GNNs, several methods have been explored recently. But they are developed from either techniques in convolutional neural networks or heuristic strategies. There is no generalizable and theoretical principle to guide the design of deep GNNs. To this end, we analyze the bottleneck of deep GNNs by leveraging the Dirichlet energy of node embeddings, and propose a generalizable principle to guide the training of deep GNNs. Based on it, a novel deep GNN framework -- Energetic Graph Neural Networks (EGNN) is designed. It could provide lower and upper constraints in terms of Dirichlet energy at each layer to avoid over-smoothing. Experimental results demonstrate that EGNN achieves state-of-the-art performance by using deep layers.

AAAI Conference 2021 Short Paper

Melodic Phrase Attention Network for Symbolic Data-based Music Genre Classification (Student Abstract)

Li Li
Rui Zhang
Zhenyu Wang

Compared with audio data-based music genre classification, researches on symbolic data-based music are scarce. Existing methods generally utilize manually extracted features, which is very time-consuming and laborious, and use traditional classifiers for label prediction without considering specific music features. To tackle this issue, we propose the Melodic Phrase Attention Network (MPAN) for symbolic data-based music genre classification. Our model is trained in three steps: First, we adopt representation learning, instead of the traditional musical feature extraction method, to obtain a vectorized representation of the music pieces. Second, the music pieces are divided into several melodic phrases through melody segmentation. Finally, the Melodic Phrase Attention Network is designed according to music characteristics, to identify the reflection of each melodic phrase on the music genre, thereby generating more accurate predictions. Experimental results show that our proposed method is superior to baseline symbolic data-based music genre classification approaches, and has achieved significant performance improvements on two large datasets.

NeurIPS Conference 2021 Conference Paper

Towards understanding retrosynthesis by energy-based models

Ruoxi Sun
Hanjun Dai
Li Li
Steven Kearnes
Bo Dai

Retrosynthesis is the process of identifying a set of reactants to synthesize a target molecule. It is of vital importance to material design and drug discovery. Existing machine learning approaches based on language models and graph neural networks have achieved encouraging results. However, the inner connections of these models are rarely discussed, and rigorous evaluations of these models are largely in need. In this paper, we propose a framework that unifies sequence- and graph-based methods as energy-based models (EBMs) with different energy functions. This unified view establishes connections and reveals the differences between models, thereby enhancing our understanding of model design. We also provide a comprehensive assessment of performance to the community. Moreover, we present a novel dual variant within the framework that performs consistent training to induce the agreement between forward- and backward-prediction. This model improves the state-of-the-art of template-free methods with or without reaction types.

EAAI Journal 2020 Journal Article

A novel and effective optimization algorithm for global optimization and its engineering applications: Turbulent Flow of Water-based Optimization (TFWO)

Mojtaba Ghasemi
Iraj Faraji Davoudkhani
Ebrahim Akbari
Abolfazl Rahimnejad
Sahand Ghavidel
Li Li

In this study we present a new and effective grouping optimization algorithm (namely, the Turbulent Flow of Water-based Optimization (TFWO)), inspired from a nature search phenomenon, i. e. whirlpools created in turbulent flow of water, for global real-world optimization problems. In the proposed algorithm, the problem of selecting control parameters is eliminated, the convergence power is increased and the algorithm have a fixed structure. The proposed algorithm is used to find the global solutions of real-parameter benchmark functions with different dimensions. Besides, in order to further investigate the effectiveness of TFWO, it was used to solve various types of nonlinear Economic Load Dispatch (ELD) optimization problems in power systems and Reliability–RedundancyAllocation Optimization (RRAO) for the overspeed protection system of a gas turbine, as two real-world engineering optimization problems. The results of TFWO are compared with other algorithms, which provide evidence for efficient performance with superior solution quality of the proposed TFWO algorithm in solving a great range of real-parameter benchmark and real-world engineering problems. Also, the results prove the competitive performance and robustness of TFWO algorithm compared to other state-of-the-art optimization algorithms in this study. The source codes of the TFWO algorithm are publicly available at https: //github. com/ebrahimakbary/TFWO.

EAAI Journal 2020 Journal Article

A novel hybrid optimization approach for reactive power dispatch problem considering voltage stability index

Mostafa Nasouri Gilvaei
Hossein Jafari
Mojtaba Jabbari Ghadi
Li Li

This paper proposes a novel, reliable, and effective hybrid approach based on the integration of the firefly algorithm (FA) and the adaptive particularly tunable fuzzy particle swarm optimization (APT-FPSO) method to address reactive power dispatch (RPD) problem, a crucial optimization problem in the operation of power systems. Similar to many other original meta-heuristic optimization techniques, the standard FA suffers from some severe drawbacks, most importantly being easily trapped into a locally optimal solution. In order to tackle these difficulties, in the current study, an improved version of fuzzy-based particle swarm optimization is utilized in the internal structure of the original FA. The developed hybrid approach, which is capable of avoiding premature convergence of the original FA by enhancing exploration and exploitation procedures, is employed to determine the optimum control variables (i. e. , the voltage of generation buses, tap positions of tap-changer transformers, and reactive power output of shunt compensators) through optimizing three distinct objective functions consisting of total transmission real power loss, the voltage magnitude deviations as well as voltage stability index. To validate the accuracy and competency of the proposed hybrid approach, it is firstly used for solving several benchmark optimization functions and then applied to three test systems at different scales, consisting of IEEE 30-bus, IEEE 57-bus, and IEEE 118-bus power systems, for solving the RPD problem. Eventually, the results of the presented hybrid method will be compared to those obtained by other implemented swarm intelligence-based approaches. The statistical analysis of this research substantiates the robustness and effectiveness of the developed algorithm to handle sophisticated optimization problems, particularly the RPD problem.

IJCAI Conference 2020 Conference Paper

Can Cross Entropy Loss Be Robust to Label Noise?

Lei Feng
Senlin Shu
Zhuoyi Lin
Fengmao Lv
Li Li
Bo An

Trained with the standard cross entropy loss, deep neural networks can achieve great performance on correctly labeled data. However, if the training data is corrupted with label noise, deep models tend to overfit the noisy labels, thereby achieving poor generation performance. To remedy this issue, several loss functions have been proposed and demonstrated to be robust to label noise. Although most of the robust loss functions stem from Categorical Cross Entropy (CCE) loss, they fail to embody the intrinsic relationships between CCE and other loss functions. In this paper, we propose a general framework dubbed Taylor cross entropy loss to train deep models in the presence of label noise. Specifically, our framework enables to weight the extent of fitting the training labels by controlling the order of Taylor Series for CCE, hence it can be robust to label noise. In addition, our framework clearly reveals the intrinsic relationships between CCE and other loss functions, such as Mean Absolute Error (MAE) and Mean Squared Error (MSE). Moreover, we present a detailed theoretical analysis to certify the robustness of this framework. Extensive experimental results on benchmark datasets demonstrate that our proposed approach significantly outperforms the state-of-the-art counterparts.

PDF Details DOI

IS Journal 2020 Journal Article

Joint Intelligence Ranking by Federated Multiplicative Update

Chi Zhang
Yu Liu
Le Wang
Yuehu Liu
Li Li
Nanning Zheng

The joint intelligence ranking of intelligent systems like autonomous driving is of great importance for building a more general, extensive, and universally accepted intelligence evaluation scheme. However, due to issues such as privacy security and industry or area competition, the integration of isolated test results may face large unimaginable difficulty in information security and encrypted model training. To address this, we derive the federated multiplicative update (FMU) algorithm with boundary constraints to solve the nonnegative matrix factorization based joint intelligence ranking. The encrypted learning process is developed to alternate original computation steps in multiplicative update algorithms. Owning feasible property for the fast convergence and secure exchange of variables, the proposed framework outperforms the previous work on both real and simulated data. Further experimental analysis reveals that the introduced federated mechanism does not harm the overall time efficiency.

JMLR Journal 2020 Journal Article

Learning Data-adaptive Non-parametric Kernels

Fanghui Liu
Xiaolin Huang
Chen Gong
Jie Yang
Li Li

In this paper, we propose a data-adaptive non-parametric kernel learning framework in margin based kernel methods. In model formulation, given an initial kernel matrix, a data-adaptive matrix with two constraints is imposed in an entry-wise scheme. Learning this data-adaptive matrix in a formulation-free strategy enlarges the margin between classes and thus improves the model flexibility. The introduced two constraints are imposed either exactly (on small data sets) or approximately (on large data sets) in our model, which provides a controllable trade-off between model flexibility and complexity with theoretical demonstration. In algorithm optimization, the objective function of our learning framework is proven to be gradient-Lipschitz continuous. Thereby, kernel and classifier/regressor learning can be efficiently optimized in a unified framework via Nesterov's acceleration. For the scalability issue, we study a decomposition-based approach to our model in the large sample case. The effectiveness of this approximation is illustrated by both empirical studies and theoretical guarantees. Experimental results on various classification and regression benchmark data sets demonstrate that our non-parametric kernel learning framework achieves good performance when compared with other representative kernel learning based algorithms. [abs] [ pdf ][ bib ] &copy JMLR 2020. ( edit, beta )

AAAI Conference 2018 Conference Paper

Nonlinear Pairwise Layer and Its Training for Kernel Learning

Fanghui Liu
Xiaolin Huang
Chen Gong
Jie Yang
Li Li

Kernel learning is a fundamental technique that has been intensively studied in the past decades. For the complicated practical tasks, the traditional “shallow” kernels (e. g. , Gaussian kernel and sigmoid kernel) are not ﬂexible enough to produce satisfactory performance. To address this shortcoming, this paper introduces a nonlinear layer in kernel learning to enhance the model ﬂexibility. This layer is pairwise, which fully considers the coupling information among examples. So our model contains a ﬁxed single mapping layer (i. e. a Gaussian kernel) as well as a nonlinear pairwise layer, thereby achieving better ﬂexibility than the existing kernel structures. Moreover, the proposed structure can be seamlessly embedded to Support Vector Machines (SVM), of which the training process can be formulated as a joint optimization problem including nonlinear function learning and standard SVM optimization. We theoretically prove that the objective function is gradient-Lipschitz continuous, which further guides us how to accelerate the optimization process in a deep kernel architecture. Experimentally, we ﬁnd that the proposed structure outperforms other state-ofthe-art kernel-based algorithms on various benchmark datasets, and thus the effectiveness of the incorporated pairwise layer with its training approach is demonstrated.

YNICL Journal 2017 Journal Article

Radiation-induced abnormal cortical thickness in patients with nasopharyngeal carcinoma after radiotherapy

Jiabao Lin
Xiaofei Lv
Meiqi Niu
Lizhi Liu
Jun Chen
Fei Xie
Miao Zhong
Shijun Qiu

Conventional MRI studies showed that radiation-induced brain necrosis in patients with nasopharyngeal carcinoma (NPC) in years after radiotherapy (RT) could involve brain gray matter (GM) and impair brain function. However, it is still unclear the radiation-induced brain morphological changes in NPC patients with normal-appearing GM in the early period after RT. In this study, we acquired high-resolution brain structural MRI data from three groups of patients, 22 before radiotherapy (pre-RT) NPC patients with newly diagnosed but not yet medically treated, 22 NPC patients in the early-delayed stage after radiotherapy (post-RT-ED), and 20 NPC patients in the late-delayed stage after radiotherapy (post-RT-LD), and then analyzed the radiation-induced cortical thickness alteration in NPC patients after RT. Using a vertex-wise surface-based morphometry (SBM) approach, we detected significantly decreased cortical thickness in the precentral gyrus (PreCG) in the post-RT-ED group compared to the pre-RT group. And the post-RT-LD group showed significantly increased cortical thickness in widespread brain regions, including the bilateral inferior parietal, left isthmus of the cingulate, left bank of the superior temporal sulcus and left lateral occipital regions, compared to the pre-RT group, and in the bilateral PreCG compared to the post-RT-ED group. Similar analysis with ROI-wise SBM method also found the consistent results. These results indicated that radiation-induced brain injury mainly occurred in the post-RT-LD group and the cortical thickness alterations after RT were dynamic in different periods. Our findings may reflect the pathogenesis of radiation-induced brain injury in NPC patients with normal-appearing GM and an early intervention is necessary for protecting GM during RT.

EAAI Journal 2016 Journal Article

A genetic algorithm for the multi-objective optimization of mixed-model assembly line based on the mental workload

Xiaosong Zhao
Chia-Yu Hsu
Pei-Chann Chang
Li Li

The increasing complexity of product varieties and productions leads to higher mental workload in the mixed-model assembly line (MMAL). Mental workload can improve product quality and guarantee the efficiency simultaneously. However, little research has been done on balancing the production quality and efficiency based on the effect of mental workload and complexity in the MMAL. This study aims to propose a mathematical model to formulate the multi-objective MMAL problem and the genetic algorithm is applied for problem solving due to the computational complexities. A numerical example is used to demonstrate the effectiveness of the proposed approach. The results show that incorporating the impact of mental workload on performance into account can make the rolled throughput yield (RTY) and efficiency balance when designing the MMAL. Moreover, we also verify that improving the experience of the operators can mitigate the impact of mental workload on the quality and efficiency.

AAAI Conference 2016 Conference Paper

BRBA: A Blocking-Based Association Rule Hiding Method

Peng Cheng
Ivan Lee
Li Li
Kuo-Kun Tseng
Jeng-Shyang Pan

Privacy preserving in association rule mining is an important research topic in the database security ﬁeld. This paper has proposed a blocking-based method to solve the association rule hiding problem for data sharing. It aims at reducing undesirable side effects and increasing desirable side effects, while ensuring to conceal all sensitive rules. The candidate transactions are selected for sanitization based on their relations with border rules. Comparative experiments on real datasets demonstrate that the proposed method can achieve its goals.

IS Journal 2011 Journal Article

Toward Cognitive Vehicles

Ding Wen
Gongjun Yan
Nan-Ning Zheng
Lin-Cheng Shen
Li Li

As a result of more cars on the road, traffic becomes more congested and streets become more risky. In addition, new communication and entertainment applications make drivers ever-more over-burdened and distracted. To relieve the continually increasing stress on drivers and reduce the number of accidents, current intelligent vehicle research is attempting to understand and model drivers. This article surveys recent works on cognitive vehicles that model drivers in a stimuli-decision-reaction mode and, on vehicle system side, improve perception, suggestion, and function delegation of traffic environment. The authors illustrate the relationships between recent models and methods and list related research challenges, while introducing applications of the driver-cognition models in intelligent vehicle control systems.

LORI Conference 2009 Conference Paper

From the Logical Point of View: The Chain Store Paradox Revisited

Li Li
Robert C. Koons
Jianjun Zhang

Abstract The standard approach to a rational action paradox in game theory (namely, the chain store paradox) has presupposed that the player’s beliefs are probabilities represented by functions with values between 0 and 1. However, a general solution must include the possibility that the subjective probabilities take only the values 1 and 0, requiring a non-Bayesian account of belief revision. In this paper, we propose a situation-theoretic diagnosis and solution to the paradox, based on the conception of Austinian propositions relativized to particular situations, as developed by Barwise and Etchemendy.

IS Journal 2009 Journal Article

IVS 09: Future Research in Vehicle Vision Systems

Zhiheng Li
Li Li
Yi Zhang

Visual sensing plays an essential role in intelligent vehicles. With the aid of visual sensors, driver assistance systems can alert the driver to dangerous situations or actions (such as swerving out of the lane or disregarding traffic signs or lights), or even independently take control of the vehicle. Currently, research on vehicle vision systems is receiving continuous global interest. This paper discusses some future directions for this field based on trends we observed at the 2008 IEEE Conference on Intelligent Transportation Systems (ITSC 08) and the 2009 IEEE Intelligent Vehicles Symposium (IVS 09). Our aim is to attract researchers to several critical questions that are important yet difficult. Solving these problems could be of great benefit to both academia and the vehicle industries.

AAAI Conference 2008 Short Paper

Generating Plans in Concurrent, Probabilistic, Over-Subscribed Domains

Li Li

Planning in realistic domains involves reasoning under uncertainty, operating under time and resource constraints, and finding the optimal set of goals to be achieved. In this paper, we provide an AO* based algorithm that can deal with durative actions, concurrent execution, over-subscribed goals, and probabilistic outcomes in a unified way. We explore plan optimization by introducing two novel aspects to the model. First, we introduce parallel steps that serve the same goal and increase the probability of success in addition to parallel steps that serve different goals and decrease execution time. Second, we introduce plan steps to terminate concurrent steps that are no longer useful so that resources can be conserved. Our algorithm called CPOAO* (Concurrent, Probabilistic, Oversubscription AO*) can deal with the aforementioned extensions and relies on the AO* framework to reduce the size of the search space using informative heuristic functions. We describe our framework, implementation, the heuristic functions we use, the experimental results, and potential research on heuristics that can further reduce the size of search space.

AAAI Conference 2008 Conference Paper

Generating Plans in Concurrent, Probabilistic, Over-Subscribed Domains

Li Li

Planning in realistic domains involves reasoning under uncertainty, operating under time and resource constraints, and finding the optimal set of goals to be achieved. In this paper, we provide an AO* based algorithm that can deal with durative actions, concurrent execution, over-subscribed goals, and probabilistic outcomes in a unified way. We explore plan optimization by introducing two novel aspects to the model. First, we introduce parallel steps that serve the same goal and increase the probability of success in addition to parallel steps that serve different goals and decrease execution time. Second, we introduce plan steps to terminate concurrent steps that are no longer useful so that resources can be conserved. Our algorithm called CPOAO* (Concurrent, Probabilistic, Oversubscription AO*) can deal with the aforementioned extensions and relies on the AO* framework to reduce the size of the search space using informative heuristic functions. We describe our framework, implementation, the heuristic functions we use, the experimental results, and potential research on heuristics that can further reduce the size of search space.

IS Journal 2005 Journal Article

IVS 05: new developments and research trends for intelligent vehicles

Li Li
Jingyan Song
Fei-Yue Wang
Wolfgang Niehsen
Nan-Ning Zheng

We discuss several selected topics from IVS 05 to provide a broad overview of intelligent-vehicle research perspectives and innovative projects. Specifically, we focus on advances in vehicle sensing, vehicle motion control and communications, and driver assistance and monitoring.

IS Journal 2003 Journal Article

Creating a digital-vehicle proving ground

Fei-Yue Wang
Xiaojing Wang
Li Li
P. Mirchandani

This installment presents the state of the art of ITS research in China, particularly the facilities and the proving ground for testing automated vehicles. To combine their strengths, in 2002 the ITSC, the Chinese Academy of Sciences, and the University of Arizona agreed to conduct joint research on a digital automobile proving ground (DAPG) for automated-vehicle driving tests based on their Beijing and Tucson facilities. The paper describes this international collaboration's status and progress.