Arrow Research search

Author name cluster

Yi Liu

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.

83 papers
2 author rows

Possible papers

83

AAAI Conference 2026 Conference Paper

AIM: Manifold-based Data Filtering for Representation Finetuning

  • Qing Li
  • Qibin Zheng
  • Yi Liu
  • Xingchun Diao

Representation Finetuning (ReFT) has recently emerged as an efficient paradigm for adapting pretrained language models by editing hidden representations rather than model weights. However, our preliminary experiments reveal that ReFT is notably more sensitive to training data quality compared to traditional parameter-efficient finetuning methods, particularly to samples with incorrect labels, which can severely degrade performance. Inspired by prior work demonstrating that the hidden representations of generalizable neural networks exhibit low-dimensional manifold structures, we hypothesize that effective generalization in ReFT requires geometrically structured transformations between pre- and post-intervention representations. This implies that the intervention vectors representing these transformations should form a low-dimensional manifold, rendering the inconsistent transformations induced by label noise as detectable geometric outliers. To leverage this insight, we introduce Aligning Interventions on a learned Manifold (AIM), a representation-based data filtering method for ReFT, which identifies high-quality training samples by measuring the geometric consistency of their intervention vectors with respect to a robust reference manifold derived via principal component analysis on trusted data. Extensive experiments on both commonsense and arithmetic reasoning tasks confirm the effectiveness of AIM, showing consistent improvements over strong data selection baselines across multiple model scales.

PDF Details DOI

AAAI Conference 2026 Conference Paper

Answering the Unanswerable Is to Err Knowingly: Analyzing and Mitigating Abstention Failures in Large Reasoning Models

  • Yi Liu
  • Xiangyu Liu
  • Zequn Sun
  • Wei Hu

Large reasoning models (LRMs) have shown remarkable progress on complex reasoning tasks. However, some questions posed to LRMs are inherently unanswerable, such as math problems lacking sufficient conditions. We find that LRMs continually fail to provide appropriate abstentions when confronted with these unanswerable questions. In this paper, we systematically analyze, investigate, and resolve this issue for trustworthy AI. We first conduct a detailed analysis of the distinct response behaviors of LRMs when facing unanswerable questions. Then, we show that LRMs possess sufficient cognitive capabilities to recognize the flaws in these questions. However, they fail to exhibit appropriate abstention behavior, revealing a misalignment between their internal cognition and external response. Finally, to resolve this issue, we propose a lightweight, two-stage method that combines cognitive monitoring with inference-time intervention. Experimental results demonstrate that our method significantly improves the abstention rate while maintaining the reasoning performance.

PDF Details DOI

EAAI Journal 2026 Journal Article

Dynamic heuristic phased Double Deep Q Network path planning algorithm based on Gaussian mixture regression in discrete traffic environment

  • Ruixin Zhang
  • Qing Xu
  • Kai Sun
  • Zhilin He
  • Yi Liu
  • Youneng Su
  • Jingyi Wang
  • Xinming Zhu

In outdoor path planning, effectively balancing the relationship between path capacity and path length remains a significant challenge in current research. To address this challenge, we propose a phased DDQN (Double Deep Q Network) path planning algorithm that integrates Gaussian mixture regression and dynamic heuristics within a discrete traffic environment: GMR-DDQN (Double deep Q network with Gaussian mixture regression). First, the influence of land cover and terrain on vehicle capacity is considered, leading to the formulation of a reward function and the development of a multi-coupled discrete toll cost model. Next, toll costs are incorporated into the DDQN using a greedy strategy, while heuristic methods are applied to generate multiple demonstration trajectories with minimal training. Then, Gaussian mixture regression is employed to gather demonstration trajectory data, construct the proposed path, and generate a buffer-constrained sampling area. The model is further trained to refine the trajectories. Finally, the performance of the algorithm is evaluated in three different traffic environments. Experimental results indicate that, compared to the improved DQN (Deep Q Network) and DDQN algorithms, GMR-DDQN reduces code execution time by 50. 27 %–74. 93 %, shortens path passage time by 27. 12 %–37. 29 %, decreases the average number of steps by 78. 12 %–88. 33 %, and increases the average reward by at least 18, 000 points. Additionally, compared to the unconstrained GMR-DDQN, GMR-DDQN cuts code running time by 25. 79 %–50. 27 %. The results demonstrate that GMR-DDQN can efficiently utilize the road network to enhance traffic efficiency.

Details DOI

EAAI Journal 2026 Journal Article

Dynamic multi-prototype guided domain incremental learning for Electroencephalogram-based disease classification

  • Fan Yang
  • Anping Zeng
  • Chunlin He
  • Yi Liu
  • Chaorong Li
  • Xingjie Wang

Electroencephalogram (EEG) plays a pivotal role in the early screening, clinical diagnosis, and prognostic evaluation of neurological disorders. Although EEG-based classification algorithms have achieved remarkable progress in recent years, existing models are primarily designed for static offline scenarios and struggle to adapt to the dynamic characteristics of evolving data distributions over time in clinical settings. While continual learning offers a potential solution, the significant inter-individual variability, non-stationarity, and temporal heterogeneity of EEG signals pose challenges to existing continual learning methods in terms of model adaptability, stability, and the balance between old and new knowledge. To address these issues, this paper proposes a dynamic multi-prototype guided domain-incremental learning method for continual EEG series classification, which employs an evolvable multi-prototype representation guidance mechanism to steer the model. Specifically, we first design a multi-prototype representation strategy that maintains multiple prototypes per class and integrates momentum updates with similarity gating mechanisms to achieve continuous optimization of prototype representations, thereby precisely capturing the dynamic intra-class distribution evolution. Next, we adopt a decoupled training framework for the feature extractor and classifier, leveraging prototype-guided mechanisms to encourage the feature extractor to learn stable inter-task shared representations. Finally, we construct a nearest prototype contrastive loss function to enhance the model’s discriminative capability for decision boundaries and feature structures by optimizing intra-class compactness and inter-class separability. Extensive experimental evaluations on four benchmark datasets demonstrate the effectiveness and efficiency of our proposed method.

Details DOI

AAAI Conference 2026 Conference Paper

DynamicRTL: RTL Representation Learning for Dynamic Circuit Behavior

  • Ruiyang Ma
  • Yunhao Zhou
  • Yipeng Wang
  • Yi Liu
  • Zhengyuan Shi
  • Ziyang Zheng
  • Kexin Chen
  • Zhiqiang He

There is a growing body of work on using Graph Neural Networks (GNNs) to learn representations of circuits, focusing primarily on their static characteristics. However, these models fail to capture circuit runtime behavior, which is crucial for tasks like circuit verification and optimization. To address this limitation, we introduce DR-GNN (DynamicRTL-GNN), a novel approach that learns RTL circuit representations by incorporating both static structures and multi-cycle execution behaviors. DR-GNN leverages an operator-level Control Data Flow Graph (CDFG) to represent Register Transfer Level (RTL) circuits, enabling the model to capture dynamic dependencies and runtime execution. To train and evaluate DR-GNN, we build the first comprehensive dynamic circuit dataset, comprising over 6,300 Verilog designs and 63,000 simulation traces. Our results demonstrate that DR-GNN outperforms existing models in branch hit prediction and toggle rate prediction. Furthermore, its learned representations transfer effectively to related dynamic circuit tasks, achieving strong performance in power estimation and assertion prediction.

PDF Details DOI

AAAI Conference 2026 Conference Paper

Finding Time Series Anomalies Using Granular-Ball Vector Data Description

  • Lifeng Shen
  • Liang Peng
  • Ruiwen Liu
  • Shuyin Xia
  • Yi Liu

Modeling normal behavior in dynamic, nonlinear time series data is challenging for effective anomaly detection. Traditional methods, such as nearest neighbor and clustering approaches, often depend on rigid assumptions, such as a predefined number of reliable neighbors or clusters, which frequently break down in complex temporal scenarios. To address these limitations, we introduce the Granular-ball One-Class Network (GBOC), a novel approach based on a data-adaptive representation called Granular-ball Vector Data Description (GVDD). GVDD partitions the latent space into compact, high-density regions represented by granular-balls, which are generated through a density-guided hierarchical splitting process and refined by removing noisy structures. Each granular-ball serves as a prototype for local normal behavior, naturally positioning itself between individual instances and clusters while preserving the local topological structure of the sample set. During training, GBOC improves the compactness of representations by aligning samples with their nearest granular-ball centers. During inference, anomaly scores are computed based on the distance to the nearest granular-ball. By focusing on dense, high-quality regions and significantly reducing the number of prototypes, GBOC delivers both robustness and efficiency in anomaly detection. Extensive experiments validate the effectiveness and superiority of the proposed method, highlighting its ability to handle the challenges of time series anomaly detection.

PDF Details DOI

AAAI Conference 2026 Conference Paper

FineRef: Fine-Grained Error Reflection and Correction for Long-Form Generation with Citations

  • Yixing Peng
  • Licheng Zhang
  • Shancheng Fang
  • Yi Liu
  • Peijian Gu
  • Quan Wang

Generating with citations is crucial for trustworthy Large Language Models (LLMs), yet even advanced LLMs often produce mismatched or irrelevant citations. Existing methods over-optimize citation fidelity while overlooking relevance to the user query, which degrades answer quality and robustness in real-world settings with noisy or irrelevant retrieved content. Moreover, the prevailing single-pass paradigm struggles to deliver optimal answers in long-form generation that requiring multiple citations. To address these limitations, we propose FineRef, a framework based on Fine-grained error Reflection, which explicitly teaches the model to self-identify and correct two key citation errors—mismatch and irrelevance—on a per-citation basis. FineRef follows a two-stage training strategy. The first stage instills an “attempt–reflect–correct” behavioral pattern via supervised fine-tuning, using fine-grained and controllable reflection data constructed by specialized lightweight models. An online self-reflective bootstrapping strategy is designed to improve generalization by iteratively enriching training data with verified, self-improving examples. To further enhance the self-reflection and correction capability, the second stage applies process-level reinforcement learning with a multi-dimensional reward scheme that promotes reflection accuracy, answer quality, and correction gain. Experiments on the ALCE benchmark demonstrate that FineRef significantly improves both citation performance and answer accuracy. Our 7B model outperforms GPT-4 by up to 18% in Citation F1 and 4% in EM Recall, while also surpassing the state-of-the-art model across key evaluation metrics. FineRef also exhibits strong generalization and robustness in domain transfer settings and noisy retrieval scenarios.

PDF Details DOI

AAAI Conference 2026 Conference Paper

In-Token Rationality Optimization: Towards Accurate and Concise LLM Reasoning via Self-Feedback

  • Mingye Zhu
  • Yi Liu
  • Zheren Fu
  • Quan Wang
  • Yongdong Zhang

Training Large Language Models (LLMs) for chain-of-thought reasoning presents a significant challenge: supervised fine-tuning on a single "golden" rationale hurts generalization as it penalizes equally valid alternatives, whereas reinforcement learning with verifiable rewards struggles with credit assignment and prohibitive computational cost. To tackle these limitations, we introduce InTRO (In-Token Rationality Optimization), a new framework that enables both token-level exploration and self-feedback for accurate and concise reasoning. Instead of directly optimizing an intractable objective over all valid reasoning paths, InTRO leverages correction factors—token-wise importance weights estimated by the information discrepancy between the generative policy and its answer-conditioned counterpart, for informative next-token selection. This approach allows the model to perform token-level exploration and receive self-generated feedback within a single forward pass, ultimately encouraging accurate and concise rationales. Across six math-reasoning benchmarks, InTRO consistently outperforms other baselines, raising solution accuracy by up to 20% relative to the base model. Its chains of thought are also notably more concise, exhibiting reduced verbosity. Beyond this, InTRO enables cross-domain transfer, successfully adapting to out-of-domain reasoning tasks that extend beyond the realm of mathematics, demonstrating robust generalization.

PDF Details DOI

JBHI Journal 2026 Journal Article

Piezoelectric Ceramic Sensor Array Based Obstructive Sleep Apnea Event Detection

  • Yi Liu
  • Zhengdong Li
  • Xiaomao Fan
  • Yingying Shao
  • Dikun Hu
  • Rong Huang
  • Yi Xiao
  • Boxuan Lv

Obstructive sleep apnea (OSA) is one of the major sleep disorders, which has been demonstrated to be a high-risk factor for cardiovascular disease, hypertension, and motor vehicle accidents. Pressure sensors in a contactless manner are a promising way to monitor sleep conditions outside of the hospital. However, previous studies mainly based on limited sensors are often subjected to noise contamination and constrained by the sleeper position to pressure sensors. The acquired pressure signals are of poor quality or even lost, which are not appropriate for the downstream task of OSA event detection. To address this issue, we designed a sensitive piezoelectric ceramic sensor array (PCSA) by aligning sixteen sensors embedded into a mat covering the chest and abdomen area, which can capture the changes of weak pressure signals under a sleeping mattress with a thickness of up to 30 cm. Based on PCSA, we recruited 36 adult volunteers from the Peking Union Medical College Hospital and conducted a pilot study to acquire overnight pressure signals along with polysomnography recordings. Subsequently, we developed an automated OSA event detection method named DRFNet. The main advantage of DRFNet is that it can well capture the time-domain and frequency-domain features from different views by fusing ResNet18 and DenseNet121 networks. Experiment results showed that DRFNet can achieve 75. 19 % sensitivity, 87. 78 % specificity, and 81. 48 % accuracy, which is competitive with existing state-of-the-art methods. Combined with PCSA, it can be potentially deployed into an embedded device and provide contactless sleep monitoring service in home settings.

Details DOI

AAAI Conference 2026 Conference Paper

ProtSAE: Disentangling and Interpreting Protein Language Models via Semantically-Guided Sparse Autoencoders

  • Xiangyu Liu
  • Haodi Lei
  • Yi Liu
  • Yang Liu
  • Wei Hu

Sparse Autoencoder (SAE) has emerged as a powerful tool for mechanistic interpretability of large language models. Recent works apply SAE to protein language models (PLMs), aiming to extract and analyze biologically meaningful features from their latent spaces. However, SAE suffers from semantic entanglement, where individual neurons often mix multiple nonlinear concepts, making it difficult to reliably interpret or manipulate model behaviors. In this paper, we propose a semantically-guided SAE, called ProtSAE. Unlike existing SAE which requires annotation datasets to filter and interpret activations, we guide semantic disentanglement during training using both annotation datasets and domain knowledge to mitigate the effects of entangled attributes. We design interpretability experiments showing that ProtSAE learns more biologically relevant and interpretable hidden features compared to previous methods. Performance analyses further demonstrate that ProtSAE maintains high reconstruction fidelity while achieving better results in interpretable probing. We also show the potential of ProtSAE in steering PLMs for downstream generation tasks.

PDF Details DOI

AAAI Conference 2026 Conference Paper

Sortblock: Similarity-Aware Feature Reuse for Diffusion Model

  • Hanqi Chen
  • Xu Zhang
  • Xiaoliu Guan
  • Lielin Jiang
  • Guanzhong Wang
  • Zeyu Chen
  • Yi Liu

Diffusion Transformers (DiTs) have demonstrated remarkable generative capabilities, particularly benefiting from Transformer architectures that enhance visual and artistic fidelity. However, their inherently sequential denoising process results in high inference latency, limiting their deployment in real-time scenarios. Existing training-free acceleration approaches typically reuse intermediate features at fixed timesteps or layers, overlooking the evolving semantic focus across denoising stages and Transformer blocks.To address this, we propose Sortblock, a training-free inference acceleration framework that dynamically caches block-wise features based on their similarity across adjacent timesteps. By ranking the evolution of residuals, Sortblock adaptively determines a recomputation ratio, selectively skipping redundant computations while preserving generation quality. Furthermore, we incorporate a lightweight linear prediction mechanism to reduce accumulated errors in skipped blocks.Extensive experiments across various tasks and DiT architectures demonstrate that Sortblock achieves over 2 times inference speedup with minimal degradation in output quality, offering an effective and generalizable solution for accelerating diffusion-based generative models.

PDF Details DOI

EAAI Journal 2026 Journal Article

Spatiotemporal physics-guided graph neural network for wind farm power prediction

  • Yi Liu
  • Yingning Qiu
  • Yanhui Feng
  • Huanyu Zhou
  • Xue-Lu Xiong

Recent efforts to enhance the interpretability of Graph Neural Networks (GNN) have focused on boosting their trustworthiness and traceability while maintaining predictive accuracy. However, mainstream approaches largely rely on data-driven strategies, with limited integration of physical prior knowledge or critical examination of GNN interpretability within physics-constrained frameworks. This paper introduces a novel Physics-guided GNN that incorporates spatiotemporal wind farm dynamics using a cutting-edge three-dimensional wake analytical model to guide the GNN architecture. This integration ensures compliance with physical laws during training, reducing the uncertainty and complexity associated with purely data-driven learning and addressing scalability challenges. By employing spatiotemporal directed local subgraphs and physics-induced attention weight learning, the model effectively considers the spatiotemporal wake coupling processes in wind farms, enabling high-accuracy power prediction. The proposed model outperforms other neural network structures in predicting both overall wind farm power production and individual turbine output. This research offers an efficient solution for power prediction in complex wind farms and demonstrates the potential of embedding domain-specific physical knowledge into GNN across broader multi-physics scenarios. It provides valuable insights into integrating physical theories with GNNs, enhancing model precision and transparency.

Details DOI

AAAI Conference 2026 Conference Paper

STEAMROLLER: A Multi-Agent System for Inclusive Automatic Speech Recognition for People Who Stutter

  • Ziqi Xu
  • Yi Liu
  • Yuekang Li
  • Ling Shi
  • Kailong Wang
  • Yongxin Zhao

People who stutter (PWS) face systemic exclusion in today’s voice-driven society, where access to voice assistants, authentication systems, and remote work tools increasingly depends on fluent speech. Current automatic speech recognition (ASR) systems, trained predominantly on fluent speech, fail to serve millions of PWS worldwide. We present STEAMROLLER, a real time system that transforms stuttered speech into fluent output through a novel multi-stage, multi-agent AI pipeline. Our approach addresses three critical technical challenges: (1) the difficulty of direct speech to speech conversion for disfluent input, (2) semantic distortions introduced during ASR transcription of stuttered speech, and (3) latency constraints for real time communication. STEAMROLLER employs a three stage architecture comprising ASR transcription, multi-agent text repair, and speech synthesis, where our core innovation lies in a collaborative multi-agent framework that iteratively refines transcripts while preserving semantic intent. Experiments on the FluencyBank dataset and a user study demonstrates clear word error rate (WER) reduction and strong user satisfaction. Beyond immediate accessibility benefits, fine tuning ASR on STEAMROLLER repaired speech further yields additional WER improvements, creating a pathway toward inclusive AI ecosystems.

PDF Details DOI

AAAI Conference 2026 Conference Paper

Synergizing Multigrid Algorithms with Vision Transformer: A Novel Approach to Enhance the Seismic Foundation Model

  • Huiwen Wu
  • Shuo Zhang
  • Yi Liu
  • Hongbin Ye

Due to the rapid advancement and homogenization of Artificial Intelligence (AI) technology development, transformer-based foundation models have revolutionized scientific applications, such as drug discovery, materials research, and astronomy. However, seismic data presents unique characteristics that require specialized processing techniques for pretraining foundation models in seismic contexts with high- and low-frequency features playing crucial roles. Existing Vision Transformer (ViT) with sequential image tokenization fails to efficiently and effectively capture both high- and low-frequency seismic information because they ignore the intrinsic structural patterns of seismograms. This work introduces ADATG, a novel adaptive two-grid training strategy with Hilbert encoding, explicitly tailored for seismogram data and leveraging the hierarchical structures inherent in seismic data. Specifically, our approach employs spectrum decomposition to separate high- and low-frequency components, and hierarchical Hilbert encoding to represent the data effectively. Moreover, inspired by the frequency principle, we propose an adaptive training strategy that initially emphasizes coarse-level information and then progressively refines the model's focus on fine-level features. Extensive experiments demonstrate the effectiveness and efficiency of our method. This research highlights the importance of data encoding and training strategies informed by the distinct characteristics of high- and low-frequency features in seismic images, ultimately enhancing the pretraining of visual seismic foundation models.

PDF Details DOI

AAAI Conference 2026 Conference Paper

Who Should I Trust? Explicit Confidence-Focused Multimodal Intent Recognition

  • Yi Liu
  • Qimeng Yang
  • Lanlan Lu

Multimodal intent recognition is aimed at understanding user intentions by integrating information from multiple modalities. It has attracted increasing attention in recently developed dialog systems. The existing studies have focused mainly on modeling semantic interactions within and across modalities, but they often overlook the reliability of each modality. In real-world scenarios, inputs may be corrupted by noisy audio, blurred or occluded videos, or ambiguous text, making it difficult for the employed model to determine who to trust and how much to trust. To address this challenge, we propose a method called explicit confidence-focused multimodal intent recognition (ECFMIR). The core idea of this approach is to assign each modality and each cross-modal associations feature a dedicated confidence lens (CLens) that explicitly estimates the confidence level in a hypothetical manner. This design helps reduce the degree of uncertainty and mitigate the risk of incorrect predictions when addressing conflicting inputs. Comprehensive experiments conducted on two benchmark multimodal intent recognition datasets demonstrate the effectiveness of our method. A further analysis reveals that ECFMIR achieves significant advantages for high-conflict categories and under low-resource conditions.

PDF Details DOI

EAAI Journal 2025 Journal Article

A multimodal industrial anomaly detection method based on mask training and teacher–student joint memory

  • Yi Liu
  • Changsheng Zhang
  • Xingjun Dong
  • Yufei Yang

In recent years, the teacher–student framework has been applied to both single-modality detection and multimodal detection, which realizes anomaly detection based on the feature difference between the teacher model and the student model. However, current multimodal teacher–student models use the same teacher model to extract two-dimensional (2D) image and three-dimensional (3D) point cloud features. The point cloud features extracted by the teacher model pre-trained on images are not the optimal feature representation. To further improve the performance of the teacher–student framework on the multimodal anomaly detection task, this paper proposes Multimodal Teacher-Student Joint Memory (MTSJM). MTSJM constructs a teacher–student joint memory bank for each modality, the feature distance between the test sample and the memory bank is used as the anomaly indicator. This distance reflects the feature differences between the test sample and the normal sample at multiple levels, including the teacher–teacher, teacher–student, and student–student levels. Then, this paper proposes a mask-based student model training method. While ensuring that the student learns the feature of normal regions, mask training increases the feature difference of non-normal regions between the student and the teacher. On the MVTec 3D Anomaly Detection (MVTec 3D-AD) dataset, the proposed MTSJM achieves effective anomaly detection performance, reaching 95. 7% mean Image-level Area Under the Receiver Operator Curve (I-AUROC) and 97. 2% mean Area Under the Per-Region Overlap (AUPRO). In addition, MTSJM achieves 99. 3% I-AUROC and 99. 6% Pixel-level AUROC (P-AUROC) on a real-world vehicle stamping part task, which further illustrates the applicability of MTSJM on the multimodal anomaly detection task.

Details DOI

EAAI Journal 2025 Journal Article

A novel convolutional neural network with global perception for bearing fault diagnosis

  • Xianguo Li
  • Ying Chen
  • Yi Liu

Bearings are key support components in rotating machinery, and their stability is crucial to the reliability of the entire mechanical system. To address the limitations of existing Transformer architectures in edge-side optimization and convolutional neural networks in global feature extraction, especially the resulting poor real-time performance and low accuracy in bearing fault diagnosis based on acoustic signals, this paper proposes a novel global-aware convolutional neural network based on residual masking and position-aware strategies (ParC-ReSMNet). Firstly, the network is based on the residual network (ResNet-18) and designed with a residual mask block combined with an improved spatial pyramid mask attention mechanism, which effectively removes redundant spatial information and focuses on critical fault features, thereby enhancing the robustness and generalization of the network. Secondly, a position-aware circular module is introduced to replace specific residual blocks in the original network, achieving an effective fusion of positional and global information, thereby augmenting the modeling capability of the convolutional neural network. Experiments are conducted on a self-made belt conveyor idler dataset and the Detection and Classification of Acoustic Scenes and Events (DCASE) 2023 Task2 bearing dataset, with results showing that ParC-ReSMNet achieves 95. 49% and 96. 67% accuracy, respectively. Compared to seven state-of-the-art models, it has the highest precision and recall, along with good real-time performance, which demonstrates great application value for fault monitoring of belt conveyors used in coal mines, power plants, ports, and other rotating machinery. The code library is available at: https: //github. com/xgli411/Parc-ResMNet.

Details DOI

EAAI Journal 2025 Journal Article

Asymmetric multimodal guidance fusion network for realtime visible and thermal semantic segmentation

  • Biao Yang
  • Yuanhui Guo
  • Rongrong Ni
  • Yi Liu
  • Guofa Li
  • Chuan Hu

In nighttime traffic scenarios, achieving real-time visible and thermal (RGB-Thermal) images semantic segmentation is a pivotal and intricate challenge within autonomous driving systems, demanding a delicate balance between precision and efficiency. However, previous methods have predominantly focused on enhancing model accuracy through increasing the number of parameters while neglecting the burden on inference time. Nevertheless, neglecting the guidance provided by semantic cues during the fusion process inevitably leads to imprecise segmentation. To improve the accuracy and speed of segmentation, this paper introduces a novel Asymmetric Guided Fusion Network (AGFNet) for real-time RGB-Thermal semantic segmentation. Specifically, we design an asymmetric dual-stream encoder to extract RGB and thermal features, thereby decreasing inference time. Afterward, we propose the Multi-scale pooling Cross-select Enhancement Module (MCEM) for mining semantic cues. For feature fusion, we introduce the Guidance Control Fusion Module (GCFM) to leverage semantic cues in guiding the fusion of RGB and thermal features, absorbing complementary relationships between the two modalities. AGFNet achieves 37. 4 frames per second (FPS), 57. 8% mean Intersection over Union (mIoU), and 48. 1 FPS, 56. 3% mIoU on the Multi spectral Fusion dataset(MFdataset), achieving a state-of-the-art balance between efficiency and accuracy. Our code will be available at https: //github. com/Mask0913/AGFNet.

Details DOI

JBHI Journal 2025 Journal Article

Attention-Guided 3D CNN With Lesion Feature Selection for Early Alzheimer's Disease Prediction Using Longitudinal sMRI

  • Jinwei Liu
  • Yashu Xu
  • Yi Liu
  • Huating Luo
  • Wenxiang Huang
  • Lizhong Yao

Predicting the progression from mild cognitive impairment (MCI) to Alzheimer's disease (AD) is critical for early intervention. Towards this end, various deep learning models have been applied in this domain, typically relying on structural magnetic resonance imaging (sMRI) data from a single time point whereas neglecting the dynamic changes in brain structure over time. Current longitudinal studies inadequately explore disease evolution dynamics and are burdened by high computational complexity. This paper introduces a novel lightweight 3D convolutional neural network specifically designed to capture the evolution of brain diseases for modeling the progression of MCI. First, a longitudinal lesion feature selection strategy is proposed to extract core features from temporal data, facilitating the detection of subtle differences in brain structure between two time points. Next, to refine the model for a more concentrated emphasis on lesion features, a disease trend attention mechanism is introduced to learn the dependencies between overall disease trends and local variation features. Finally, disease prediction visualization techniques are employed to improve the interpretability of the final predictions. Extensive experiments demonstrate that the proposed model achieves state-of-the-art performance in terms of area under the curve (AUC), accuracy, specificity, precision, and F1 score. This study confirms the efficacy of our early diagnostic method, utilizing only two follow-up sMRI scans to predict the disease status of MCI patients 24 months later with an AUC of 79. 03%.

Details DOI

AAAI Conference 2025 Conference Paper

CALM: Curiosity-Driven Auditing for Large Language Models

  • Xiang Zheng
  • Longxiang Wang
  • Yi Liu
  • Xingjun Ma
  • Chao Shen
  • Cong Wang

Auditing Large Language Models (LLMs) is a crucial and challenging task. In this study, we focus on auditing black-box LLMs without access to their parameters, only to the provided service. We treat this type of auditing as a black-box optimization problem where the goal is to automatically uncover input-output pairs of the target LLMs that exhibit illegal, immoral, or unsafe behaviors. For instance, we may seek a non-toxic input that the target LLM responds to with a toxic output or an input that induces the hallucinative response from the target LLM containing politically sensitive individuals. This black-box optimization is challenging due to the scarcity of feasible points, the discrete nature of the prompt space, and the large search space. To address these challenges, we propose Curiosity-Driven Auditing for Large Language Models (CALM), which uses intrinsically motivated reinforcement learning to finetune an LLM as the auditor agent to uncover potential harmful and biased input-output pairs of the target LLM. CALM successfully identifies derogatory completions involving celebrities and uncovers inputs that elicit specific names under the black-box setting. This work offers a promising direction for auditing black-box LLMs.

PDF Details DOI

AAAI Conference 2025 Conference Paper

Controllable Protein Sequence Generation with LLM Preference Optimization

  • Xiangyu Liu
  • Yi Liu
  • Silei Chen
  • Wei Hu

Designing proteins with specific attributes offers an important solution to address biomedical challenges. Pre-trained protein large language models (LLMs) have shown promising results on protein sequence generation. However, to control sequence generation for specific attributes, existing work still exhibits poor functionality and structural stability. In this paper, we propose a novel controllable protein design method called CtrlProt. We finetune a protein LLM with a new multi-listwise preference optimization strategy to improve generation quality and support multi-attribute controllable generation. Experiments demonstrate that CtrlProt can meet functionality and structural stability requirements effectively, achieving state-of-the-art performance in both single-attribute and multi-attribute protein sequence generation.

PDF Details DOI

IJCAI Conference 2025 Conference Paper

D3: Diversity, Difficulty, and Dependability-Aware Data Selection for Sample-Efficient LLM Instruction Tuning

  • Jia Zhang
  • Chen-Xi Zhang
  • Yao Liu
  • Yi-Xuan Jin
  • Xiao-Wen Yang
  • Bo Zheng
  • Yi Liu
  • Lan-Zhe Guo

Recent advancements in instruction tuning for large language models (LLMs) suggest that a small, high-quality dataset can significantly equip LLMs with instruction-following capabilities, outperforming large datasets often burdened by quality and redundancy issues. However, the challenge lies in automatically identifying valuable subsets from large datasets to boost both the effectiveness and efficiency of instruction tuning. In this paper, we first establish data selection criteria based on three distinct aspects of data value: diversity, difficulty, and dependability, and then propose the D3 method comprising two key steps of scoring and selection. Specifically, in the scoring step, we define the diversity function to measure sample distinctiveness and introduce the uncertainty-based prediction difficulty to evaluate sample difficulty by mitigating the interference of context-oriented generation diversity. Additionally, we integrate an external LLM for dependability assessment. In the selection step, we formulate the D3 weighted coreset objective, which jointly optimizes three aspects of data value to solve for the most valuable subset. The two steps of D3 can iterate multiple rounds, incorporating feedback to refine the selection focus adaptively. Experiments on both public datasets and the real-world Taobao Live application demonstrate the effectiveness of D3 in endowing LLMs with competitive or even superior instruction-following capabilities using less than 10% of the entire dataset.

PDF Details DOI

TMLR Journal 2025 Journal Article

Dynamic Schwartz-Fourier Neural Operator for Enhanced Expressive Power

  • Wenhan Gao
  • Jian Luo
  • Ruichen Xu
  • Yi Liu

Recently, neural operators have emerged as a prevailing approach for learning discretization-invariant mappings between function spaces. A particular example is the Fourier Neural Operator (FNO), which constrains integral kernels to be convolutions and learns the kernel directly in the frequency domain. Due to the capacity of Fourier transforms to effectively reduce the dimensionality and preserve information, FNOs demonstrate superior performance in terms of both efficiency and accuracy. In FNOs, the convolution kernel is fixed as a point-wise multiplication in the frequency domain; however, these translation-invariant kernels might limit the expressiveness of FNOs. For instance, if the underlying system lacks translational symmetries, the kernels learned by the FNO will still exhibit translational invariance, thereby limiting the model's expressive power. We propose a dynamic Schwartz operator that induces interactions between modes to enhance the expressiveness of FNOs. In this work, we introduce a novel approach that equips FNOs with Schwartz operators to learn dynamic kernels, termed Dynamic Kernel Fourier Neural Operators (DSFNOs). By incorporating this dynamic mechanism, our model gains the ability to capture relevant frequency information patterns, facilitating a better understanding and representation of complex physical phenomena. Through experiments, we demonstrate that DSFNOs can improve FNOs on a range of tasks, highlighting the effectiveness of our proposed approach. The code is available at https://github.com/wenhangao21/TMLR25_DSFNO.

PDF Details

TMLR Journal 2025 Journal Article

Efficient Exploration in Multi-Agent Reinforcement Learning via Farsighted Self-Direction

  • Tiancheng Lao
  • Xudong Guo
  • Mengge Liu
  • Junjie Yu
  • Yi Liu
  • Wenhui Fan

Multi-agent reinforcement learning faces greater challenges with efficient exploration compared to single-agent counterparts, primarily due to the exponential growth in state and action spaces. Methods based on intrinsic rewards have been proven to enhance exploration efficiency in multi-agent scenarios effectively. However, these methods are plagued by instability during training and biases in exploration direction. To address these challenges, we propose Farsighted Self-Direction (FSD), a novel model-free method that utilizes a long-term exploration bonus to achieve coordinated exploration. Since prediction error against individual Q-values indicates a potential bonus for committed exploration, it is taken into account in action selection to directly guide the coordinated exploration. Further, we also use clipped double Q-learning to reduce noise in prediction error. We validate the method on didactic examples and demonstrate the outperformance of our method on challenging StarCraft II micromanagement tasks.

PDF Details

NeurIPS Conference 2025 Conference Paper

Efficient Knowledge Transfer in Federated Recommendation for Joint Venture Ecosystem

  • Yichen Li
  • Yijing Shan
  • Yi Liu
  • Haozhao Wang
  • Cheng Wang
  • Yi Wang
  • Ruixuan Li

The current Federated Recommendation System (FedRS) focuses on personalized recommendation services and assumes clients are personalized IoT devices (e. g. , Mobile phones). In this paper, we deeply dive into new but practical FedRS applications within the joint venture ecosystem. Subsidiaries engage as participants with their users and items. However, in such a situation, merely exchanging item embedding is insufficient, as user bases always exhibit both overlaps and exclusive segments, demonstrating the complexity of user information. Meanwhile, directly uploading user information is a violation of privacy and unacceptable. To tackle the above challenges, we propose an efficient and privacy-enhanced federated recommendation for the joint venture ecosystem (FR-JVE) that each client transfers more common knowledge from other clients with a distilled user's \textit{rating preference} from the local dataset. More specifically, we first transform the local data into a new format and apply model inversion techniques to distill the rating preference with frozen user gradients before the federated training. Then, a bridge function is employed on each client side to align the local rating preference and aggregated global preference in a privacy-friendly manner. Finally, each client matches similar users to make a better prediction for overlapped users. From a theoretical perspective, we analyze how effectively FR-JVE can guarantee user privacy. Empirically, we show that FR-JVE achieves superior performance compared to state-of-the-art methods.

PDF Details

NeurIPS Conference 2025 Conference Paper

Enhancing Privacy in Multimodal Federated Learning with Information Theory

  • Tianzhe Xiao
  • Yichen Li
  • Yining Qi
  • Yi Liu
  • Haozhao Wang
  • Yi Wang
  • Ruixuan Li

Multimodal federated learning (MMFL) has gained increasing popularity due to its ability to leverage the correlation between various modalities, meanwhile preserving data privacy for different clients. However, recent studies show that correlation between modalities increase the vulnerability of federated learning against Gradient Inversion Attack (GIA). The complicated situation of MMFL privacy preserving can be summarized as follows: 1) different modality transmits different amounts of information, thus requires various protection strength; 2) correlation between modalities should be taken into account. This paper introduces an information theory perspective to analyze the leaked privacy in process of MMFL, and tries to propose a more reasonable protection method \textbf{Sec-MMFL} based on assessing different information leakage possibilities of each modality by conditional mutual information and adjust the corresponding protection strength. Moreover, we use mutual information to reduce the cross-modality information leakage in MMFL. Experiments have proven that our method can bring more balanced and comprehensive protection at an acceptable cost.

PDF Details

AAAI Conference 2025 Conference Paper

Exploring Enhanced Contextual Information for Video-Level Object Tracking

  • Ben Kang
  • Xin Chen
  • Simiao Lai
  • Yang Liu
  • Yi Liu
  • Dong Wang

Contextual information at the video level has become increasingly crucial for visual object tracking. However, existing methods typically use only a few tokens to convey this information, which can lead to information loss and limit their ability to fully capture the context. To address this issue, we propose a new video-level visual object tracking framework called MCITrack. It leverages Mamba's hidden states to continuously record and transmit extensive contextual information throughout the video stream, resulting in more robust object tracking. The core component of MCITrack is the Contextual Information Fusion module, which consists of the mamba layer and the cross-attention layer. The mamba layer stores historical contextual information, while the cross-attention layer integrates this information into the current visual features of each backbone block. This module enhances the model's ability to capture and utilize contextual information at multiple levels through deep integration with the backbone. Experiments demonstrate that MCITrack achieves competitive performance across numerous benchmarks. For instance, it gets 76.6% AUC on LaSOT and 80.0% AO on GOT-10k, establishing a new state-of-the-art performance.

PDF Details DOI

TIST Journal 2025 Journal Article

Knowledge Enhancement and Temporal Aware for Multi-Behavior Contrastive Recommendation

  • Hongrui Xuan
  • Bohan Li
  • Wenlong Wu
  • Yi Liu
  • Hongzhi Yin

A well-designed recommender system can accurately learn the embeddings of users and items, reflecting the unique preferences of users. Traditional recommendation techniques usually focus on modeling the singular type of behaviors between users and items. However, in many practical recommendation scenarios (e.g., social media, e-commerce), there exist multi-typed interactive behaviors in user–item relationships, such as click, tag-as-favorite, and purchase in online shopping platforms. Thus, how to make full use of multi-behavior information for recommendation is of great importance to the existing system, which presents challenges in two aspects that need to be explored: (1) Utilizing users’ personalized preferences to capture multi-behavioral dependencies; (2) Dealing with the insufficient recommendation caused by sparse supervision signal for target behavior. In this work, we propose the Knowledge Enhancement Multi-Behavior Contrastive Learning (KMCL) framework, including two Contrastive Learning tasks and three functional modules to tackle the above challenges, respectively. In particular, we design the multi-behavior learning module to extract users’ personalized behavior information for user-embedding enhancement and utilize knowledge graph in the knowledge enhancement module to derive more robust knowledge-aware representations for items. In addition, in the optimization stage, we also model the coarse-grained commonalities and the fine-grained differences between multi-behavior of users to further improve the recommendation effect and propose a joint training paradigm to enhance the learning effect of KMCLR in the joint learning module. Besides, we also considered how to make full use of temporal signals to enhance the effectiveness of multi-behavior recommendations in scenarios with time information and designed a novel encoder to address this issue. Extensive experiments and ablation tests on the three real-world datasets indicate that our KMCLR outperforms various state-of-the-art recommendation methods and verify the effectiveness of our method.

Details DOI

NeurIPS Conference 2025 Conference Paper

Leveraging Importance Sampling to Detach Alignment Modules from Large Language Models

  • Yi Liu
  • Dianqing Liu
  • Mingye Zhu
  • Junbo Guo
  • Yongdong Zhang
  • Zhendong Mao

The widespread adoption of large language models (LLMs) across industries has increased the demand for high-quality and customizable outputs. However, traditional alignment methods often require retraining large pretrained models, making it difficult to quickly adapt and optimize LLMs for diverse applications. To address this limitation, we propose a novel \textit{Residual Alignment Model} (\textit{RAM}) that formalizes the alignment process as a type of importance sampling. In this framework, the unaligned upstream model serves as the proposal distribution, while the alignment process is framed as secondary sampling based on an autoregressive alignment module that acts as an estimator of the importance weights. This design enables a natural detachment of the alignment module from the target aligned model, improving flexibility and scalability. Based on this model, we derive an efficient sequence-level training strategy for the alignment module, which operates independently of the proposal module. Additionally, we develop a resampling algorithm with iterative token-level decoding to address the common first-token latency issue in comparable methods. Experimental evaluations on two leading open-source LLMs across diverse tasks, including instruction following, domain adaptation, and preference optimization, demonstrate that our approach consistently outperforms baseline models.

PDF Details

NeurIPS Conference 2025 Conference Paper

Leveraging robust optimization for llm alignment under distribution shifts

  • Mingye Zhu
  • Yi Liu
  • Zheren Fu
  • Yongdong Zhang
  • Zhendong Mao

Preference alignment methods are increasingly critical for steering large language models (LLMs) to generate outputs consistent with human values. While recent approaches often rely on synthetic data generated by LLMs for scalability and cost-efficiency reasons, this reliance can introduce distributional shifts that undermine the nuanced representation of human preferences needed for desirable outputs. In this paper, we propose a novel distribution-aware optimization framework that improves preference alignment despite such shifts. Our approach first leverages well-learned classifiers to assign a calibration value to each training sample, quantifying its alignment with the target human-preferred distribution. These values are then incorporated into a robust optimization objective that minimizes the worst-case loss over regions of the data space most relevant to human preferences. By explicitly focusing optimization on the target distribution, our approach mitigates the impact of distributional mismatch and improves the generation of responses that better reflect intended values.

PDF Details

JBHI Journal 2025 Journal Article

LiMT: A Multi-Task Liver Image Benchmark Dataset

  • Zhe Liu
  • Kai Han
  • Siqi Ma
  • Yan Zhu
  • Jun Chen
  • Chongwen Lyu
  • Xinyi Qiu
  • Chengxuan Qian

Computer-aided diagnosis (CAD) technology can assist clinicians in evaluating liver lesions and intervening with treatment in time. Although CAD technology has advanced in recent years, the application scope of existing datasets remains relatively limited, typically supporting only single tasks, which has somewhat constrained the development of CAD technology. To address the above limitation, in this paper, we construct a multi-task liver dataset (LiMT) used for liver and tumor segmentation, multi-label lesion classification, and lesion detection based on arterial phase-enhanced computed tomography (CT), potentially providing an exploratory solution that is able to explore the correlation between tasks and does not need to worry about the heterogeneity between task-specific datasets during training. The dataset includes CT volumes from 150 different cases, comprising four types of liver diseases as well as normal cases. Each volume has been carefully annotated and calibrated by experienced clinicians. This public multi-task dataset may become a valuable resource for the medical imaging research community in the future. In addition, this paper not only provides relevant baseline experimental results but also reviews existing datasets and methods related to liver-related tasks. Our dataset is available at https://drive.google.com/drive/folders/1l9HRK13uaOQTNShf5pwgSz3OTanWjkag? usp=sharing.

Details DOI

NeurIPS Conference 2025 Conference Paper

PubSub-VFL: Towards Efficient Two-Party Split Learning in Heterogeneous Environments via Publisher/Subscriber Architecture

  • Yi Liu
  • Yang Liu
  • Leqian Zheng
  • Jue Hong
  • Junjie Shi
  • Qingyou Yang
  • Ye Wu
  • Cong Wang

With the rapid advancement of the digital economy, data collaboration between organizations has become a well-established business model, driving the growth of various industries. However, privacy concerns make direct data sharing impractical. To address this, Two-Party Split Learning (a. k. a. Vertical Federated Learning (VFL)) has emerged as a promising solution for secure collaborative learning. Despite its advantages, this architecture still suffers from low computational resource utilization and training efficiency. Specifically, its synchronous dependency design increases training latency, while resource and data heterogeneity among participants further hinder efficient computation. To overcome these challenges, we propose \texttt{PubSub-VFL}, a novel VFL paradigm with a Publisher/Subscriber architecture optimized for two-party collaborative learning with high computational efficiency. \texttt{PubSub-VFL} leverages the decoupling capabilities of the Pub/Sub architecture and the data parallelism of the parameter server architecture to design a hierarchical asynchronous mechanism, reducing training latency and improving system efficiency. Additionally, to mitigate the training imbalance caused by resource and data heterogeneity, we formalize an optimization problem based on participants’ system profiles, enabling the selection of optimal hyperparameters while preserving privacy. We conduct a theoretical analysis to demonstrate that \texttt{PubSub-VFL} achieves stable convergence and is compatible with security protocols such as differential privacy. Extensive case studies on five benchmark datasets further validate its effectiveness, showing that \texttt{PubSub-VFL} compared to state-of-the-art baselines not only accelerates training by $2 \sim 7\times$ without compromising accuracy but also achieves computational resource utilization by up to 91. 07\%.

PDF Details

IJCAI Conference 2025 Conference Paper

SAP: Privacy-Preserving Fine-Tuning on Language Models with Split-and-Privatize Framework

  • Xicong Shen
  • Yang Liu
  • Yi Liu
  • Peiran Wang
  • Huiqi Liu
  • Jue Hong
  • Bing Duan
  • Zirui Huang

Pre-trained Language Models (PLM) have enabled a cost-effective approach to handling various downstream applications via Parameter-Efficient-Fine-Tuning (PEFT) techniques. In this context, service providers have introduced a popular fine-tuning-based product service known as Model-as-a-Service (MaaS). This service offers users access to extensive PLMs and training resources. With MaaS, users can fine-tune, deploy, and utilize their customized models seamlessly, leveraging a one-stop platform that allows them to work with their private datasets efficiently. However, this service paradigm has recently been exposed to the possibility of leaking user private data. To this end, we identify the data privacy leakage risks in MaaS-based PEFT and propose a Split-and-Privatize (SAP) framework, mitigating the privacy leakage by integrating split learning and differential privacy into MaaS PEFT. Furthermore, we propose Contributing-Token-Identification (CTI), a novel method to balance model utility degradation and privacy leakage. As a result, the proposed framework is comprehensively evaluated, demonstrating a 65% improvement in empirical privacy with only a 1% degradation in model performance on the Stanford Sentiment Treebank dataset, outperforming existing state-of-the-art baselines.

PDF Details DOI

AAAI Conference 2025 Conference Paper

SUTrack: Towards Simple and Unified Single Object Tracking

  • Xin Chen
  • Ben Kang
  • Wanting Geng
  • Jiawen Zhu
  • Yi Liu
  • Dong Wang
  • Huchuan Lu

In this paper, we propose a simple yet unified single object tracking (SOT) framework, dubbed SUTrack. It consolidates five SOT tasks (RGB-based, RGB-Depth, RGB-Thermal, RGB-Event, RGB-Language Tracking) into a unified model trained in a single session. Due to the distinct nature of the data, current methods typically design individual architectures and train separate models for each task. This fragmentation results in redundant training processes, repetitive technological innovations, and limited cross-modal knowledge sharing. In contrast, SUTrack demonstrates that a single model with a unified input representation can effectively handle various SOT tasks, eliminating the need for task-specific designs and separate training sessions. Additionally, we introduce a task-recognition training strategy and a soft token type embedding to further enhance SUTrack's performance with minimal overhead. Experiments show that SUTrack outperforms previous task-specific counterparts across 11 datasets spanning five SOT tasks. Moreover, we provide a range of models catering edge devices as well as high-performance GPUs, striking a good trade-off between speed and accuracy. We hope SUTrack could serve as a strong foundation for further compelling research into unified tracking models.

PDF Details DOI

IJCAI Conference 2025 Conference Paper

Template3D-AD: Point Cloud Template Matching Method Based on Center Points for 3D Anomaly Detection

  • Yi Liu
  • Changsheng Zhang
  • Yufei Yang

Existing 3D anomaly detection methods mainly include reconstruction-based methods and memory-based methods. However, reconstruction-based methods rely on anomaly simulation strategies, while the memory bank of memory-based methods cannot cover the features of all points. Different from existing methods, this paper proposes Template3D-AD, a 3D anomaly detection method based on template matching. Template3D-AD matches the test sample with the template based on center points, and extracts the global features and local features of the center point respectively. Considering that the appearance of anomalies is related to the change of surface shape, this paper proposes a curvature-based local feature representation method, which increases the feature difference between abnormal surfaces and normal surfaces. Then, this paper designs a global-local detection strategy, which combines global feature differences and local feature differences for anomaly detection. Extensive experiments show that Template3D-AD outperforms the state-of-the-art methods, achieving 84. 4% (1. 5% ↑) I-AUROC on the Real3D-AD dataset and 86. 5% (11. 6% ↑) I-AUROC on the Anomaly-ShapeNet dataset. Code at https: //github. com/CaedmonLY/Template3D-AD.

PDF Details DOI

JBHI Journal 2025 Journal Article

Towards High-Quality MRI Reconstruction With Anisotropic Diffusion-Assisted Generative Adversarial Networks and Its Multi-Modal Images Extension

  • Yuyang Luo
  • Gengshen Wu
  • Yi Liu
  • Wenjian Liu
  • Jungong Han

Recently, fast Magnetic Resonance Imaging reconstruction technology has emerged as a promising way to improve the clinical diagnostic experience by significantly reducing scan times. While existing studies have used Generative Adversarial Networks to achieve impressive results in reconstructing MR images, they still suffer from challenges such as blurred zones/boundaries and abnormal spots caused by inevitable noise in the reconstruction process. To this end, we propose a novel deep framework termed Anisotropic Diffusion-Assisted Generative Adversarial Networks, which aims to maximally preserve valid high-frequency information and structural details while minimizing noises in reconstructed images by optimizing a joint loss function in a unified framework. In doing so, it enables more authentic and accurate MR image generation. To specifically handle unforeseeable noises, an Anisotropic Diffused Reconstruction Module is developed and added aside the backbone network as a denoise assistant, which improves the final image quality by minimizing reconstruction losses between targets and iteratively denoised generative outputs with no extra computational complexity during the testing phase. To make the most of valuable MRI data, we extend its application to support multi-modal learning to boost reconstructed image quality by aggregating more valid information from images of diverse modalities. Extensive experiments on public datasets show that the proposed framework can achieve superior performance in polishing up the quality of reconstructed MR images. For example, the proposed method obtains average PSNR and mSSIM values of 35. 785 dB and 0. 9765 on the MRNet dataset, which are at least about 2. 9 dB and 0. 07 higher than those from the baselines.

Details DOI

AAAI Conference 2025 Conference Paper

VHM: Versatile and Honest Vision Language Model for Remote Sensing Image Analysis

  • Chao Pang
  • Xingxing Weng
  • Jiang Wu
  • Jiayu Li
  • Yi Liu
  • Jiaxing Sun
  • Weijia Li
  • Shuai Wang

This paper develops a Versatile and Honest vision language Model (VHM) for remote sensing image analysis. VHM is built on a large-scale remote sensing image-text dataset with rich-content captions (VersaD), and an honest instruction dataset comprising both factual and deceptive questions (HnstD). Unlike prevailing remote sensing image-text datasets, in which image captions focus on a few prominent objects and their relationships, VersaD captions provide detailed information about image properties, object attributes, and the overall scene. This comprehensive captioning enables VHM to thoroughly understand remote sensing images and perform diverse remote sensing tasks. Moreover, different from existing remote sensing instruction datasets that only include factual questions, HnstD contains additional deceptive questions stemming from the non-existence of objects. This feature prevents VHM from producing affirmative answers to nonsense queries, thereby ensuring its honesty. In our experiments, VHM significantly outperforms various vision language models on common tasks of scene classification, visual question answering, and visual grounding. Additionally, VHM achieves competent performance on several unexplored tasks, such as building vectorizing, multi-label classification and honest question answering.

PDF Details DOI

EAAI Journal 2024 Journal Article

A novel failure mode and effect analysis model based on extended interval-valued q-rung orthopair fuzzy approach for risk analysis

  • Yuan Rong
  • Liying Yu
  • Yi Liu
  • Vladimir Simic
  • Dragan Pamucar
  • Harish Garg

The failure mode and effect analysis (FMEA) is an essential technique for risk assessment that can not only determine the potential failure modes, causes of failure and effects, but also expunge the potential failure modes in a system. Imprecise evaluation outcomes may be achieved due to the inherent imperfections caused by the traditional FMEA approach. Accordingly, a hybrid FMEA model based upon the interval-valued q-rung orthopair fuzzy (IVq-ROF) set and extended weighted aggregated sum product assessment (WASPAS) method is propounded to provide decision support for risk assessment. The main dedications of this investigation are as below. First, an innovative IVq-ROF score function is presented to compare the IVq-ROF numbers and the related properties of the proposed function are also discussed. Second, the IVq-ROF Aczel–Alsina operation laws are investigated and then some novel IVq-ROF Hamy mean operators are propounded on based on the advanced operations. Third, the IVq-ROF logarithmic percentage change-driven objective weighting (LOPCOW) method is advanced based on the presented score function to determine the objective importance of the risk criteria and the fuzzy full consistency method (FUCOM) is utilized to ascertain the subjective importance of the risk criteria. Fourth, a novel FMEA model based on extended WASPAS method is presented to ascertain the prioritization of failure modes. Lastly, an empirical about risk assessment in the shipbuilding process is implemented to verify the applicability and feasibility of the designed FMEA model. The comparison analysis and sensibility study are executed to demonstrate the significant merits and stability of the developed model. The best advantage of the presented model is that it not only obtains a consistent outcome and takes into account the interrelations among the risk criteria, but also addresses more problems where the weights of the criteria and experts are not determined in advance.

Details DOI

EAAI Journal 2024 Journal Article

A pharmaceutical cold-chain logistics service quality model using a q-rung orthopair fuzzy framework with distance measure

  • Yuan Rong
  • Liying Yu
  • Yi Liu
  • Vladimir Simic
  • Dragan Pamucar

The assessment of pharmaceutical cold chain logistics (PCCL) service quality can provide timely scientific and valid rectification suggestions for enterprises and promote the continuous development and improvement of PCCL systems. PCCL service quality evaluation can be seen as a multi-expert, multi-criteria decision-making process with significant uncertainty and ambiguity. Accordingly, this paper propounds an innovative q-rung orthopair fuzzy (q-ROF) assessment framework based on the improved decision-making trial and evaluation laboratory (DEMATEL) technique and Taxonomy approach. Firstly, a novel distance measure considering both the influence of interactions and the volatility of information is presented to distinguish between q-ROF sets. Then, the q-ROF-DEMATEL technique is proffered to estimate the subjective weights of assessment criteria. A comprehensive criteria weight technique based on game theory is developed by integrating the subjective weights obtained using the q-ROF-DEMATEL method with objective weights derived using a combination of two optimizations. Furthermore, based upon the combinative weight model, a multi-stage assessment framework is designed by fusing the combination weight model and a distance-measure-based Taxonomy approach. An empirical concerning the PCCL service quality assessment of enterprises for the medicine management department is implemented to verify the practicability of the designed evaluation framework. The comparison discussion with sensitivity analysis is implemented to exemplify that the proffered framework is a valid, feasible, and stable method for PCCL service quality assessment.

Details DOI

AAAI Conference 2024 Conference Paper

Analysis of Differentially Private Synthetic Data: A Measurement Error Approach

  • Yangdi Jiang
  • Yi Liu
  • Xiaodong Yan
  • Anne-Sophie Charest
  • Linglong Kong
  • Bei Jiang

Differentially private (DP) synthetic datasets have been receiving significant attention from academia, industry, and government. However, little is known about how to perform statistical inference using DP synthetic datasets. Naive approaches that do not take into account the induced uncertainty due to the DP mechanism will result in biased estimators and invalid inferences. In this paper, we present a class of maximum likelihood estimator (MLE)-based easy-to-implement bias-corrected DP estimators with valid asymptotic confidence intervals (CI) for parameters in regression settings, by establishing the connection between additive DP mechanisms and measurement error models. Our simulation shows that our estimator has comparable performance to the widely used sufficient statistic perturbation (SSP) algorithm in some scenarios but with the advantage of releasing a synthetic dataset and obtaining statistically valid asymptotic CIs, which can achieve better coverage when compared to the naive CIs obtained by ignoring the DP mechanism.

PDF Details DOI

AAAI Conference 2024 Conference Paper

Benchmarking Large Language Models on Controllable Generation under Diversified Instructions

  • Yihan Chen
  • Benfeng Xu
  • Quan Wang
  • Yi Liu
  • Zhendong Mao

While large language models (LLMs) have exhibited impressive instruction-following capabilities, it is still unclear whether and to what extent they can respond to explicit constraints that might be entailed in various instructions. As a significant aspect of LLM alignment, it is thus important to formulate such a specialized set of instructions as well as investigate the resulting behavior of LLMs. To address this vacancy, we propose a new benchmark CoDI-Eval to systematically and comprehensively evaluate LLMs' responses to instructions with various constraints. We construct a large collection of constraints-attributed instructions as a test suite focused on both generalization and coverage. Specifically, we advocate an instruction diversification process to synthesize diverse forms of constraint expression and also deliberate the candidate task taxonomy with even finer-grained sub-categories. Finally, we automate the entire evaluation process to facilitate further developments. Different from existing studies on controllable text generation, CoDI-Eval extends the scope to the prevalent instruction-following paradigm for the first time. We provide extensive evaluations of representative LLMs (e.g., ChatGPT, Vicuna) on CoDI-Eval, revealing their limitations in following instructions with specific constraints and there is still a significant gap between open-source and commercial closed-source LLMs. We believe this benchmark will facilitate research into improving the controllability of LLMs' responses to instructions. Our data and code are available at https://github.com/Xt-cyh/CoDI-Eval.

PDF Details DOI

EAAI Journal 2024 Journal Article

Class-specific feature selection using fuzzy information-theoretic metrics

  • Xi-Ao Ma
  • Hao Xu
  • Yi Liu
  • Justin Zuopeng Zhang

Fuzzy information-theoretic metrics have been demonstrated to be effective in evaluating feature relevance and redundancy in both categorical and numerical feature selection tasks. Most existing feature selection methods based on fuzzy information-theoretic metrics treat all classes as a single entity, resulting in the selection of a consistent feature subset for all classes. This approach overlooks the fact that diverse classes may exhibit distinct discriminative characteristics, which requires the selection of diverse feature subsets for each class. Consequently, these methods lack the capability to handle this variability in feature selection required for optimal performance. To address this limitation, this paper proposes a class-specific feature selection method based on fuzzy information-theoretic metrics. To be more specific, we introduce several class-specific fuzzy information-theoretic metrics. Building upon these metrics, we formulate a class-specific feature selection algorithm, which is referred to as class-specific fuzzy information-theoretic feature selection. This algorithm enables the selection of highly relevant feature subsets tailored to each individual class. Furthermore, we present a class-specific ensemble classification framework that integrates the classification results obtained from the feature subsets generated by our method. Finally, we conduct extensive tests on 21 publicly available datasets using four popular classifiers to compare the performance of our method with eight up-to-date classification-specific methods. Test results demonstrate that our method outperforms the eight compared classification-specific methods in terms of performance.

Details DOI

TMLR Journal 2024 Journal Article

Coordinate Transform Fourier Neural Operators for Symmetries in Physical Modelings

  • Wenhan Gao
  • Ruichen Xu
  • Hong Wang
  • Yi Liu

Symmetries often arise in many natural sciences; rather than relying on data augmentation or regularization for learning these symmetries, incorporating these inherent symmetries directly into the neural network architecture simplifies the learning process and enhances model performance. The laws of physics, including partial differential equations (PDEs), remain unchanged regardless of the coordinate system employed to depict them, and symmetries sometimes can be natural to illuminate in other coordinate systems. Moreover, symmetries often are associated with the underlying domain shapes. In this work, we consider physical modelings with neural operators (NOs), and we propose an approach based on coordinate transforms (CT) to work on different domain shapes and symmetries. Canonical coordinate transforms are applied to convert both the domain shape and symmetries. For example, a sphere can be naturally converted to a square with periodicities across its edges. The resulting CT-FNO scheme barely increases computational complexity and can be applied to different domain shapes while respecting the symmetries.

PDF Details

ICRA Conference 2024 Conference Paper

Distribution-Aware Continual Test-Time Adaptation for Semantic Segmentation

  • Jiayi Ni
  • Senqiao Yang
  • Ran Xu 0013
  • Jiaming Liu 0003
  • Xiaoqi Li 0020
  • Wenyu Jiao
  • Zehui Chen
  • Yi Liu

Since autonomous driving systems usually face dynamic and ever-changing environments, continual test-time adaptation (CTTA) has been proposed as a strategy for transferring deployed models to continually changing target domains. However, the pursuit of long-term adaptation often introduces catastrophic forgetting and error accumulation problems, which impede the practical implementation of CTTA in the real world. Recently, existing CTTA methods mainly focus on utilizing a majority of parameters to fit target domain knowledge through self-training. Unfortunately, these approaches often amplify the challenge of error accumulation due to noisy pseudo-labels, and pose practical limitations stemming from the heavy computational costs associated with entire model updates. In this paper, we propose a distribution-aware tuning (DAT) method to make the semantic segmentation CTTA efficient and practical in real-world applications. DAT adaptively selects and updates two small groups of trainable parameters based on data distribution during the continual adaptation process, including domain-specific parameters (DSP) and task-relevant parameters (TRP). Specifically, DSP exhibits sensitivity to outputs with substantial distribution shifts, effectively mitigating the problem of error accumulation. In contrast, TRP are allocated to positions that are responsive to outputs with minor distribution shifts, which are fine-tuned to avoid the catastrophic forgetting problem. In addition, since CTTA is a temporal task, we introduce the Parameter Accumulation Update (PAU) strategy to collect the updated DSP and TRP in target domain sequences. We conducted extensive experiments on two widely-used semantic segmentation CTTA benchmarks, achieving competitive performance and efficiency compared to previous state-of-the-art methods.

Details

NeurIPS Conference 2024 Conference Paper

Empowering Active Learning for 3D Molecular Graphs with Geometric Graph Isomorphism

  • Ronast Subedi
  • Lu Wei
  • Wenhan Gao
  • Shayok Chakraborty
  • Yi Liu

Molecular learning is pivotal in many real-world applications, such as drug discovery. Supervised learning requires heavy human annotation, which is particularly challenging for molecular data, e. g. , the commonly used density functional theory (DFT) is highly computationally expensive. Active learning (AL) automatically queries labels for most informative samples, thereby remarkably alleviating the annotation hurdle. In this paper, we present a principled AL paradigm for molecular learning, where we treat molecules as 3D molecular graphs. Specifically, we propose a new diversity sampling method to eliminate mutual redundancy built on distributions of 3D geometries. We first propose a set of new 3D graph isometries for 3D graph isomorphism analysis. Our method is provably at least as expressive as the Geometric Weisfeiler-Lehman (GWL) test. The moments of the distributions of the associated geometries are then extracted for efficient diversity computing. To ensure our AL paradigm selects samples with maximal uncertainties, we carefully design a Bayesian geometric graph neural network to compute uncertainties specifically for 3D molecular graphs. We pose active sampling as a quadratic programming (QP) problem using the proposed components. Experimental results demonstrate the effectiveness of our AL paradigm, as well as the proposed diversity and uncertainty methods.

PDF Details DOI

AAAI Conference 2024 Conference Paper

Gradual Residuals Alignment: A Dual-Stream Framework for GAN Inversion and Image Attribute Editing

  • Hao Li
  • Mengqi Huang
  • Lei Zhang
  • Bo Hu
  • Yi Liu
  • Zhendong Mao

GAN-based image attribute editing firstly leverages GAN Inversion to project real images into the latent space of GAN and then manipulates corresponding latent codes. Recent inversion methods mainly utilize additional high-bit features to improve image details preservation, as low-bit codes cannot faithfully reconstruct source images, leading to the loss of details. However, during editing, existing works fail to accurately complement the lost details and suffer from poor editability. The main reason is they inject all the lost details indiscriminately at one time, which inherently induces the position and quantity of details to overfit source images, resulting in inconsistent content and artifacts in edited images. This work argues that details should be gradually injected into both the reconstruction and editing process in a multi-stage coarse-to-fine manner for better detail preservation and high editability. Therefore, a novel dual-stream framework is proposed to accurately complement details at each stage. The Reconstruction Stream is employed to embed coarse-to-fine lost details into residual features and then adaptively add them to the GAN generator. In the Editing Stream, residual features are accurately aligned by our Selective Attention mechanism and then injected into the editing process in a multi-stage manner. Extensive experiments have shown the superiority of our framework in both reconstruction accuracy and editing quality compared with existing methods.

PDF Details DOI

EAAI Journal 2024 Journal Article

Improving real-time driver distraction detection via constrained attention mechanism

  • Hang Gao
  • Yi Liu

Real-time driving distraction detection has garnered significant attention due to its potential to build various driving safety protections such as distraction warnings and driver assistance systems. Recent studies have focused on the development of neural networks for vision-based detection, though achieving a balance between performance and efficiency has proven challenging. In this paper, we propose a novel constrained attention (CA) mechanism for real-time driver distraction detection, which aims to achieve better performance meanwhile ensuring the computation efficiency. Specifically, we conduct some case studies by generating class activation maps to check the model attention, and three potential factors affecting performance are mined, which are ambiguous attention signal, excessive attention region, and similar attention between different classes. Two regularization terms are designed to optimize the three obstacles. Firstly, a concentrative regularization is introduced to limit the size of the attention region, meanwhile, pixels in the region have clear attention degrees. Secondly, an orthogonal regularization is proposed to optimize the attention of different classes to be discriminative. To further inspiring the model, we design an intersample constraint, which optimizes the attention of images with the same ground truth to be similar. Experiments are conducted on two driver distraction detection datasets, and the experimental results showed that our CA mechanism can bring significant performance improvement. More importantly, there will be no additional computational burden when the trained model is deployed in actual scenarios. Codes are released at https: //github. com/gaohangcodes/CAN4DDD.

Details DOI

ICML Conference 2024 Conference Paper

Multi-Source Conformal Inference Under Distribution Shift

  • Yi Liu
  • Alexander Levis
  • Sharon-Lise T. Normand
  • Larry Han

Recent years have experienced increasing utilization of complex machine learning models across multiple sources of data to inform more generalizable decision-making. However, distribution shifts across data sources and privacy concerns related to sharing individual-level data, coupled with a lack of uncertainty quantification from machine learning predictions, make it challenging to achieve valid inferences in multi-source environments. In this paper, we consider the problem of obtaining distribution-free prediction intervals for a target population, leveraging multiple potentially biased data sources. We derive the efficient influence functions for the quantiles of unobserved outcomes in the target and source populations, and show that one can incorporate machine learning prediction algorithms in the estimation of nuisance functions while still achieving parametric rates of convergence to nominal coverage probabilities. Moreover, when conditional outcome invariance is violated, we propose a data-adaptive strategy to upweight informative data sources for efficiency gain and downweight non-informative data sources for bias reduction. We highlight the robustness and efficiency of our proposals for a variety of conformal scores and data-generating mechanisms via extensive synthetic experiments. Hospital length of stay prediction intervals for pediatric patients undergoing a high-risk cardiac surgical procedure between 2016-2022 in the U. S. illustrate the utility of our methodology.

Details

ICLR Conference 2024 Conference Paper

Neural Field Classifiers via Target Encoding and Classification Loss

  • Xindi Yang
  • Zeke Xie
  • Xiong Zhou
  • Boyu Liu
  • Buhua Liu
  • Yi Liu
  • Haoran Wang 0004
  • Yunfeng Cai

Neural field methods have seen great progress in various long-standing tasks in computer vision and computer graphics, including novel view synthesis and geometry reconstruction. As existing neural field methods try to predict some coordinate-based continuous target values, such as RGB for Neural Radiance Field (NeRF), all of these methods are regression models and are optimized by some regression loss. However, are regression models really better than classification models for neural field methods? In this work, we try to visit this very fundamental but overlooked question for neural fields from a machine learning perspective. We successfully propose a novel Neural Field Classifier (NFC) framework which formulates existing neural field methods as classification tasks rather than regression tasks. The proposed NFC can easily transform arbitrary Neural Field Regressor (NFR) into its classification variant via employing a novel Target Encoding module and optimizing a classification loss. By encoding a continuous regression target into a high-dimensional discrete encoding, we naturally formulate a multi-label classification task. Extensive experiments demonstrate the impressive effectiveness of NFC at the nearly free extra computational costs. Moreover, NFC also shows robustness to sparse inputs, corrupted images, and dynamic scenes.

Details

AAAI Conference 2024 Conference Paper

Towards Efficient and Effective Text-to-Video Retrieval with Coarse-to-Fine Visual Representation Learning

  • Kaibin Tian
  • Yanhua Cheng
  • Yi Liu
  • Xinglin Hou
  • Quan Chen
  • Han Li

In recent years, text-to-video retrieval methods based on CLIP have experienced rapid development. The primary direction of evolution is to exploit the much wider gamut of visual and textual cues to achieve alignment. Concretely, those methods with impressive performance often design a heavy fusion block for sentence (words)-video (frames) interaction, regardless of the prohibitive computation complexity. Nevertheless, these approaches are not optimal in terms of feature utilization and retrieval efficiency. To address this issue, we adopt multi-granularity visual feature learning, ensuring the model's comprehensiveness in capturing visual content features spanning from abstract to detailed levels during the training phase. To better leverage the multi-granularity features, we devise a two-stage retrieval architecture in the retrieval phase. This solution ingeniously balances the coarse and fine granularity of retrieval content. Moreover, it also strikes a harmonious equilibrium between retrieval effectiveness and efficiency. Specifically, in training phase, we design a parameter-free text-gated interaction block (TIB) for fine-grained video representation learning and embed an extra Pearson Constraint to optimize cross-modal representation learning. In retrieval phase, we use coarse-grained video representations for fast recall of top-k candidates, which are then reranked by fine-grained video representations. Extensive experiments on four benchmarks demonstrate the efficiency and effectiveness. Notably, our method achieves comparable performance with the current state-of-the-art methods while being nearly 50 times faster.

PDF Details DOI

ICML Conference 2024 Conference Paper

Tuning-free Estimation and Inference of Cumulative Distribution Function under Local Differential Privacy

  • Yi Liu
  • Qirui Hu
  • Linglong Kong

We introduce a novel algorithm for estimating Cumulative Distribution Function (CDF) values under Local Differential Privacy (LDP) by exploiting an unexpected connection between LDP and the current status problem, a classical survival data problem in statistics. This connection leads to the development of tools for constrained isotonic estimation based on binary queries. Through mathematical proofs and extensive numerical testing, we demonstrate that our method achieves uniform and $L_2$ error bounds when estimating the entire CDF curve. By employing increasingly dense grids, the error bound can be improved, exhibiting an asymptotic normal distribution of the proposed estimator. Theoretically, we show that the error bound smoothly changes as the number of grids increases relative to the sample size $n$. Computationally, we demonstrate that our constrained isotonic estimator can be efficiently computed deterministically, eliminating the need for hyperparameters or random optimization.

Details

ICRA Conference 2023 Conference Paper

Efficient Preference-Based Reinforcement Learning Using Learned Dynamics Models

  • Yi Liu
  • Gaurav Datta
  • Ellen R. Novoseller
  • Daniel S. Brown

Preference-based reinforcement learning (PbRL) can enable robots to learn to perform tasks based on an individual's preferences without requiring a hand-crafted re-ward function. However, existing approaches either assume access to a high-fidelity simulator or analytic model or take a model-free approach that requires extensive, possibly unsafe online environment interactions. In this paper, we study the benefits and challenges of using a learned dynamics model when performing PbRL. In particular, we provide evidence that a learned dynamics model offers the following benefits when performing PbRL: (1) preference elicitation and policy optimization require significantly fewer environment interactions than model-free PbRL, (2) diverse preference queries can be synthesized safely and efficiently as a byproduct of standard model-based RL, and (3) reward pre-training based on suboptimal demonstrations can be performed without any environmental interaction. Our paper provides empirical ev-idence that learned dynamics models enable robots to learn customized policies based on user preferences in ways that are safer and more sample efficient than prior preference learning approaches. Supplementary materials and code are available at https://sites.google.com/berkeley.edu/mop-rl.

Details

NeurIPS Conference 2023 Conference Paper

Gaussian Differential Privacy on Riemannian Manifolds

  • Yangdi Jiang
  • Xiaotian Chang
  • Yi Liu
  • Lei Ding
  • Linglong Kong
  • Bei Jiang

We develop an advanced approach for extending Gaussian Differential Privacy (GDP) to general Riemannian manifolds. The concept of GDP stands out as a prominent privacy definition that strongly warrants extension to manifold settings, due to its central limit properties. By harnessing the power of the renowned Bishop-Gromov theorem in geometric analysis, we propose a Riemannian Gaussian distribution that integrates the Riemannian distance, allowing us to achieve GDP in Riemannian manifolds with bounded Ricci curvature. To the best of our knowledge, this work marks the first instance of extending the GDP framework to accommodate general Riemannian manifolds, encompassing curved spaces, and circumventing the reliance on tangent space summaries. We provide a simple algorithm to evaluate the privacy budget $\mu$ on any one-dimensional manifold and introduce a versatile Markov Chain Monte Carlo (MCMC)-based algorithm to calculate $\mu$ on any Riemannian manifold with constant curvature. Through simulations on one of the most prevalent manifolds in statistics, the unit sphere $S^d$, we demonstrate the superior utility of our Riemannian Gaussian mechanism in comparison to the previously proposed Riemannian Laplace mechanism for implementing GDP.

PDF Details

TAAS Journal 2023 Journal Article

GLDAP: Global Dynamic Action Persistence Adaptation for Deep Reinforcement Learning

  • Junbo Tong
  • Daming Shi
  • Yi Liu
  • Wenhui Fan

In the implementation of deep reinforcement learning (DRL), action persistence strategies are often adopted so agents maintain their actions for a fixed or variable number of steps. The choice of the persistent duration for agent actions usually has notable effects on the performance of reinforcement learning algorithms. Aiming at the research gap of global dynamic optimal action persistence and its application in multi-agent systems, we propose a novel framework: global dynamic action persistence (GLDAP), which achieves global action persistence adaptation for deep reinforcement learning. We introduce a closed-loop method that is used to learn the estimated value and the corresponding policy of each candidate action persistence. Our experiment shows that GLDAP achieves an average of 2.5%~90.7% performance improvement and 3~20 times higher sampling efficiency over several baselines across various single-agent and multi-agent domains. We also validate the ability of GLDAP to determine the optimal action persistence through multiple experiments.

Details DOI

EAAI Journal 2023 Journal Article

Instance-specific algorithm configuration via unsupervised deep graph clustering

  • Wen Song
  • Yi Liu
  • Zhiguang Cao
  • Yaoxin Wu
  • Qiqiang Li

Instance-specific Algorithm Configuration (AC) methods are effective in automatically generating high-quality algorithm parameters for heterogeneous NP-hard problems from multiple sources. However, existing works rely on manually designed features to describe training instances, which are simple numerical attributes and cannot fully capture structural differences. Targeting at Mixed-Integer Programming (MIP) solvers, this paper proposes a novel instances-specific AC method based on end-to-end deep graph clustering. By representing an MIP instance as a bipartite graph, a random walk algorithm is designed to extract raw features with both numerical and structural information from the instance graph. Then an auto-encoder is designed to learn dense instance embeddings unsupervisedly, which facilitates clustering heterogeneous instances into homogeneous clusters for training instance-specific configurations. Experimental results on multiple benchmarks show that the proposed method can improve the solving efficiency of CPLEX on highly heterogeneous instances, and outperform existing instance specific AC methods.

Details DOI

ICML Conference 2023 Conference Paper

Online Local Differential Private Quantile Inference via Self-normalization

  • Yi Liu
  • Qirui Hu
  • Lei Ding 0013
  • Linglong Kong

Based on binary inquiries, we developed an algorithm to estimate population quantiles under Local Differential Privacy (LDP). By self-normalizing, our algorithm provides asymptotically normal estimation with valid inference, resulting in tight confidence intervals without the need for nuisance parameters to be estimated. Our proposed method can be conducted fully online, leading to high computational efficiency and minimal storage requirements with $\mathcal{O}(1)$ space. We also proved an optimality result by an elegant application of one central limit theorem of Gaussian Differential Privacy (GDP) when targeting the frequently encountered median estimation problem. With mathematical proof and extensive numerical testing, we demonstrate the validity of our algorithm both theoretically and experimentally.

Details

ICML Conference 2023 Conference Paper

Semi-Offline Reinforcement Learning for Optimized Text Generation

  • Changyu Chen
  • Xiting Wang
  • Yiqiao Jin
  • Victor Ye Dong
  • Li Dong
  • Jie Cao
  • Yi Liu
  • Rui Yan 0001

Existing reinforcement learning (RL) mainly utilize online or offline settings. The online methods explore the environment with expensive time cost, and the offline methods efficiently obtain reward signals by sacrificing the exploration capability. We propose semi-offline RL, a novel paradigm that can smoothly transit from the offline setting to the online setting, balances the exploration capability and training cost, and provides a theoretical foundation for comparing different RL settings. Based on the semi-offline MDP formulation, we present the RL setting that is optimal in terms of optimization cost, asymptotic error, and overfitting error bound. Extensive experiments show that our semi-offline RL approach is effective in various text generation tasks and datasets, and yields comparable or usually better performance compared with the state-of-the-art methods.

Details

JBHI Journal 2022 Journal Article

A Home-based Tele-rehabilitation System With Enhanced Therapist-patient Remote Interaction: A Feasibility Study

  • Yi Liu
  • Shuxiang Guo
  • Ziyi Yang
  • Hideyuki Hirata
  • Takashi Tamiya

As a promising alternative to hospital-based manual therapy, robot-assisted tele-rehabilitation therapy has shown significant benefits in reducing the therapist's workload and accelerating the patient's recovery process. However, existing telerobotic systems for rehabilitation face barriers to implementing appropriate therapy treatment due to the lack of effective therapist-patient interactive capabilities. In this paper, we develop a home-based tele-rehabilitation system that implements two alternative training methods, including a haptic-enabled guided training that allows the therapist to adjust the intensity of therapeutic movements provided by the rehabilitation device and a surface electromyography (sEMG)-based supervised training that explores remote assessment of the patient's kinesthetic awareness. Preliminary experiments were conducted to demonstrate the feasibility of the proposed alternative training methods and evaluate the functionality of the developed tele-rehabilitation system. Results showed that the proposed tele-rehabilitation system enabled therapist-in-the-loop to dynamically adjust the rehabilitation intensity and provided more interactivity in therapist-patient remote interaction.

Details DOI

JBHI Journal 2022 Journal Article

An Effective Semi-Supervised Approach for Liver CT Image Segmentation

  • Kai Han
  • Lu Liu
  • Yuqing Song
  • Yi Liu
  • Chengjian Qiu
  • Yangyang Tang
  • Qiaoying Teng
  • Zhe Liu

Despite the substantial progress made by deep networks in the field of medical image segmentation, they generally require sufficient pixel-level annotated data for training. The scale of training data remains to be the main bottleneck to obtain a better deep segmentation model. Semi-supervised learning is an effective approach that alleviates the dependence on labeled data. However, most existing semi-supervised image segmentation methods usually do not generate high-quality pseudo labels to expand training dataset. In this paper, we propose a deep semi-supervised approach for liver CT image segmentation by expanding pseudo-labeling algorithm under the very low annotated-data paradigm. Specifically, the output features of labeled images from the pretrained network combine with corresponding pixel-level annotations to produce class representations according to the mean operation. Then pseudo labels of unlabeled images are generated by calculating the distances between unlabeled feature vectors and each class representation. To further improve the quality of pseudo labels, we adopt a series of operations to optimize pseudo labels. A more accurate segmentation network is obtained by expanding the training dataset and adjusting the contributions between supervised and unsupervised loss. Besides, the novel random patch based on prior locations is introduced for unlabeled images in the training procedure. Extensive experiments show our method has achieved more competitive results compared with other semi-supervised methods when fewer labeled slices of LiTS dataset are available.

Details DOI

NeurIPS Conference 2022 Conference Paper

Boosting the Transferability of Adversarial Attacks with Reverse Adversarial Perturbation

  • Zeyu Qin
  • Yanbo Fan
  • Yi Liu
  • Li Shen
  • Yong Zhang
  • Jue Wang
  • Baoyuan Wu

Deep neural networks (DNNs) have been shown to be vulnerable to adversarial examples, which can produce erroneous predictions by injecting imperceptible perturbations. In this work, we study the transferability of adversarial examples, which is significant due to its threat to real-world applications where model architecture or parameters are usually unknown. Many existing works reveal that the adversarial examples are likely to overfit the surrogate model that they are generated from, limiting its transfer attack performance against different target models. To mitigate the overfitting of the surrogate model, we propose a novel attack method, dubbed reverse adversarial perturbation (RAP). Specifically, instead of minimizing the loss of a single adversarial point, we advocate seeking adversarial example located at a region with unified low loss value, by injecting the worst-case perturbation (the reverse adversarial perturbation) for each step of the optimization procedure. The adversarial attack with RAP is formulated as a min-max bi-level optimization problem. By integrating RAP into the iterative process for attacks, our method can find more stable adversarial examples which are less sensitive to the changes of decision boundary, mitigating the overfitting of the surrogate model. Comprehensive experimental comparisons demonstrate that RAP can significantly boost adversarial transferability. Furthermore, RAP can be naturally combined with many existing black-box attack techniques, to further boost the transferability. When attacking a real-world image recognition system, Google Cloud Vision API, we obtain 22% performance improvement of targeted attacks over the compared method. Our codes are available at https: //github. com/SCLBD/Transfer attack RAP.

PDF Details

NeurIPS Conference 2022 Conference Paper

ComENet: Towards Complete and Efficient Message Passing for 3D Molecular Graphs

  • Limei Wang
  • Yi Liu
  • Yuchao Lin
  • Haoran Liu
  • Shuiwang Ji

Many real-world data can be modeled as 3D graphs, but learning representations that incorporates 3D information completely and efficiently is challenging. Existing methods either use partial 3D information, or suffer from excessive computational cost. To incorporate 3D information completely and efficiently, we propose a novel message passing scheme that operates within 1-hop neighborhood. Our method guarantees full completeness of 3D information on 3D graphs by achieving global and local completeness. Notably, we propose the important rotation angles to fulfill global completeness. Additionally, we show that our method is orders of magnitude faster than prior methods. We provide rigorous proof of completeness and analysis of time complexity for our methods. As molecules are in essence quantum systems, we build the \underline{com}plete and \underline{e}fficient graph neural network (ComENet) by combing quantum inspired basis functions and the proposed message passing scheme. Experimental results demonstrate the capability and efficiency of ComENet, especially on real-world datasets that are large in both numbers and sizes of graphs. Our code is publicly available as part of the DIG library (\url{https: //github. com/divelab/DIG}).

PDF Details

JBHI Journal 2022 Journal Article

Data Integration Using Tensor Decomposition for the Prediction of miRNA-Disease Associations

  • Jiawei Luo
  • Yi Liu
  • Pei Liu
  • Zihan Lai
  • Hao Wu

Dysfunction of miRNAs has an important relationship with diseases by impacting their target genes. Identifying disease-related miRNAs is of great significance to prevent and treat diseases. Integrating information of genes related miRNAs and/or diseases in calculational methods for miRNA-disease association studies is meaningful because of the complexity of biological mechanisms. Therefore, in this study, we propose a novel method based on tensor decomposition, termed TDMDA, to integrate multi-type data for identifying pathogenic miRNAs. First, we construct a three-order association tensor to express the associations of miRNA-disease pairs, the associations of miRNA-gene pairs, and the associations of gene-disease pairs simultaneously. Then, a tensor decomposition-based method with auxiliary information is applied to reconstruct the association tensor for predicting miRNA-disease associations, and the auxiliary information includes biological similarity information and adjacency information. The performance of TDMDA is compared with other advanced methods under 5-fold cross-validations. The experimental results indicate the TDMDA is a competitive method.

Details DOI

ICRA Conference 2022 Conference Paper

GOMP-FIT: Grasp-Optimized Motion Planning for Fast Inertial Transport

  • Jeffrey Ichnowski
  • Yahav Avigal
  • Yi Liu
  • Ken Goldberg

High-speed motions in pick-and-place operations are critical to making robots cost-effective in many automation scenarios, from warehouses and manufacturing to hospitals and homes. However, motions can be too fast-such as when the object being transported has an open-top, is fragile, or both. One way to avoid spills or damage, is to move the arm slowly. We propose an alternative: Grasp-Optimized Motion Planning for Fast Inertial Transport (GOMP-FIT), a time-optimizing motion planner based on our prior work, that includes con-straints based on accelerations at the robot end-effector. With GOMP-FIT, a robot can perform high-speed motions that avoid obstacles and use inertial forces to its advantage. In experiments transporting open-top containers with varying tilt tolerances, whereas GOMP computes sub-second motions that spill up to 90 % of the contents during transport, GOMP-FIT generates motions that spill 0 % of contents while being slowed by as little as 0 % when there are few obstacles, 30 % when there are high obstacles and 45-degree tolerances, and 50 % when there 15-degree tolerances and few obstacles. Videos and more at: https://berkeleyautomation.github.io/gomp-fit/.

Details

NeurIPS Conference 2022 Conference Paper

Hierarchical Channel-spatial Encoding for Communication-efficient Collaborative Learning

  • Qihua Zhou
  • Song Guo
  • Yi Liu
  • Jie Zhang
  • Jiewei Zhang
  • Tao Guo
  • Zhenda Xu
  • Xun Liu

It witnesses that the collaborative learning (CL) systems often face the performance bottleneck of limited bandwidth, where multiple low-end devices continuously generate data and transmit intermediate features to the cloud for incremental training. To this end, improving the communication efficiency by reducing traffic size is one of the most crucial issues for realistic deployment. Existing systems mostly compress features at pixel level and ignore the characteristics of feature structure, which could be further exploited for more efficient compression. In this paper, we take new insights into implementing scalable CL systems through a hierarchical compression on features, termed Stripe-wise Group Quantization (SGQ). Different from previous unstructured quantization methods, SGQ captures both channel and spatial similarity in pixels, and simultaneously encodes features in these two levels to gain a much higher compression ratio. In particular, we refactor feature structure based on inter-channel similarity and bound the gradient deviation caused by quantization, in forward and backward passes, respectively. Such a double-stage pipeline makes SGQ hold a sublinear convergence order as the vanilla SGD-based optimization. Extensive experiments show that SGQ achieves a higher traffic reduction ratio by up to 15. 97 times and provides 9. 22 times image processing speedup over the uniform quantized training, while preserving adequate model accuracy as FP32 does, even using 4-bit quantization. This verifies that SGQ can be applied to a wide spectrum of edge intelligence applications.

PDF Details

AAAI Conference 2022 Conference Paper

How Does Knowledge Graph Embedding Extrapolate to Unseen Data: A Semantic Evidence View

  • Ren Li
  • Yanan Cao
  • Qiannan Zhu
  • Guanqun Bi
  • Fang Fang
  • Yi Liu
  • Qian Li

Knowledge Graph Embedding (KGE) aims to learn representations for entities and relations. Most KGE models have gained great success, especially on extrapolation scenarios. Specifically, given an unseen triple (h, r, t), a trained model can still correctly predict t from (h, r, ?), or h from (? , r, t), such extrapolation ability is impressive. However, most existing KGE works focus on the design of delicate triple modeling function, which mainly tells us how to measure the plausibility of observed triples, but offers limited explanation of why the methods can extrapolate to unseen data, and what are the important factors to help KGE extrapolate. Therefore in this work, we attempt to study the KGE extrapolation of two problems: 1. How does KGE extrapolate to unseen data? 2. How to design the KGE model with better extrapolation ability? For the problem 1, we first discuss the impact factors for extrapolation and from relation, entity and triple level respectively, propose three Semantic Evidences (SEs), which can be observed from train set and provide important semantic information for extrapolation. Then we verify the effectiveness of SEs through extensive experiments on several typical KGE methods. For the problem 2, to make better use of the three levels of SE, we propose a novel GNNbased KGE model, called Semantic Evidence aware Graph Neural Network (SE-GNN). In SE-GNN, each level of SE is modeled explicitly by the corresponding neighbor pattern, and merged sufficiently by the multi-layer aggregation, which contributes to obtaining more extrapolative knowledge representation. Finally, through extensive experiments on FB15k- 237 and WN18RR datasets, we show that SE-GNN achieves state-of-the-art performance on Knowledge Graph Completion task and performs a better extrapolation ability. Our code is available at https: //github. com/renli1024/SE-GNN.

PDF Details

NeurIPS Conference 2022 Conference Paper

Identification, Amplification and Measurement: A bridge to Gaussian Differential Privacy

  • Yi Liu
  • Ke Sun
  • Bei Jiang
  • Linglong Kong

Gaussian differential privacy (GDP) is a single-parameter family of privacy notions that provides coherent guarantees to avoid the exposure of sensitive individual information. Despite the extra interpretability and tighter bounds under composition GDP provides, many widely used mechanisms (e. g. , the Laplace mechanism) inherently provide GDP guarantees but often fail to take advantage of this new framework because their privacy guarantees were derived under a different background. In this paper, we study the asymptotic properties of privacy profiles and develop a simple criterion to identify algorithms with GDP properties. We propose an efficient method for GDP algorithms to narrow down possible values of an optimal privacy measurement, $\mu$ with an arbitrarily small and quantifiable margin of error. For non GDP algorithms, we provide a post-processing procedure that can amplify existing privacy guarantees to meet the GDP condition. As applications, we compare two single-parameter families of privacy notions, $\epsilon$-DP, and $\mu$-GDP, and show that all $\epsilon$-DP algorithms are intrinsically also GDP. Lastly, we show that the combination of our measurement process and the composition theorem of GDP is a powerful and convenient tool to handle compositions compared to the traditional standard and advanced composition theorems.

PDF Details

EAAI Journal 2022 Journal Article

MARCOS approach based upon cubic Fermatean fuzzy set and its application in evaluation and selecting cold chain logistics distribution center

  • Yuan Rong
  • Liying Yu
  • Wenyao Niu
  • Yi Liu
  • Tapan Senapati
  • Arunodaya Raj Mishra

The evaluation and selection of cold chain logistics distribution center (CCLDC) possesses important strategic significance for enterprises to optimize logistics network. This paper firstly introduces an innovative uncertain information representation model called cubic Fermatean fuzzy set (CCFS) by integrating the cubic fuzzy set (CFS) and Fermatean fuzzy set (FFS) to portray the complicated indeterminacy and inaccuracy information. To begin with, the definition, score and accuracy functions, comparison laws, and generalized distance measure of CFFS are all defined successively. Then, based upon the defined Aczel-Alsina operations of CFFS, several cubic Fermatean fuzzy Aczel-Alsina aggregation operators are presented to flexibly integrate the cubic Fermatean fuzzy information, as well as some elegant properties of those operators are investigated at length. Again, two weight identification models are introduced based upon the defined score function and distance measure to calculate the weight of experts and criteria. In addition, an extended measurement alternatives and ranking based on the compromise solution (MARCOS) method by means of the presented operators, score function, and distance measure is brought forward within cubic Fermatean fuzzy circumstance. Consequently, an empirical study about the evaluation and selection of CCLDC is applied to demonstrate the efficacy and practicability of the presented method. Moreover, the comparative study and sensibility analysis are implemented to validate the rationality and superiority of the propounded method in addressing the problem of selecting the optimal CCLDC. The outcomes reflect that the developed approach provides a synthetic, robust and operable group decision framework for the evaluation and selection of CCLDC within an uncertain context.

Details DOI

NeurIPS Conference 2022 Conference Paper

Periodic Graph Transformers for Crystal Material Property Prediction

  • Keqiang Yan
  • Yi Liu
  • Yuchao Lin
  • Shuiwang Ji

We consider representation learning on periodic graphs encoding crystal materials. Different from regular graphs, periodic graphs consist of a minimum unit cell repeating itself on a regular lattice in 3D space. How to effectively encode these periodic structures poses unique challenges not present in regular graph representation learning. In addition to being E(3) invariant, periodic graph representations need to be periodic invariant. That is, the learned representations should be invariant to shifts of cell boundaries as they are artificially imposed. Furthermore, the periodic repeating patterns need to be captured explicitly as lattices of different sizes and orientations may correspond to different materials. In this work, we propose a transformer architecture, known as Matformer, for periodic graph representation learning. Our Matformer is designed to be invariant to periodicity and can capture repeating patterns explicitly. In particular, Matformer encodes periodic patterns by efficient use of geometric distances between the same atoms in neighboring cells. Experimental results on multiple common benchmark datasets show that our Matformer outperforms baseline methods consistently. In addition, our results demonstrate the importance of periodic invariance and explicit repeating pattern encoding for crystal representation learning. Our code is publicly available at https: //github. com/YKQ98/Matformer.

PDF Details

ICRA Conference 2022 Conference Paper

Towards Efficient 3D Human Motion Prediction using Deformable Transformer-based Adversarial Network

  • Hua Yu 0006
  • Xuanzhe Fan
  • Yaqing Hou
  • Yi Liu
  • Cai Kang
  • Dongsheng Zhou
  • Qiang Zhang 0008

Human motion prediction is a crucial step for achieving human-robot interactions. While recent transformer-based methods have shown great potentials in 3D human motion prediction, they still suffer from mode collapse to non-plausible poses and quadratically computational complexity with respect to the increasing length of input sequences. In this paper, we propose a novel spatio-temporal deformable transformer-based adversarial network (STDTA) for 3D human motion prediction. First, we design a spatio-temporal deformable transformer module to capture the correlations between human joints while reducing the computational costs. Second, we introduce the adversarial training mechanism and design fidelity and continuity discriminators to maintain smoothness and stability for the long-term prediction. Finally, extensive experiments on Human 3. 6M and AMASS benchmarks demonstrate that the proposed STDTA achieves state-of-the-art performance.

Details

YNIMG Journal 2022 Journal Article

Transfer learning for cognitive reserve quantification

  • Xi Zhu
  • Yi Liu
  • Christian G. Habeck
  • Yaakov Stern
  • Seonjoo Lee
  • for-the-Alzheimer's-Disease-Neuroimaging-Initiative

Cognitive reserve (CR) has been introduced to explain individual differences in susceptibility to cognitive or functional impairment in the presence of age or pathology. We developed a deep learning model to quantify the CR as residual variance in memory performance using the Structural Magnetic Resonance Imaging (sMRI) data from a lifespan healthy cohort. The generalizability of the sMRI-based deep learning model was tested in two independent healthy and Alzheimer's cohorts using transfer learning framework. Structural MRIs were collected from three cohorts: 495 healthy adults (age: 20-80) from RANN, 620 healthy adults (age: 36-100) from lifespan Human Connectome Project Aging (HCPA), and 941 adults (age: 55-92) from Alzheimer's Disease Neuroimaging Initiative (ADNI). Region of interest (ROI)-specific cortical thickness and volume measures were extracted using the Desikan-Killiany Atlas. CR was quantified by residuals which subtract the predicted memory from the true memory. Cascade neural network (CNN) models were used to train RANN dataset for memory prediction. Transfer learning was applied to transfer the T1 imaging-based model from source domain (RANN) to the target domains (HCPA or ADNI). The CNN model trained on the RANN dataset exhibited strong linear correlation between true and predicted memory based on the T1 cortical thickness and volume predictors. In addition, the model generated from healthy lifespan data (RANN) was able to generalize to an independent healthy lifespan data (HCPA) and older demented participants (ADNI) across different scanner types. The estimated CR was correlated with CR proxies such education and IQ across all three datasets. The current findings suggest that the transfer learning approach is an effective way to generalize the residual-based CR estimation. It is applicable to various diseases and may flexibly incorporate different imaging modalities such as fMRI and PET, making it a promising tool for scientific and clinical purposes.

Details DOI

JBHI Journal 2021 Journal Article

A Home-Based Bilateral Rehabilitation System With sEMG-based Real-Time Variable Stiffness

  • Yi Liu
  • Shuxiang Guo
  • Ziyi Yang
  • Hideyuki Hirata
  • Takashi Tamiya

Bilateral rehabilitation allows patients with hemiparesis to exploit the cooperative capabilities of both arms to promote the recovery process. Although various approaches have been proposed to facilitate synchronized robot-assisted bilateral movements, few studies have focused on addressing the varying joint stiffness resulting from dynamic motions. This paper presents a novel bilateral rehabilitation system that implements a surface electromyography (sEMG)-based stiffness control to achieve real-time stiffness adjustment based on the user's dynamic motion. An sEMG-driven musculoskeletal model that incorporates the muscle activation and muscular contraction dynamics is developed to provide reference signals for the robot's real-time stiffness control. Preliminary experiments were conducted to evaluate the system performance in tracking accuracy and comfortability, which showed the proposed rehabilitation system with sEMG-based real-time stiffness variation achieved fast adaption to the patient's dynamic movement as well as improving the comfort in robot-assisted bilateral training.

Details DOI

NeurIPS Conference 2021 Conference Paper

Damped Anderson Mixing for Deep Reinforcement Learning: Acceleration, Convergence, and Stabilization

  • Ke Sun
  • Yafei Wang
  • Yi Liu
  • Yingnan Zhao
  • Bo Pan
  • Shangling Jui
  • Bei Jiang
  • Linglong Kong

Anderson mixing has been heuristically applied to reinforcement learning (RL) algorithms for accelerating convergence and improving the sampling efficiency of deep RL. Despite its heuristic improvement of convergence, a rigorous mathematical justification for the benefits of Anderson mixing in RL has not yet been put forward. In this paper, we provide deeper insights into a class of acceleration schemes built on Anderson mixing that improve the convergence of deep RL algorithms. Our main results establish a connection between Anderson mixing and quasi-Newton methods and prove that Anderson mixing increases the convergence radius of policy iteration schemes by an extra contraction factor. The key focus of the analysis roots in the fixed-point iteration nature of RL. We further propose a stabilization strategy by introducing a stable regularization term in Anderson mixing and a differentiable, non-expansive MellowMax operator that can allow both faster convergence and more stable behavior. Extensive experiments demonstrate that our proposed method enhances the convergence, stability, and performance of RL algorithms.

PDF Details

JMLR Journal 2021 Journal Article

DIG: A Turnkey Library for Diving into Graph Deep Learning Research

  • Meng Liu
  • Youzhi Luo
  • Limei Wang
  • Yaochen Xie
  • Hao Yuan
  • Shurui Gui
  • Haiyang Yu
  • Zhao Xu

Although there exist several libraries for deep learning on graphs, they are aiming at implementing basic operations for graph deep learning. In the research community, implementing and benchmarking various advanced tasks are still painful and time-consuming with existing libraries. To facilitate graph deep learning research, we introduce DIG: Dive into Graphs, a turnkey library that provides a unified testbed for higher level, research-oriented graph deep learning tasks. Currently, we consider graph generation, self-supervised learning on graphs, explainability of graph neural networks, and deep learning on 3D graphs. For each direction, we provide unified implementations of data interfaces, common algorithms, and evaluation metrics. Altogether, DIG is an extensible, open-source, and turnkey library for researchers to develop new methods and effortlessly compare with common baselines using widely used datasets and evaluation metrics. Source code is available at https://github.com/divelab/DIG. [abs] [ pdf ][ bib ] [ code ] &copy JMLR 2021. ( edit, beta )

PDF Details

IJCAI Conference 2020 Conference Paper

Internal and Contextual Attention Network for Cold-start Multi-channel Matching in Recommendation

  • Ruobing Xie
  • Zhijie Qiu
  • Jun Rao
  • Yi Liu
  • Bo Zhang
  • Leyu Lin

Real-world integrated personalized recommendation systems usually deal with millions of heterogeneous items. It is extremely challenging to conduct full corpus retrieval with complicated models due to the tremendous computation costs. Hence, most large-scale recommendation systems consist of two modules: a multi-channel matching module to efficiently retrieve a small subset of candidates, and a ranking module for precise personalized recommendation. However, multi-channel matching usually suffers from cold-start problems when adding new channels or new data sources. To solve this issue, we propose a novel Internal and contextual attention network (ICAN), which highlights channel-specific contextual information and feature field interactions between multiple channels. In experiments, we conduct both offline and online evaluations with case studies on a real-world integrated recommendation system. The significant improvements confirm the effectiveness and robustness of ICAN, especially for cold-start channels. Currently, ICAN has been deployed on WeChat Top Stories used by millions of users. The source code can be obtained from https: //github. com/zhijieqiu/ICAN.

PDF Details DOI

JBHI Journal 2020 Journal Article

Single Volume Image Generator and Deep Learning-Based ASD Classification

  • Md Rishad Ahmed
  • Yuan Zhang
  • Yi Liu
  • Hongen Liao

Autism spectrum disorder (ASD) is an intricate neuropsychiatric brain disorder characterized by social deficits and repetitive behaviors. Deep learning approaches have been applied in clinical or behavioral identification of ASD; most erstwhile models are inadequate in their capacity to exploit the data richness. On the other hand, classification techniques often solely rely on region-based summary and/or functional connectivity analysis of functional magnetic resonance imaging (fMRI). Besides, biomedical data modeling to analyze big data related to ASD is still perplexing due to its complexity and heterogeneity. Single volume image consideration has not been previously investigated in classification purposes. By deeming these challenges, in this work, firstly, we design an image generator to generate single volume brain images from the whole-brain image by considering the voxel time point of each subject separately. Then, to classify ASD and typical control participants, we evaluate four deep learning approaches with their corresponding ensemble classifiers comprising one amended Convolutional Neural Network (CNN). Finally, to check out the data variability, we apply the proposed CNN classifier with leave-one-site-out 5-fold cross-validation across the sites and validate our findings by comparing with literature reports. We showcase our approach on large-scale multi-site brain imaging dataset (ABIDE) by considering four preprocessing pipelines, which outperforms the state-of-the-art methods. Hence, it is robust and consistent.

Details DOI

AAAI Conference 2019 Conference Paper

A Human-Like Semantic Cognition Network for Aspect-Level Sentiment Classification

  • Zeyang Lei
  • Yujiu Yang
  • Min Yang
  • Wei Zhao
  • Jun Guo
  • Yi Liu

In this paper, we propose a novel Human-like Semantic Cognition Network (HSCN) for aspect-level sentiment classification, motivated by the principles of human beings’ reading cognitive process (pre-reading, active reading, post-reading). We first design a word-level interactive perception module to capture the correlation between context words and the given target words, which can be regarded as pre-reading. Second, to mimic the process of active reading, we propose a targetaware semantic distillation module to produce the targetspecific context representation for aspect-level sentiment prediction. Third, we further devise a semantic deviation metric module to measure the semantic deviation between the targetspecific context representation and the given target, which evaluates the degree we understand the target-specific context semantics. The measured semantic deviation is then used to fine-tune the above active reading process in a feedback regulation way. To verify the effectiveness of our approach, we conduct extensive experiments on three widely used datasets. The experiments demonstrate that HSCN achieves impressive results compared to other strong competitors.

PDF Details

AAAI Conference 2018 Conference Paper

Aida: Intelligent Image Analysis to Automatically Detect Poems in Digital Archives of Historic Newspapers

  • Leen-Kiat Soh
  • Elizabeth Lorang
  • Yi Liu

We describe an intelligent image analysis approach to automatically detect poems in digitally archived historic newspapers. Our application, Image Analysis for Archival Discovery, or Aida, integrates computer vision to capture visual cues based on visual structures of poetic works— instead of the meaning or content—and machine learning to train an artificial neural network to determine whether an image has poetic text. We have tested our application on almost 17, 000 image snippets and obtained promising accuracies, precision, and recall. The application is currently being deployed at two institutions for digital library and literary research.

PDF Details

EAAI Journal 2016 Journal Article

Parameter identification of a nonlinear model of hydraulic turbine governing system with an elastic water hammer based on a modified gravitational search algorithm

  • Chaoshun Li
  • Li Chang
  • Zhengjun Huang
  • Yi Liu
  • Nan Zhang

The hydraulic turbine governing system (HTGS) is a crucial control system of hydroelectric generating units (HGUs). Parameter identification of HTGS is an important issue for the modeling and control of HGUs. The parameter identification problem of HTGS is more difficult if the elastic water hammer model is considered in the system, and existing algorithms are not effective to solve it. To solve this new problem, a modified gravitational search algorithm (MGSA) has been proposed in which modifications have been made to improve the performance of the GSA from two aspects. First, the constant attenuation factor is replaced by a hyperbolic function to generate a better gravitational constant to balance the global exploration and local exploitation during different searching stages. Second, agent mutation is introduced to increase the diversity of agents and to strengthen the ability to jump out of the local minima of the GSA. The performance of the MGSA has been verified by 13 typical benchmark problems, and the experimental results and statistical analysis demonstrate that the proposed MGSA significantly outperforms the standard GSA and some other popular optimization algorithms. The MGSA is then employed in the parameter identification of a nonlinear model of HTGS with an elastic water hammer, and the experimental results indicate that MGSA locates more precise parameter values than the compared methods.

Details DOI

YNIMG Journal 2015 Journal Article

Mortality salience enhances racial in-group bias in empathic neural responses to others' suffering

  • Xiaoyang Li
  • Yi Liu
  • Siyang Luo
  • Bing Wu
  • Xinhuai Wu
  • Shihui Han

Behavioral research suggests that mortality salience (MS) leads to increased in-group identification and in-group favoritism in prosocial behavior. What remains unknown is whether and how MS influences brain activity that mediates emotional resonance with in-group and out-group members and is associated with in-group favoritism in helping behavior. The current work investigated MS effects on empathic neural responses to racial in-group and out-group members' suffering. Experiments 1 and 2 respectively recorded event related potentials (ERPs) and blood oxygen level dependent signals to pain/neutral expressions of Asian and Caucasian faces from Chinese adults who had been primed with MS or negative affect (NA). Experiment 1 found that an early frontal/central activity (P2) was more strongly modulated by pain vs. neutral expressions of Asian than Caucasian faces, but this effect was not affected by MS vs. NA priming. However, MS relative to NA priming enhanced racial in-group bias in long-latency neural response to pain expressions over the central/parietal regions (P3). Experiment 2 found that MS vs. NA priming increased racial in-group bias in empathic neural responses to pain expression in the anterior and mid-cingulate cortex. Our findings indicate that reminding mortality enhances brain activity that differentiates between racial in-group and out-group members' emotional states and suggest a neural basis of in-group favoritism under mortality threat.

Details DOI

AAAI Conference 2008 Conference Paper

Dimension Amnesic Pyramid Match Kernel

  • Yi Liu

With the success of local features in object recognition, feature-set representations are widely used in computer vision and related domains. Pyramid match kernel (PMK) is an efficient approach to quantifying the similarity between two unordered feature-sets, which allows well established kernel machines to learn with such representations. However, the approximation of PMK to the optimal feature matches deteriorates linearly with the dimension of local features, which prohibits the direct use of high dimensional features. In this paper, we propose a general, data-independent kernel to quantify the feature-set similarities, which gives an upper bound of approximation error independent of the dimension of local features. The key idea is to employ the technique of normal random projection to construct a number of low dimensional subspaces, and perform the original PMK algorithm therein. By leveraging on the invariance property of p-stable distributions, our approach achieves the desirable dimension-free property. Extensive experiments on the ETH-80 image database solidly demonstrate the advantage of our approach to high dimensional features.

PDF Details

AAAI Conference 2006 Conference Paper

Semi-supervised Multi-label Learning by Constrained Non-negative Matrix Factorization

  • Yi Liu

We present a novel framework for multi-label learning that explicitly addresses the challenge arising from the large number of classes and a small size of training data. The key assumption behind this work is that two examples tend to have large overlap in their assigned class memberships if they share high similarity in their input patterns. We capitalize this assumption by first computing two sets of similarities, one based on the input patterns of examples, and the other based on the class memberships of the examples. We then search for the optimal assignment of class memberships to the unlabeled data that minimizes the difference between these two sets of similarities. The optimization problem is formulated as a constrained Non-negative Matrix Factorization (NMF) problem, and an algorithm is presented to efficiently find the solution. Compared to the existing approaches for multi-label learning, the proposed approach is advantageous in that it is able to explore both the unlabeled data and the correlation among different classes simultaneously. Experiments with text categorization show that our approach performs significantly better than several state-of-the-art classification techniques when the number of classes is large and the size of training data is small.

PDF Details

AAAI Conference 2005 Conference Paper

Query Translation Disambiguation as Graph Partitioning

  • Yi Liu

Resolving ambiguity in the process of query translation is crucial to cross-language information retrieval when only a bilingual dictionary is available. In this paper we propose a novel approach for query translation disambiguation, named “spectral query translation model”. The proposed approach views the problem of query translation disambiguation as a graph partitioning problem. For a given query, a weighted graph is first created for all possible translations of query words based on the co-occurrence statistics of the translation words. The best translation of the query is then determined by the most strongly connected component within the graph. The proposed approach distinguishes from previous approaches in that the translations of all query words are estimated simultaneously. Furthermore, translation probabilities are introduced in the proposed approach to capture the uncertainty in translating queries. Empirical studies with TREC datasets have shown that the spectral query translation model achieves a relative 20% - 50% improvement in cross-language information retrieval, compared to other approaches that also exploit word co-occurrence statistics for query translation disambiguation.

PDF Details

ICRA Conference 1988 Conference Paper

Stereo vision correspondence using a multichannel graph matching technique

  • Nasser M. Nasrabadi
  • Yi Liu
  • Jen-hi Chiang

A multichannel feature-based stereo vision technique is described wherein curve segments are used as the feature primitives in the matching process. Curve segments are extracted by tracking the zero-crossings of the left and right images. The generalized Hough transform of each curve and the length of the segment in the left image are used as a local feature vector to represent the distinctive characteristics of the segment. The feature vector of each segment is used as a constraint to find an instance of the same segment in the right image. The epipolar constraint on the centroids of the curve segment is used to limit the searching space in the right image. A relational graph is formed from the left image by treating the centroids as the nodes of the graph. The local features of the segments are used to represent the local properties of the nodes, and the relationship between the nodes represents the structural properties of the object in the scene. A similar graph is formed from the right image curve segments. A graph isomorphism is then formed between the two graphs. >

Details
v2026.09.13