Arrow Research search

Author name cluster

Yu Zhou

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.

60 papers
2 author rows

Possible papers

60

EAAI Journal 2026 Journal Article

A Local-Global Fusion Vision Mamba UNet Framework for medical image segmentation

  • Yanbo Li
  • Zihan Mao
  • Feiwei Qin
  • Yong Peng
  • Guodao Zhang
  • Xugang Xi
  • Xiaoqin Ma
  • Huanhuan Yu

As a State Space Model (SSM) that achieves long-range dependency modeling with linear computational complexity, Mamba demonstrates significant efficiency advantages in medical image segmentation. However, while Mamba-based methods enable long-range modeling with linear complexity, their global dependency mechanisms often lead to local feature attenuation, particularly affecting the processing of complex anatomical structures. Existing multi-scale fusion methods also exhibit limited compatibility with State Space Models. To address these challenges, this paper proposes the Local-Global Fusion Vision Mamba UNet (LGFVM-UNet) framework. Its core innovation lies in the Dynamic Gating-enhanced Local-Global Fusion Visual State Space (LGF-VSS) block, which enables the synergistic modeling of global context and local details. Additionally, we designed a Multi-level Cross-scale Feature Fusion Block (MCFB) that enhances multi-scale feature representation through bidirectional resampling and spatial-channel dual attention mechanisms. Additionally, we propose a Gradient Statistics-based Adaptive Hierarchical Loss that dynamically adjusts multi-level supervision weights to optimize the learning process. The proposed method is experimentally validated on five public medical image segmentation datasets spanning diverse imaging modalities and anatomical structures. Results demonstrate that our approach outperforms state-of-the-art methods, excelling in long-range dependency modeling, local detail capture, and multi-scale feature fusion. The source code of our work is available at https: //github. com/NicoleDyson/LGFVM-UNet.

Details DOI

EAAI Journal 2026 Journal Article

An efficient physics-informed neural network model for predicting methane and carbon dioxide adsorption in shale: Simultaneous enhancement of recovery and carbon sequestration

  • Yu Zhou
  • Jinyang Li
  • Wenchuan Liu
  • Xinlong Lu
  • Zizuo Liu
  • Xiaoping Li
  • Dengwei Jing

Machine learning (ML) models offer rapid and low-cost prediction of methane (CH4) and carbon dioxide (CO2) adsorption in shale, which is crucial for enhancing recovery and achieving CO2 geological sequestration. However, with adsorption mechanisms model not yet fully established, existing purely data-driven ML lacks reliable physical constraints and exhibits weak interpretability, limited accuracy, and poor generalization. To address this gap, a novel fractal supercritical Dubinin-Radushkevich-Langmuir (FSDR-L) model was derived to describe the adsorption behaviors of CH4 and CO2 in shale and to directly quantify the critical pore size for gas adsorption mechanism transition. The results indicate that increasing temperature shifts CH4/CO2 molecules in shale toward monolayer adsorption, while reducing the contribution of pore-filling. Subsequently, a physics-informed neural network (PINN) model guided by the insights of the FSDR-L model was developed for the first time to predict CH4/CO2 adsorption amounts in shale. The findings reveal that the PINN model achieved reductions of 38. 93 % in mean absolute percentage error, 39. 47 % in mean absolute error, and 57. 46 % in root mean square error compared to the best-performing conventional ML model, demonstrating superior predictive performance and generalization capability in capturing complex shale-gas adsorption behaviors. Finally, to enhance the interpretability of the PINN model, a variance-based sensitivity analysis was conducted, revealing that total organic carbon, pressure, temperature, and pore volume are the key factors governing CH4/CO2 adsorption capacity in shale.

Details DOI

AAAI Conference 2026 Conference Paper

Non-Monotonicity in Fair Division of Graphs

  • Hadi Hosseini
  • Shraddha Pathak
  • Yu Zhou

We consider the problem of fairly allocating the vertices of a graph among n agents, where the value of a bundle is determined by its cut value---the number of edges with exactly one endpoint in the bundle. This model naturally captures applications such as team formation and network partitioning, where valuations are inherently non-monotonic: the marginal values may be positive, negative, or zero depending on the composition of the bundle. We focus on the fairness notion of envy-freeness up to one item (EF1) and explore its compatibility with several efficiency concepts such as Transfer Stability (TS) that prohibits any transfer of item that benefit one agent at another's expense. For general graphs, our results uncover a non-monotonic relationship between the number of agents n and the existence of allocations satisfying EF1 and transfer stability (TS): such allocations always exist for n=2, may fail to exist for n=3, but exist again for all n>= 4. We further show that existence can be guaranteed for any n by slightly weakening the efficiency requirement or by restricting the graph to forests. All of our positive results are achieved via efficient algorithms.

PDF Details DOI

AAAI Conference 2026 Conference Paper

ST-SAM: Multimodal Scene Text Segmentation with Dense Visual and Sparse Textual Prompts via SAM

  • Jin Wei
  • Yaqiang Wu
  • Jiayi Yan
  • Zeng Li
  • Zhen Xu
  • Yu Zhou
  • Lingling Zhang
  • Qianying Wang

Scene text segmentation is a critical preprocessing step in various text-based applications. Specialist text segmentation methods, often relying on a detect-then-segment paradigm, tend to exhibit reduced robustness and can lead to cascading errors. The introduction of the Segment Anything Model (SAM) has revolutionized general segmentation by leveraging vision foundation models. However, SAM still falls short when applied to domain-specific tasks such as scene text segmentation. To bridge this gap between SAM and specialized scene text segmentation approaches, we propose ST-SAM (Scene Text SAM), a parameter-efficient fine-tuning framework tailored to adapt SAM for high-quality scene text segmentation without relying on explicit text detection. ST-SAM incorporates a multimodal prompting mechanism: a lightweight visual encoder generates multi-scale spatial features to provide precise visual context; and textual prompts generated by a large language model offer high-level semantic guidance. We demonstrate the advantages of the proposed ST-SAM as follows: (1) ST-SAM achieves new state-of-the-art performance on multiple scene text segmentation benchmarks, including 85.30% fgIoU on Total-Text and 91.03% fgIoU on TextSeg, outperforming both specialist and generalist models. (2) ST-SAM enables effective domain adaptation by flexibly adapting the general SAM architecture to the domain of scene text. (3) By discarding the detect-then-segment pipeline, ST-SAM simplifies the inference process while still achieving robust performance on complex text cases.

PDF Details DOI

AAAI Conference 2026 Conference Paper

SUGAR: Learning Skeleton Representation with Visual-Motion Knowledge for Action Recognition

  • Qilang Ye
  • Yu Zhou
  • Lian He
  • Jie Zhang
  • Xuanming Guo
  • Jiayu Zhang
  • Mingkui Tan
  • Weicheng Xie

Large Language Models (LLMs) hold rich implicit knowledge and powerful transferability. In this paper, we explore the combination of LLMs with the human skeleton to perform action classification and description. However, when treating LLM as a recognizer, two questions arise: 1) How can LLMs understand the skeleton? 2) How can LLMs distinguish among actions? To address these problems, we introduce a novel paradigm named learning Skeleton representation with visual-motion knowledge for Action Recognition (SUGAR). In our pipeline, we first utilize off-the-shelf large-scale video models as a knowledge base to generate visual, motion information related to actions. Then, we propose to supervise skeleton learning through this prior knowledge to yield discrete representations. Finally, we use the LLM with untouched pre-training weights to understand these representations and generate the desired action targets and descriptions. Notably, we present a Temporal Query Projection (TQP) module to continuously model the skeleton signals with long sequences. Experiments on several skeleton-based action classification benchmarks demonstrate the efficacy of our SUGAR. Moreover, experiments on zero-shot scenarios show that SUGAR is more versatile than linear-based methods.

PDF Details DOI

AAAI Conference 2026 Conference Paper

Task-Aware 3D Affordance Segmentation via 2D Guidance and Geometric Refinement

  • Lian He
  • Meng Liu
  • Qilang Ye
  • Yu Zhou
  • Xiang Deng
  • Gangyi Ding

Understanding 3D scene-level affordances from natural language instructions is essential for enabling embodied agents to interact meaningfully in complex environments. However, this task remains challenging due to the need for semantic reasoning and spatial grounding. Existing methods mainly focus on object-level affordances or merely lift 2D predictions to 3D, neglecting rich geometric structure information in point clouds and incurring high computational costs. To address these limitations, we introduce Task-Aware 3D Scene-level Affordance segmentation (TASA), a novel geometry-optimized framework that jointly leverages 2D semantic cues and 3D geometric reasoning in a coarse-to-fine manner. To improve the affordance detection efficiency, TASA features a task-aware 2D affordance detection module to identify manipulable points from language and visual inputs, guiding the selection of task-relevant views. To fully exploit 3D geometric information, a 3D affordance refinement module is proposed to integrate 2D semantic priors with local 3D geometry, resulting in accurate and spatially coherent 3D affordance masks. Experiments on SceneFun3D demonstrate that TASA significantly outperforms the baselines in both accuracy and efficiency in scene-level affordance segmentation.

PDF Details DOI

AAAI Conference 2026 Conference Paper

When Eyes and Ears Disagree: Can MLLMs Discern Audio-Visual Confusion?

  • Qilang Ye
  • Wei Zeng
  • Meng Liu
  • Jie Zhang
  • Yupeng Hu
  • Zitong Yu
  • Yu Zhou

Can Multimodal Large Language Models (MLLMs) discern confused objects that are visually present but audio-absent? To study this, we introduce a new benchmark, AV-ConfuseBench, which simulates an “Audio-Visual Confusion” scene by modifying the corresponding sound of an object in the video, e.g., mute the sounding object and ask MLLMs “Is there a/an {muted-object} sound”. Experimental results reveal that MLLMs, such as Qwen2.5-Omni and Gemini 2.5, struggle to discriminate non-existent audio due to visually dominated reasoning. Motivated by this observation, we introduce RL-CoMM, a Reinforcement Learning-based Collaborative Multi-MLLM that is built upon the Qwen2.5-Omni foundation. RL-CoMM includes two stages: 1) To alleviate visually dominated ambiguities, we introduce an external model, a Large Audio Language Model (LALM), as the reference model to generate audio-only reasoning. Then, we design a Step-wise Reasoning Reward function that enables MLLMs to self-improve audio-visual reasoning with the audio-only reference. 2) To ensure an accurate answer prediction, we introduce Answer-centered Confidence Optimization to reduce the uncertainty of potential heterogeneous reasoning differences. Extensive experiments on audio-visual question answering and audio-visual hallucination show that RL-CoMM improves accuracy by 10~30% over the baseline model with limited training data.

PDF Details DOI

EAAI Journal 2025 Journal Article

A novel multimodal deep learning framework for predicting residual strength of corroded rectangular hollow-section columns

  • Yu-Jia Zhang
  • Lei Zhang
  • Yu Zhou
  • Tian-Xiang Li
  • Reece Lincoln
  • Jing-Zhong Tong
  • Jia-Jia Shen

Corrosion, recognized as a thermodynamically spontaneous process, is one of the key issues affecting the health of rectangular hollow steel section columns under working conditions, and has attracted much attention in recent years. Traditional approaches, such as multi-layer perceptron, often rely solely on the degree of volume loss to predict residual strength, overlooking the spatial complexity of actual corrosion patterns. To address these limitations, this study presents a novel multimodal deep learning network for accurately predicting the residual strength of corroded hollow steel section columns with random, nonuniform corrosion distributions. Our approach integrates (i) image‐based corrosion distributions on four steel walls, and (ii) tabular geometric parameters, through five distinct data-fusion methods proposed in this work, three employing Late Fusion (via a novel multi‐head attention module) and two using Early Fusion (via pixel−level merging). The image information extraction core is built upon a lightweight convolutional neural network and a channel−spatial attention block, while the tabular extraction module leverages a revised multi-layer perceptron architecture. After Bayesian hyperparameter optimization, the best‐performing model achieves a coefficient of determination of 0. 971 on the test set, surpassing conventional machine learning and other multimodal fusion techniques by 0. 01–0. 161. Further analysis shows that the reverse visualization technique highlights corrosion−critical regions that closely coincide with the experimentally validated failure zones. Consequently, the proposed framework not only predicts residual strength with high accuracy but also localizes vulnerable areas for targeted reinforcement. This methodology holds promise for large‐scale corrosion monitoring and structural health assessment of steel infrastructure.

Details DOI

EAAI Journal 2025 Journal Article

A novel semi-local centrality to identify influential nodes in complex networks by integrating multidimensional factors

  • Kun Zhang
  • Zaiyi Pu
  • Chuan Jin
  • Yu Zhou
  • Zhenyu Wang

This study addresses the critical problem of identifying influential nodes in complex networks, a task that plays a pivotal role in understanding network dynamics, optimizing information spread, and controlling epidemic outbreaks. Although semi-local centrality metrics are a valid approach for identifying influential nodes, they face challenges such as inefficiency when dealing with large-scale networks and neglecting semantic relationships, often relying on unidimensional criteria that limit their effectiveness. To tackle this challenge, this study presents a novel Semi-Local Centrality metric designed to identify influential nodes in complex networks by incorporating Multidimensional Factors (SLCMF). SLCMF combines structural, social, and semantic factors to find seed nodes in complex networks. To improve scalability, SLCMF utilizes distributed local subgraphs and redefines semi-local centrality by employing the average shortest path theory. Additionally, SLCMF incorporates a semantic graph embedding model by an augmented graph to capture distant and latent relationships among nodes. Extensive experiments on real-world networks demonstrate the effectiveness and efficiency of the proposed centrality metric, showcasing its superior performance in ranking influential nodes. Specifically, SLCMF outperforms the best traditional and advanced centrality metrics, improving Kendall's correlation coefficient by 8. 94% and 1. 61%, respectively. Additionally, the proposed metric demonstrates enhanced efficiency, reducing runtime by 4. 7% and 0. 21% compared to the top-performing traditional and advanced metrics, respectively.

Details DOI

NeurIPS Conference 2025 Conference Paper

An Adaptive Quantum Circuit of Dempster's Rule of Combination for Uncertain Pattern Classification

  • Fuyuan Xiao
  • Yu Zhou
  • Witold Pedrycz

In pattern classification, efficient uncertainty reasoning plays a critical role, particularly in real-time applications involving noisy data, ambiguous class boundaries, or overlapping categories. Leveraging the advanced computational power of quantum computing, an Adaptive Quantum Circuit for Dempster’s Rule of Combination (AQC-DRC) is proposed to address efficient classification under uncertain environments. The AQC-DRC is developed within the framework of quantum evidence theory (QET) and facilitates decision-making based on quantum basic probability and plausibility levels, which is a generalized Bayesian inference method. The AQC-DRC provides a deterministic computation of DRC, ensuring that quantum fusion outcomes in uncertain pattern classification are exactly aligned with those of the classical method, while simultaneously achieving exponential reductions in the computational complexity of evidence combination and significantly improving fusion efficiency. It is founded that the quantum basic probability amplitude function in QET, as a generalized quantum probability amplitude, can be naturally utilized to express the quantum amplitude encoding. In addition, the quantum basic probability in QET, as a generalized quantum probability, naturally forms a quantum basic probability distribution and can be used to represent quantum measurement outcomes for quantum basic probability level decision-making. Furthermore, the quantum plausibility function in QET also can be naturally used to express the quantum measurement outcomes for quantum plausibility level decision-making. These findings enrich the physical understanding of quantum amplitude encoding and quantum measurement outcomes, offering broad application prospects for representing and processing uncertain knowledge in pattern classification.

PDF Details

AAAI Conference 2025 Conference Paper

Arbitrary Reading Order Scene Text Spotter with Local Semantics Guidance

  • Jiahao Lyu
  • Wei Wang
  • Dongbao Yang
  • Jinwen Zhong
  • Yu Zhou

Scene text spotting has attracted the enthusiasm of relative researchers in recent years. Most existing scene text spotters follow the detection-then-recognition paradigm, where the vanilla detection module hardly determines the reading order and leads to failure recognition. After rethinking the auto-regressive scene text recognition method, we find that a well-trained recognizer can implicitly perceive the local semantics of all characters in a complete word or a sentence without a character-level detection module. Local semantic knowledge not only includes text content but also spatial information in the right reading order. Motivated by the above analysis, we propose the Local Semantics Guided scene text Spotter (LSGSpotter), which auto-regressively decodes the position and content of characters guided by the local semantics. Specifically, two effective modules are proposed in LSGSpotter. On the one hand, we design a Start Point Localization Module (SPLM) for locating text start points to determine the right reading order. On the other hand, a Multi-scale Adaptive Attention Module (MAAM) is proposed to adaptively aggregate text features in a local area. In conclusion, LSGSpotter achieves the arbitrary reading order spotting task without the limitation of sophisticated detection, while alleviating the cost of computational resources with the grid sampling strategy. Extensive experiment results show LSGSpotter achieves state-of-the-art performance on the InverseText benchmark. Moreover, our spotter demonstrates superior performance on English benchmarks for arbitrary-shaped text, achieving improvements of 0.7% and 2.5% on Total-Text and SCUT-CTW1500, respectively. These results validate our text spotter is effective for scene texts in arbitrary reading order and shape.

PDF Details DOI

IROS Conference 2025 Conference Paper

Bridging the Reality Gap: Communication-Aware Task Allocation with Multi-Objective Asynchronous Policy Learning

  • Zehao Xiong
  • Yu Zhou
  • Yexun Xi
  • Yizhe Cao
  • Chang Wang
  • Jie Li

Distributed task allocation in the UAV swarm is sensitive to excessive communication overhead and frequent transmissions. Combining reinforcement learning and task allocation demonstrates great potential in enhancing algorithm performance and optimizing communication. However, existing studies rely on ideal communication assumptions and the nonphysical environment, making training and validation impractical in applying networked swarms. This paper proposes the Communication-Aware Task Allocation, which aims to train a gating mechanism policy to coordinate the transmission timing, improving robustness and timelessness of the task allocation. First, the policy learning problem is formalized as a POMDP, for which the channel access and other features are designed for observations, actions are inter-agent adaptive gating mechanisms, and the shared reward reflects global task conflicts. Second, to address the asynchronous learning under the CTDE, an asynchronous experience collection and splicing method is proposed to align trajectories. Then, the MOCPPO is proposed, which combines a primal-dual operator with proximal policy optimization, updating the optimal Lagrange multiplier and strategy parameters to simultaneously minimize task conflicts and communication overhead. Finally, sim-to-real experiments are conducted in the HIL environment, and results illustrate the best trade-off optimization of the proposed method over all state-of-the-art approaches.

Details

AAAI Conference 2025 Conference Paper

DCA: Dividing and Conquering Amnesia in Incremental Object Detection

  • Aoting Zhang
  • Dongbao Yang
  • Chang Liu
  • Xiaopeng Hong
  • Miao Shang
  • Yu Zhou

Incremental object detection (IOD) aims to cultivate an object detector that can continuously localize and recognize novel classes while preserving its performance on previous classes. Existing methods achieve certain success by improving knowledge distillation and exemplar replay for transformer-based detection frameworks, but the intrinsic forgetting mechanisms remain underexplored. In this paper, we dive into the cause of forgetting and discover forgetting imbalance between localization and recognition in transformer-based IOD, which means that localization is less-forgetting and can generalize to future classes, whereas catastrophic forgetting occurs primarily on recognition. Based on these insights, we propose a Divide-and-Conquer Amnesia (DCA) strategy, which redesigns the transformer-based IOD into a localization-then-recognition process. DCA can well maintain and transfer the localization ability, leaving decoupled fragile recognition to be specially conquered. To reduce feature drift in recognition, we leverage semantic knowledge encoded in pre-trained language models to anchor class representations within a unified feature space across incremental tasks. This involves designing a duplex classifier fusion and embedding class semantic features into the recognition decoding process in the form of queries. Extensive experiments validate that our approach achieves state-of-the-art performance, especially for long-term incremental scenarios. For example, under the four-step setting on MS-COCO, our DCA strategy significantly improves the final AP by 6.9%.

PDF Details DOI

NeurIPS Conference 2025 Conference Paper

Distilling LLM Prior to Flow Model for Generalizable Agent’s Imagination in Object Goal Navigation

  • Badi Li
  • Ren-Jie Lu
  • Yu Zhou
  • Jingke Meng
  • Wei-Shi Zheng

The Object Goal Navigation (ObjectNav) task challenges agents to locate a specified object in an unseen environment by imagining unobserved regions of the scene. Prior approaches rely on deterministic and discriminative models to complete semantic maps, overlooking the inherent uncertainty in indoor layouts and limiting their ability to generalize to unseen environments. In this work, we propose GOAL, a generative flow-based framework that models the semantic distribution of indoor environments by bridging observed regions with LLM-enriched full-scene semantic maps. During training, spatial priors inferred from large language models (LLMs) are encoded as two-dimensional Gaussian fields and injected into target maps, distilling rich contextual knowledge into the flow model and enabling more generalizable completions. Extensive experiments demonstrate that GOAL achieves state-of-the-art performance on MP3D and Gibson, and shows strong generalization in transfer settings to HM3D.

PDF Details

NeurIPS Conference 2025 Conference Paper

HM3: Hierarchical Multi-Objective Model Merging for Pretrained Models

  • Yu Zhou
  • Xingyu Wu
  • Jibin Wu
  • Liang Feng
  • KC Tan

Model merging is a technique that combines multiple large pretrained models into a single model, enhancing performance and broadening task adaptability without original data or additional training. However, most existing model merging methods focus primarily on exploring the parameter space, merging models with identical architectures. Despite its potential, merging in the architecture space remains in its early stages due to the vast search space and challenges related to layer compatibility. This paper designs a hierarchical model merging framework named HM3, formulating a bilevel multi-objective model merging problem across both parameter and architecture spaces. At the parameter level, HM3 integrates existing merging methods to quickly identify optimal parameters. Based on these, an actor-critic strategy with efficient policy discretization is employed at the architecture level to explore inference paths with Markov property in the layer-granularity search space for reconstructing these optimal models. By training reusable policy and value networks, HM3 learns Pareto optimal models to provide customized solutions for various tasks. Experimental results on language and vision tasks demonstrate that HM3 outperforms methods focusing solely on the parameter or architecture space.

PDF Details

ICRA Conference 2025 Conference Paper

Inverse Kinematics on Guiding Vector Fields for Robot Path Following

  • Yu Zhou
  • Jesus Bautista
  • Weijia Yao
  • Héctor García de Marina

Inverse kinematics is a fundamental technique for motion and positioning control in robotics, typically applied to end-effectors. In this paper, we extend the concept of inverse kinematics to guiding vector fields for path following in autonomous mobile robots. The desired path is defined by its implicit equation, i. e. , by a collection of points belonging to one or more zero-level sets. These level sets serve as a reference to construct an error signal that drives the guiding vector field toward the desired path, enabling the robot to converge and travel along the path by following such a vector field. We start with the formal exposition on how inverse kinematics can be applied to guiding vector fields for single-integrator robots in an $m$ -dimensional Euclidean space. Then, we leverage inverse kinematics to ensure that the level-set error signal behaves as a linear system, facilitating control over the robot's transient motion toward the desired path and allowing for the injection of feed-forward signals to induce precise motion behavior along the path. We then propose solutions to the theoretical and practical challenges of applying this technique to unicycles with constant speeds to follow 2D paths with precise transient control. We finish by validating the predicted theoretical results through real flights with fixed-wing drones.

Details

AAAI Conference 2025 Conference Paper

LDP: Generalizing to Multilingual Visual Information Extraction by Language Decoupled Pretraining

  • Huawen Shen
  • Gengluo Li
  • Jinwen Zhong
  • Yu Zhou

Visual Information Extraction (VIE) plays a crucial role in the comprehension of semi-structured documents, and several pre-trained models have been developed to enhance performance. However, most of these works are monolingual (usually English). Due to the extremely unbalanced quantity and quality of pre-training corpora between English and other languages, few works can extend to non-English scenarios. In this paper, we conduct systematic experiments to show that vision and layout modality hold invariance among images with different languages. If decoupling language bias from document images, a vision-layout-based model can achieve impressive cross-lingual generalization. Accordingly, we present a simple but effective multilingual training paradigm LDP (Language Decoupled Pre-training) for better utilization of monolingual pre-training data. Our proposed model LDM (Language Decoupled Model) is first pre-trained on the language-independent data, where the language knowledge is decoupled by a diffusion model, and then the LDM is fine-tuned on the downstream languages. Extensive experiments show that the LDM outperformed all SOTA multilingual pre-trained models, and also maintains competitiveness on downstream monolingual/English benchmarks.

PDF Details DOI

TCS Journal 2025 Journal Article

Locating two facilities on a square with a minimum distance requirement

  • Weian Li
  • Yu Zhou

Classic works on facility location problems have been focused on the basic model where facilities and agents are distributed on one line. In this work, we study a new model where one facility or two facilities with a minimum distance requirement are to be located on a square (e. g. , a plaza) to serve the agents who are distributed on a line (e. g. , a street) that crosses the square. The actual positions of the agents are their private information, and our goal is to design strategyproof mechanisms that decide the locations to build the facilities such that the agents are incentivized to report their true positions and the social welfare is (approximately) maximized. We study different settings, where the facilities can be favorable or obnoxious and the distance metrics can be Manhattan or Euclidean. Interestingly, for Manhattan distances, all but one of our mechanisms achieve the optimal social welfare. For Euclidean distances, however, the optimal algorithms are not strategyproof. Accordingly, for each setting with Euclidean distances, we design strategyproof mechanisms that guarantee constant approximations of the optimal social welfare.

Details DOI

NeurIPS Conference 2025 Conference Paper

MTL-KD: Multi-Task Learning Via Knowledge Distillation for Generalizable Neural Vehicle Routing Solver

  • yuepeng zheng
  • Fu Luo
  • Zhenkun Wang
  • Yaoxin Wu
  • Yu Zhou

Multi-Task Learning (MTL) in Neural Combinatorial Optimization (NCO) is a promising approach for training a unified model capable of solving multiple Vehicle Routing Problem (VRP) variants. However, existing Reinforcement Learning (RL)-based multi-task methods can only train light decoder models on small-scale problems, exhibiting limited generalization ability when solving large-scale problems. To overcome this limitation, this work introduces a novel multi-task learning method driven by knowledge distillation (MTL-KD), which enables efficient training of heavy decoder models with strong generalization ability. The proposed MTL-KD method transfers policy knowledge from multiple distinct RL-based single-task models to a single heavy decoder model, facilitating label-free training and effectively improving the model's generalization ability across diverse tasks. In addition, we introduce a flexible inference strategy termed Random Reordering Re-Construction (R3C), which is specifically adapted for diverse VRP tasks and further boosts the performance of the multi-task model. Experimental results on 6 seen and 10 unseen VRP variants with up to 1, 000 nodes indicate that our proposed method consistently achieves superior performance on both uniform and real-world benchmarks, demonstrating robust generalization abilities. The code is available at https: //github. com/CIAM-Group/MTLKD.

PDF Details

NeurIPS Conference 2025 Conference Paper

Orochi: Versatile Biomedical Image Processor

  • Gaole Dai
  • Chenghao Zhou
  • Yu Zhou
  • Rongyu Zhang
  • Yuan Zhang
  • Chengkai Hou
  • Tiejun Huang
  • Jianxu Chen

Deep learning has emerged as a pivotal tool for accelerating research in the life sciences, with the low-level processing of biomedical images (e. g. , registration, fusion, restoration, super-resolution) being one of its most critical applications. Platforms such as ImageJ (Fiji) and napari have enabled the development of customized plugins for various models. However, these plugins are typically based on models that are limited to specific tasks and datasets, making them less practical for biologists. To address this challenge, we introduce Orochi, the first application-oriented, efficient, and versatile image processor designed to overcome these limitations. Orochi is pre-trained on patches/volumes extracted from the raw data of over 100 publicly available studies using our Random Multi-scale Sampling strategy. We further propose Task-related Joint-embedding Pre-Training (TJP), which employs biomedical task-related degradation for self-supervision rather than relying on Masked Image Modelling (MIM), which performs poorly in downstream tasks such as registration. To ensure computational efficiency, we leverage Mamba's linear computational complexity and construct Multi-head Hierarchy Mamba. Additionally, we provide a three-tier fine-tuning framework (Full, Normal, and Light) and demonstrate that Orochi achieves comparable or superior performance to current state-of-the-art specialist models, even with lightweight parameter-efficient options. We hope that our study contributes to the development of an all-in-one workflow, thereby relieving biologists from the overwhelming task of selecting among numerous models. Our pre-trained weights and code will be released.

PDF Details

AAAI Conference 2025 Conference Paper

Specifying What You Know or Not for Multi-Label Class-Incremental Learning

  • Aoting Zhang
  • Dongbao Yang
  • Chang Liu
  • Xiaopeng Hong
  • Yu Zhou

Existing class incremental learning is mainly designed for single-label classification task, which is ill-equipped for multi-label scenarios due to the inherent contradiction of learning objectives for samples with incomplete labels. We argue that the main challenge to overcome this contradiction in multi-label class-incremental learning (MLCIL) lies in the model's inability to clearly distinguish between known and unknown knowledge. This ambiguity hinders the model's ability to retain historical knowledge, master current classes, and prepare for future learning simultaneously. In this paper, we target at specifying what is known or not to accommodate Historical, Current, and Prospective knowledge for MLCIL and propose a novel framework termed as HCP. Specifically, (i) we clarify the known classes by dynamic feature purification and recall enhancement with distribution prior, enhancing the precision and retention of known information. (ii) We design prospective knowledge mining to probe the unknown, preparing the model for future learning. Extensive experiments validate that our method effectively alleviates catastrophic forgetting in MLCIL, surpassing the previous state-of-the-art by 3.3% on average accuracy for MS-COCO B0-C10 setting without replay buffers.

PDF Details DOI

IJCAI Conference 2025 Conference Paper

The Devil is in Fine-tuning and Long-tailed Problems: A New Benchmark for Scene Text Detection

  • Tianjiao Cao
  • Jiahao Lyu
  • Weichao Zeng
  • Weimin Mu
  • Yu Zhou

Scene text detection has seen the emergence of high-performing methods that excel on academic benchmarks. However, these detectors often fail to replicate such success in real-world scenarios. We uncover two key factors contributing to this discrepancy through extensive experiments. First, a Fine-tuning Gap, where models leverage Dataset-Specific Optimization (DSO) paradigm for one domain at the cost of reduced effectiveness in others, leads to inflated performances on academic benchmarks. Second, the suboptimal performance in practical settings is primarily attributed to the longtailed distribution of texts, where detectors struggle with rare and complex categories as artistic or overlapped text. Given that the DSO paradigm might undermine the generalization ability of models, we advocate for a Joint-Dataset Learning (JDL) protocol to alleviate the Fine-tuning Gap. Additionally, an error analysis is conducted to identify three major categories and 13 subcategories of challenges in long-tailed scene text, upon which we propose a Long-Tailed Benchmark (LTB). LTB facilitates a comprehensive evaluation of ability to handle a diverse range of long-tailed challenges. We further introduce MAEDet, a self-supervised learningbased method, as a strong baseline for LTB. The code is available at https: //github. com/pd162/LTB.

PDF Details DOI

ICML Conference 2025 Conference Paper

The Four Color Theorem for Cell Instance Segmentation

  • Ye Zhang 0008
  • Yu Zhou
  • Yifeng Wang 0001
  • Jun Xiao
  • Ziyue Wang 0005
  • Yongbing Zhang 0002
  • Jianxu Chen 0001

Cell instance segmentation is critical to analyzing biomedical images, yet accurately distinguishing tightly touching cells remains a persistent challenge. Existing instance segmentation frameworks, including detection-based, contour-based, and distance mapping-based approaches, have made significant progress, but balancing model performance with computational efficiency remains an open problem. In this paper, we propose a novel cell instance segmentation method inspired by the four-color theorem. By conceptualizing cells as countries and tissues as oceans, we introduce a four-color encoding scheme that ensures adjacent instances receive distinct labels. This reformulation transforms instance segmentation into a constrained semantic segmentation problem with only four predicted classes, substantially simplifying the instance differentiation process. To solve the training instability caused by the non-uniqueness of four-color encoding, we design an asymptotic training strategy and encoding transformation method. Extensive experiments on various modes demonstrate our approach achieves state-of-the-art performance. The code is available at https: //github. com/zhangye-zoe/FCIS.

Details

AIIM Journal 2025 Journal Article

The interpretable deep learning framework and validation for seizure detection in pediatric electroencephalography: An improved accuracy and performance analysis

  • Yu Zhou
  • Yuxin Gao
  • Qiang Li
  • Ruiheng Wu
  • Aiping Yang
  • Ming-Lang Tseng

This study proposes an interpretable deep learning framework and compares the two novel models. A fully convolutional network with squeeze-and-excitation modules (SE-FCN) is designed to enhance spatial sensitivity and retain temporal resolution. In addition, a transformer-based model (TransNet) is developed to capture temporal and channel-wise dependencies via self-attention. These two models output channel saliency weights to the EEG electrode space and generate heatmaps for inferring potential epileptogenic zones. Deep learning primarily adopts convolutional neural networks (CNNs) or sequence generation networks (SGNs) and faces the limitations. For instance, CNN-based models often lack hierarchical modeling and fail to quantify channel-wise contributions, hindering spatial localization. SGN-based models struggle to capture complex spatiotemporal dependencies and typically lack adaptive attention tailored to electroencephalography (EEG) characters. Epileptic seizure detection is vital for effective clinical intervention and existing methods operated as black boxes, limiting clinical interpretability. This study evaluates the models on the CHB-MIT pediatric EEG dataset using a subject-independent cross-validation protocol. SE-FCN achieves an AUC of 0. 89 and accuracy of 86. 7 %, while TransNet achieves an AUC of 0. 92 and accuracy of 86. 4 %. Saliency maps from both models demonstrate high consistency and enable categorization of 22 patients into five groups based on inferred seizure origins.

Details DOI

IJCAI Conference 2025 Conference Paper

The Role of Video Generation in Enhancing Data-Limited Action Understanding

  • Wei Li
  • Dezhao Luo
  • Dongbao Yang
  • Zhenhang Li
  • Weiping Wang
  • Yu Zhou

Video action understanding tasks in real-world scenarios often suffer from data limitations. In this paper, we address the data-limited action understanding problem by bridging data scarcity. We propose a novel method that leverages a text-to-video diffusion transformer to generate annotated data for model training. This paradigm enables the generation of realistic annotated data on an infinite scale without human intervention. We proposed the Information Enhancement Strategy and the Uncertainty-Based Soft Target tailored to generate sample training. Through quantitative and qualitative analyzes, we discovered that real samples generally contain a richer level of information compared to generated samples. Based on this observation, the information enhancement strategy was designed to enhance the informational content of the generated samples from two perspectives: the environment and the character. Furthermore, we observed that a portion of low-quality generated samples might negatively affect model training. To address this, we devised an uncertainty-based label-smoothing strategy to increase the smoothing of these low-quality samples, thereby reducing their impact. We demonstrate the effectiveness of the proposed method on four datasets and five tasks, and achieve state-of-the-art performance for zero-shot action recognition.

PDF Details DOI

AAAI Conference 2025 Conference Paper

Track the Answer: Extending TextVQA from Image to Video with Spatio-Temporal Clues

  • Yan Zhang
  • Gangyan Zeng
  • Huawen Shen
  • Daiqing Wu
  • Yu Zhou
  • Can Ma

Video text-based visual question answering (TextVQA) is a practical task that aims to answer questions by jointly reasoning textual and visual information in a given video. Inspired by the development of TextVQA in image domain, existing Video TextVQA approaches leverage a language model (e.g. T5) to process text-rich multiple frames and generate answers auto-regressively. Nevertheless, the spatio-temporal relationships among visual entities (including scene text and objects) will be disrupted and models are susceptible to interference from unrelated information, resulting in irrational reasoning and inaccurate answering. To tackle these challenges, we propose the TEA (stands for "Track the Answer'') method that better extends the generative TextVQA framework from image to video. TEA recovers the spatio-temporal relationships in a complementary way and incorporates OCR-aware clues to enhance the quality of reasoning questions. Extensive experiments on several public Video TextVQA datasets validate the effectiveness and generalization of our framework. TEA outperforms existing TextVQA methods, video-language pretraining methods and video large language models by great margins. The code will be publicly released.

PDF Details DOI

NeurIPS Conference 2025 Conference Paper

When Semantics Mislead Vision: Mitigating Large Multimodal Models Hallucinations in Scene Text Spotting and Understanding

  • Yan Shu
  • Hangui Lin
  • Yexin Liu
  • Yan Zhang
  • Gangyan Zeng
  • Yan Li
  • Yu Zhou
  • Ser Nam Lim

Large Multimodal Models (LMMs) have achieved impressive progress in visual perception and reasoning. However, when confronted with visually ambiguous or non-semantic scene text, they often struggle to accurately spot and understand the content, frequently generating semantically plausible yet visually incorrect answers, which we refer to as semantic hallucination. In this work, we investigate the underlying causes of semantic hallucination and identify a key finding: Transformer layers in LLM with stronger attention focus on scene text regions are less prone to producing semantic hallucinations. Thus, we propose a training-free semantic hallucination mitigation framework comprising two key components: (1) ZoomText, a coarse-to-fine strategy that identifies potential text regions without external detectors; and (2) Grounded Layer Correction, which adaptively leverages the internal representations from layers less prone to hallucination to guide decoding, correcting hallucinated outputs for non-semantic samples while preserving the semantics of meaningful ones. To enable rigorous evaluation, we introduce TextHalu-Bench, a benchmark of 1, 740 samples spanning both semantic and non-semantic cases, with manually curated question–answer pairs designed to probe model hallucinations. Extensive experiments demonstrate that our method not only effectively mitigates semantic hallucination but also achieves strong performance on public benchmarks for scene text spotting and understanding.

PDF Details

IJCAI Conference 2024 Conference Paper

A Complete Landscape of EFX Allocations on Graphs: Goods, Chores and Mixed Manna

  • Yu Zhou
  • Tianze Wei
  • Minming Li
  • Bo Li

We study envy-free up to any item (EFX) allocations on graphs where vertices and edges represent agents and items respectively. An agent is only interested in items that are incident to her and all other items have zero marginal values to her. Christodoulou et al. first proposed this setting and studied the case of goods. We extend this setting to the case of mixed manna where an item may be liked or disliked by its endpoint agents. In our problem, an agent has an arbitrary valuation over her incident items such that the items she likes have non-negative marginal values to her and those she dislikes have non-positive marginal values. We provide a complete study of the four notions of EFX for mixed manna in the literature, which differ by whether the removed item can have zero marginal value. We prove that an allocation that satisfies the notion of EFX where the virtually-removed item could always have zero marginal value may not exist and determining its existence is NP-complete, while one that satisfies any of the other three notions always exists and can be computed in polynomial time. We also prove that an orientation (i. e. , a special allocation where each edge must be allocated to one of its endpoint agents) that satisfies any of the four notions may not exist, and determining its existence is NP-complete.

PDF Details DOI

AAAI Conference 2024 Conference Paper

DIUSum: Dynamic Image Utilization for Multimodal Summarization

  • Min Xiao
  • Junnan Zhu
  • Feifei Zhai
  • Yu Zhou
  • Chengqing Zong

Existing multimodal summarization approaches focus on fusing image features in the encoding process, ignoring the individualized needs for images when generating different summaries. However, whether intuitively or empirically, not all images can improve summary quality. Therefore, we propose a novel Dynamic Image Utilization framework for multimodal Summarization (DIUSum) to select and utilize valuable images for summarization. First, to predict whether an image helps produce a high-quality summary, we propose an image selector to score the usefulness of each image. Second, to dynamically utilize the multimodal information, we incorporate the hard and soft guidance from the image selector. Under the guidance, the image information is plugged into the decoder to generate a summary. Experimental results have shown that DIUSum outperforms multiple strong baselines and achieves SOTA on two public multimodal summarization datasets. Further analysis demonstrates that the image selector can reflect the improved level of summary quality brought by the images.

PDF Details DOI

NeurIPS Conference 2024 Conference Paper

Embodied Agent Interface: Benchmarking LLMs for Embodied Decision Making

  • Manling Li
  • Shiyu Zhao
  • Qineng Wang
  • Kangrui Wang
  • Yu Zhou
  • Sanjana Srivastava
  • Cem Gokmen
  • Tony Lee

We aim to evaluate Large Language Models (LLMs) for embodied decision making. While a significant body of work has been leveraging LLMs for decision making in embodied environments, we still lack a systematic understanding of their performance because they are usually applied in different domains, for different purposes, and built based on different inputs and outputs. Furthermore, existing evaluations tend to rely solely on a final success rate, making it difficult to pinpoint what ability is missing in LLMs and where the problem lies, which in turn blocks embodied agents from leveraging LLMs effectively and selectively. To address these limitations, we propose a generalized interface (Embodied Agent Interface) that supports the formalization of various types of tasks and input-output specifications of LLM-based modules. Specifically, it allows us to unify 1) a broad set of embodied decision-making tasks involving both state and temporally extended goals, 2) four commonly-used LLM-based modules for decision making: goal interpretation, subgoal decomposition, action sequencing, and transition modeling, and 3) a collection of fine-grained metrics that break down evaluation into error types, such as hallucination errors, affordance errors, and various types of planning errors. Overall, our benchmark offers a comprehensive assessment of LLMs’ performance for different subtasks, pinpointing the strengths and weaknesses in LLM-powered embodied AI systems and providing insights into the effective and selective use of LLMs in embodied decision making.

PDF Details DOI

IJCAI Conference 2024 Conference Paper

Generalized Taxonomy-Guided Graph Neural Networks

  • Yu Zhou
  • Di Jin
  • Jianguo Wei
  • Dongxiao He
  • Zhizhi Yu
  • Weixiong Zhang

Graph neural networks have been demonstrated to be effective analytic apparatus for mining network data. Most real-world networks are inherently hierarchical, offering unique opportunities to acquire latent, intrinsic network organizational properties by utilizing network taxonomies. The existing approaches for learning implicit hierarchical network structures focus on introducing taxonomy to graph neural networks but often run short of exploiting the rich network semantics and structural properties in the taxonomy, resulting in poor generalizability and reusability. To address these issues, we propose generalized Taxonomy-Guided Graph Neural Networks (TG-GNN) to integrate taxonomy into network representation learning. We first construct a taxonomy representation learning module that introduces the concept of ego network to propagate and aggregate rich semantic and structural information in the taxonomy. We then design a taxonomy-guided Markov mechanism, which encapsulates taxonomy knowledge in pairwise potential functions, to refine network embeddings. Extensive experiments on various real-world networks illustrate the effectiveness of TG-GNN over the state-of-the-art methods on scenarios involving incomplete taxonomies and inductive settings.

PDF Details DOI

TCS Journal 2024 Journal Article

Robustly reusable fuzzy extractor from isogeny

  • Yu Zhou
  • Shengli Liu
  • Shuai Han

Robustly reusable Fuzzy Extractor (rrFE) allows multiple extractions from the same fuzzy source in a reproducible way. The reusability of rrFE asks the pseudo-randomness of the extracted keys, while robustness of rrFE makes sure that active attacks can be detected during the reproduction of the extracted key. With rrFE, we are able to produce cryptographic keys for our cryptosystems from fuzzy sources, like biometrics, physical unclonable functions, etc. There are rrFE schemes from the DDH, LWE and LPN assumptions. However there is no rrFE scheme from isogeny-based assumptions up to now. In this paper, we construct the first rrFE from isogeny. To obtain such an rrFE, we propose a new framework for constructing rrFE with a core technical tool, named Enhanced Effective Group Action (EEGA). EEGA is built upon the basic Effective Group Action (EGA), and is equipped with a sampling algorithm and a derivation algorithm. With such an EEGA, the same random input can be derived to produce different pseudo-random and unpredictable outputs. The EEGA, together with other routine building blocks like secure sketch and extractor, leverages an FE to achieve reusability and robustness. We construct EEGA based on CSI-FiSh, which admits the first rrFE scheme from isogeny. Besides, we also propose another two EEGA instantiations from the DDH assumption, and this provides another approach to DDH-based rrFE, which may be of independent interest.

Details DOI

NeurIPS Conference 2024 Conference Paper

TextCtrl: Diffusion-based Scene Text Editing with Prior Guidance Control

  • Weichao Zeng
  • Yan Shu
  • Zhenhang Li
  • Dongbao Yang
  • Yu Zhou

Centred on content modification and style preservation, Scene Text Editing (STE) remains a challenging task despite considerable progress in text-to-image synthesis and text-driven image manipulation recently. GAN-based STE methods generally encounter a common issue of model generalization, while Diffusion-based STE methods suffer from undesired style deviations. To address these problems, we propose TextCtrl, a diffusion-based method that edits text with prior guidance control. Our method consists of two key components: (i) By constructing fine-grained text style disentanglement and robust text glyph structure representation, TextCtrl explicitly incorporates Style-Structure guidance into model design and network training, significantly improving text style consistency and rendering accuracy. (ii) To further leverage the style prior, a Glyph-adaptive Mutual Self-attention mechanism is proposed which deconstructs the implicit fine-grained features of the source image to enhance style consistency and vision quality during inference. Furthermore, to fill the vacancy of the real-world STE evaluation benchmark, we create the first real-world image-pair dataset termed ScenePair for fair comparisons. Experiments demonstrate the effectiveness of TextCtrl compared with previous methods concerning both style fidelity and text accuracy. Project page: https: //github. com/weichaozeng/TextCtrl.

PDF Details DOI

EAAI Journal 2024 Journal Article

YOLO_SRv2: An evolved version of YOLO_SR

  • Yong Lv
  • Yu Zhou
  • Qing Chen
  • Wenzheng Chi
  • Lining Sun
  • Lei Yu

Recently, object detection based on deep neural network has become the mainstream research direction of sweeping robots since different obstacles may produce completely contradictory sweeping decisions. However, due to the limited computational ability of the sweeping robot, the size of the object detection network is constrained, which leads to relatively low recognition accuracy and makes it difficult to be applied practically. To address this problem, we present a novel lightweight object detection network, namely YOLO_SRv2, which is an evolved version of our previously proposed You Only Look Once network for Sweeping Robots (YOLO_SR). First, an improved max-pooling structure is presented in the backbone network to avoid losing the suboptimal features. Second, a Convolutional Block Attention Module (CBAM) is added behind the backbone network to enhance the extraction of important features. Finally, a decoupled head module is included to replace the original couple head so as to solve the conflict between classification and regression tasks. In addition, we optimize the classification loss methods to improve the detection performance of hard examples. In the experimental studies, we evaluated the proposed YOLO_SRv2 with other 7 state-of-the-art methods on the ODSR-HIS dataset and PASCAL VOC dataset. The experimental results show that YOLO_SRv2 achieves 86. 1% mAP on the ODSR-HIS dataset and 78. 2% mAP on PASCAL VOC dataset with 180 FPS on RTX 2080Ti, which verifies the efficiency and the effectiveness of the proposed method as a lightweight object detection network. Source code is at https: //github. com/lvyongshjd/YOLO_SRv2.

Details DOI

TCS Journal 2023 Journal Article

Computational fuzzy extractor from LWE

  • Yu Zhou
  • Shengli Liu
  • Nan Cui

Fuzzy extractors are able to derive uniform and stable strings from noisy sources. Traditional fuzzy extractors are defined as information-theoretical primitive. Unfortunately, the components of information-theoretical fuzzy extractors consume the entropy of the sources, which results in extracted strings being too short to be applied in practice. Fuller et al. introduced the notion of Computational Fuzzy Extractor (CFE) and proposed a CFE scheme, which extracts nearly all entropy from the sources, based on the Learning with Errors (LWE) assumption. However, their construction applies to a limited type of sources and only supports sub-linear error correctness rate. In this paper, we propose a generic construction of CFE from Computationally Private Secure Sketch (CPSS) and Lossy Computational Extractor (LCExt). By instantiating CPSS and LCExt based on the LWE assumption, we obtain specific CFE schemes from LWE. Compared with the CFE scheme by Fuller et al. , our CFE scheme not only extracts almost all entropy from the source, but also supports linear error correctness rate, as long as the source has enough entropy.

Details DOI

JBHI Journal 2023 Journal Article

Cross-Hospital Sepsis Early Detection via Semi-Supervised Optimal Transport With Self-Paced Ensemble

  • Ruiqing Ding
  • Yu Zhou
  • Jie Xu
  • Yan Xie
  • Qiqiang Liang
  • He Ren
  • Yixuan Wang
  • Yanlin Chen

Leveraging machine learning techniques for Sepsis early detection and diagnosis has attracted increasing interest in recent years. However, most existing methods require a large amount of labeled training data, which may not be available for a target hospital that deploys a new Sepsis detection system. More seriously, as treated patients are diversified between hospitals, directly applying a model trained on other hospitals may not achieve good performance for the target hospital. To address this issue, we propose a novel semi-supervised transfer learning framework based on optimal transport theory and self-paced ensemble for Sepsis early detection, called SPSSOT, which can efficiently transfer knowledge from the source hospital (with rich labeled data) to the target hospital (with scarce labeled data). Specifically, SPSSOT incorporates a new optimal transport-based semi-supervised domain adaptation component that can effectively exploit all the unlabeled data in the target hospital. Moreover, self-paced ensemble is adapted in SPSSOT to alleviate the class imbalance issue during transfer learning. In a nutshell, SPSSOT is an end-to-end transfer learning method that automatically selects suitable samples from two domains (hospitals) respectively and aligns their feature spaces. Extensive experiments on two open clinical datasets, MIMIC-III and Challenge, demonstrate that SPSSOT outperforms state-of-the-art transfer learning methods by improving 1–3% of AUC.

Details DOI

IJCAI Conference 2023 Conference Paper

Divide Rows and Conquer Cells: Towards Structure Recognition for Large Tables

  • Huawen Shen
  • Xiang Gao
  • Jin Wei
  • Liang Qiao
  • Yu Zhou
  • Qiang Li
  • Zhanzhan Cheng

Recent advanced Table Structure Recognition (TSR) models adopt image-to-text solutions to parse table structure. These methods can be formulated as image caption problem, i. e. , input a single-table image and output table structure description in a specific text format, e. g. , HTML. With the impressive success of Transformer in text generation tasks, these methods use Transformer architecture to predict HTML table text in an autoregressive manner. However, tables always emerge with a large variety of shapes and sizes. Autoregressive models usually suffer from the error accumulation problem as the length of predicted text increases, which results in unsatisfactory performance for large tables. In this paper, we propose a novel image-to-text based TSR method that relieves error accumulation problems and improves performance noticeably. At the core of our method is a cascaded two-step decoder architecture with the former decoder predicting HTML table row tags non-autoregressively and the latter predicting HTML table cell tags of each row in a semi-autoregressive manner. Compared with existing methods that predict HTML text autoregressively, the superiority of our row-to-cell progressive table parsing is twofold: (1) it generates an HTML tag sequence with a vertical-and-horizontal two-step `scanning', which better fits the inherent 2D structure of image data, (2) it performs substantially better for large tables (long sequence prediction) since it alleviates error accumulation problem specific to autoregressive models. Extensive experiments demonstrate that our method achieves competitive performance on three public benchmarks.

PDF Details DOI

NeurIPS Conference 2023 Conference Paper

Fair Allocation of Indivisible Chores: Beyond Additive Costs

  • Bo Li
  • Fangxiao Wang
  • Yu Zhou

We study the maximin share (MMS) fair allocation of $m$ indivisible tasks to $n$ agents who have costs for completing the assigned tasks. It is known that exact MMS fairness cannot be guaranteed, and so far the best-known approximation for additive cost functions is $\frac{13}{11}$ by Huang and Segal-Halevi [EC, 2023]; however, beyond additivity, very little is known. In this work, we first prove that no algorithm can ensure better than $\min\{n, \frac{\log m}{\log \log m}\}$-approximation if the cost functions are submodular. This result also shows a sharp contrast with the allocation of goods where constant approximations exist as shown by Barman and Krishnamurthy [TEAC, 2020] and Ghodsi et al. [AIJ, 2022]. We then prove that for subadditive costs, there always exists an allocation that is $\min\{n, \lceil\log m\rceil\}$-approximation, and thus the approximation ratio is asymptotically tight. Besides multiplicative approximation, we also consider the ordinal relaxation, 1-out-of-$d$ MMS, which was recently proposed by Hosseini et al. [JAIR and AAMAS, 2022]. Our impossibility result implies that for any $d\ge 2$, a 1-out-of-$d$ MMS allocation may not exist. Due to these hardness results for general subadditive costs, we turn to studying two specific subadditive costs, namely, bin packing and job scheduling. For both settings, we show that constant approximate allocations exist for both multiplicative and ordinal relaxations of MMS.

PDF Details

AAAI Conference 2023 Conference Paper

One-Shot Replay: Boosting Incremental Object Detection via Retrospecting One Object

  • Dongbao Yang
  • Yu Zhou
  • Xiaopeng Hong
  • Aoting Zhang
  • Weiping Wang

Modern object detectors are ill-equipped to incrementally learn new emerging object classes over time due to the well-known phenomenon of catastrophic forgetting. Due to data privacy or limited storage, few or no images of the old data can be stored for replay. In this paper, we design a novel One-Shot Replay (OSR) method for incremental object detection, which is an augmentation-based method. Rather than storing original images, only one object-level sample for each old class is stored to reduce memory usage significantly, and we find that copy-paste is a harmonious way to replay for incremental object detection. In the incremental learning procedure, diverse augmented samples with co-occurrence of old and new objects to existing training data are generated. To introduce more variants for objects of old classes, we propose two augmentation modules. The object augmentation module aims to enhance the ability of the detector to perceive potential unknown objects. The feature augmentation module explores the relations between old and new classes and augments the feature space via analogy. Extensive experimental results on VOC2007 and COCO demonstrate that OSR can outperform the state-of-the-art incremental object detection methods without using extra wild data.

PDF Details DOI

AAMAS Conference 2023 Conference Paper

Possible Fairness for Allocating Indivisible Resources

  • Haris Aziz
  • Bo Li
  • Shiji Xing
  • Yu Zhou

Fair division of indivisible resources has attracted significant attention from multi-agent systems and computational social choice. Two popular solution concepts are envy-freeness up to any item (EFX) and maximin share (MMS) fairness which are defined using agents’ cardinal preferences. On one hand, accurate cardinal values are hard to express in real-life applications, and on the other hand, with cardinal values, MMS and EFX may not be easy to satisfy. In this work, we study a new setting where agents have arbitrary ordinal preferences for the items (possibly with indifferences), and an allocation is called possible EFX (p-EFX) or possible MMS (p- MMS) if there exist cardinal preferences that are consistent with the ordinal ones so that the allocation is EFX or MMS. We first design a polynomial-time algorithm to compute an allocation that is p-EFX and p-MMS under lexicographic preferences. This result also strengthens a result of Hosseini et al. (AAAI 2021) who proved the existence of EFX and MMS allocations under strict lexicographic preferences (i. e. , the items do not have ties). Although it has been well justified that lexicographic preferences are natural and common, there are situations where they do not fit appropriately, especially when the items have similar types. Therefore, on top of p-EFX and p-MMS, we want the allocation to be balanced (i. e. , the numbers of items allocated to the agents differ by at most one). We then design another algorithm that satisfies p-EFX, p-MMS, and balanced simultaneously.

PDF

JBHI Journal 2022 Journal Article

Fatigue-Sensitivity Comparison of sEMG and A-Mode Ultrasound based Hand Gesture Recognition

  • Jia Zeng
  • Yu Zhou
  • Yicheng Yang
  • Jipeng Yan
  • Honghai Liu

Though physiological signal based human-machine interfaces (HMIs) have recently developed rapidly, their practical use is restricted by many real-world environmental factors, one of which is muscle fatigue. This paper explores the sensitivities between surface electromyography (sEMG) and A-mode ultrasound (AUS) sensing modalities subject to muscle fatigue in the context of hand gesture recognition tasks. Two metrics, mean classification accuracy ( $mCA$ ) and decline rate ( $DR$ ), are proposed to evaluate the accuracy and muscle fatigue sensitivity between sEMG and AUS based HMIs. Muscle fatigue inducing experiment was designed and eight subjects were recruited to participate in the experiment. The gesture recognition accuracies of sEMG and AUS under non-fatigue state and fatigue state are compared through Mahalanobis distance based classifier linear discriminant analysis (LDA). In addition, Mahalanobis distance based metrics, repeatability index ( $RI$ ) and separability index ( $SI$ ), are introduced to evaluate the changes in the feature distribution during muscle fatigue and reveal the cause of the fatigue sensitivity difference between sEMG and AUS signals. The experimental results demonstrate that the fatigue robustness of AUS signal is better than that of sEMG signal. Specifically, with the employment of the LDA classifier trained under non-fatigue state, the testing accuracy of the sEMG signal on the non-fatigue state is 94. 96%, while reduce to 68. 26% on the fatigue state. The testing accuracy of the AUS signal on the corresponding states is 99. 68% and 91. 24% respectively. AUS signal attains higher $mCA$ and lower $DR$, indicating that it has advantages over sEMG signal in terms of both accuracy and muscle fatigue sensitivity. In addition, the $RI$ and $RI/SI$ analysis reveal that before and after muscle fatigue, the consistency of AUS feature distribution is better than that of sEMG. These research outcomes validate that AUS is more tolerant to feature migration caused by muscle fatigue than sEMG.

Details DOI

TIST Journal 2022 Journal Article

Graph Neural Networks: Taxonomy, Advances, and Trends

  • Yu Zhou
  • Haixia Zheng
  • Xin Huang
  • Shufeng Hao
  • Dengao Li
  • Jumin Zhao

Graph neural networks provide a powerful toolkit for embedding real-world graphs into low-dimensional spaces according to specific tasks. Up to now, there have been several surveys on this topic. However, they usually lay emphasis on different angles so that the readers cannot see a panorama of the graph neural networks. This survey aims to overcome this limitation and provide a systematic and comprehensive review on the graph neural networks. First of all, we provide a novel taxonomy for the graph neural networks, and then refer to up to 327 relevant literatures to show the panorama of the graph neural networks. All of them are classified into the corresponding categories. In order to drive the graph neural networks into a new stage, we summarize four future research directions so as to overcome the challenges faced. It is expected that more and more scholars can understand and exploit the graph neural networks and use them in their research community.

Details DOI

AAAI Conference 2022 Conference Paper

Imagine by Reasoning: A Reasoning-Based Implicit Semantic Data Augmentation for Long-Tailed Classification

  • Xiaohua Chen
  • Yucan Zhou
  • Dayan Wu
  • Wanqian Zhang
  • Yu Zhou
  • Bo Li
  • Weiping Wang

Real-world data often follows a long-tailed distribution, which makes the performance of existing classification algorithms degrade heavily. A key issue is that samples in tail categories fail to depict their intra-class diversity. Humans can imagine a sample in new poses, scenes, and view angles with their prior knowledge even if it is the first time to see this category. Inspired by this, we propose a novel reasoning-based implicit semantic data augmentation method to borrow transformation directions from other classes. Since the covariance matrix of each category represents the feature transformation directions, we can sample new directions from similar categories to generate definitely different instances. Specifically, the long-tailed distributed data is first adopted to train a backbone and a classifier. Then, a covariance matrix for each category is estimated, and a knowledge graph is constructed to store the relations of any two categories. Finally, tail samples are adaptively enhanced via propagating information from all the similar categories in the knowledge graph. Experimental results on CIFAR-100-LT, ImageNet-LT, and iNaturalist 2018 have demonstrated the effectiveness of our proposed method compared with the state-of-the-art methods.

PDF Details

YNIMG Journal 2020 Journal Article

Individual-specific and shared representations during episodic memory encoding and retrieval

  • Xiaoqian Xiao
  • Yu Zhou
  • Jing Liu
  • Zhifang Ye
  • Li Yao
  • Jiacai Zhang
  • Chuansheng Chen
  • Gui Xue

Although human memories seem unique to each individual, they are shared to a great extent across individuals. Previous studies have examined, separately, subject-specific and cross-subject shared representations during memory encoding and retrieval, but how shared memories are formed from individually encoded representations is not clearly understood. Using a unique fMRI design involving memory encoding and retrieval, and representational similarity analysis to link representations from different individuals, brain regions, and processing stages, the current study revealed that distributed brain regions showed both subject-specific and shared neural representations during both memory encoding and retrieval. Furthermore, different brain regions showed stage-specific representational strength, with the visual cortex showing greater unique and shared representations during encoding, whereas the left angular gyrus showing greater unique and shared representations during retrieval. The neural representations during encoding were transformed during retrieval, as shown by smaller cross-subject encoding-retrieval similarity (ERS) than cross-subject similarity either during encoding or during retrieval. This cross-subject and cross-stage similarity was found both within and across regions, with strong pattern similarity between the encoded representation in VVC and the retrieved representation in the angular gyrus. Simulation analysis further suggested that these patterns could be achieved by incorporating stage-specific representational strength, and cross-region reinstatement from encoding to retrieval, but not by a common transformation from encoding to retrieval across subjects. Together, our results shed light on how memory representations are encoded and transformed to maintain individual characteristics and at the same time to create shared representations to facilitate interpersonal communication.

Details DOI

IJCAI Conference 2020 Conference Paper

Knowledge Graphs Enhanced Neural Machine Translation

  • Yang Zhao
  • Jiajun Zhang
  • Yu Zhou
  • Chengqing Zong

Knowledge graphs (KGs) store much structured information on various entities, many of which are not covered by the parallel sentence pairs of neural machine translation (NMT). To improve the translation quality of these entities, in this paper we propose a novel KGs enhanced NMT method. Specifically, we first induce the new translation results of these entities by transforming the source and target KGs into a unified semantic space. We then generate adequate pseudo parallel sentence pairs that contain these induced entity pairs. Finally, NMT model is jointly trained by the original and pseudo sentence pairs. The extensive experiments on Chinese-to-English and Englishto-Japanese translation tasks demonstrate that our method significantly outperforms the strong baseline models in translation quality, especially in handling the induced entities.

PDF Details DOI

AAAI Conference 2020 Conference Paper

Multimodal Summarization with Guidance of Multimodal Reference

  • Junnan Zhu
  • Yu Zhou
  • Jiajun Zhang
  • Haoran Li
  • Chengqing Zong
  • Changliang Li

Multimodal summarization with multimodal output (MSMO) is to generate a multimodal summary for a multimodal news report, which has been proven to effectively improve users’ satisfaction. The existing MSMO methods are trained by the target of text modality, leading to the modality-bias problem that ignores the quality of model-selected image during training. To alleviate this problem, we propose a multimodal objective function with the guidance of multimodal reference to use the loss from the summary generation and the image selection. Due to the lack of multimodal reference data, we present two strategies, i. e. , ROUGE-ranking and Orderranking, to construct the multimodal reference by extending the text reference. Meanwhile, to better evaluate multimodal outputs, we propose a novel evaluation metric based on joint multimodal representation, projecting the model output and multimodal reference into a joint semantic space during evaluation. Experimental results have shown that our proposed model achieves the new state-of-the-art on both automatic and manual evaluation metrics. Besides, our proposed evaluation method can effectively improve the correlation with human judgments.

PDF Details

AAAI Conference 2020 Conference Paper

TANet: Robust 3D Object Detection from Point Clouds with Triple Attention

  • Zhe Liu
  • Xin Zhao
  • Tengteng Huang
  • Ruolan Hu
  • Yu Zhou
  • Xiang Bai

In this paper, we focus on exploring the robustness of the 3D object detection in point clouds, which has been rarely discussed in existing approaches. We observe two crucial phenomena: 1) the detection accuracy of the hard objects, e. g. , Pedestrians, is unsatisfactory, 2) when adding additional noise points, the performance of existing approaches decreases rapidly. To alleviate these problems, a novel TANet is introduced in this paper, which mainly contains a Triple Attention (TA) module, and a Coarse-to-Fine Regression (CFR) module. By considering the channel-wise, point-wise and voxel-wise attention jointly, the TA module enhances the crucial information of the target while suppresses the unstable cloud points. Besides, the novel stacked TA further exploits the multi-level feature attention. In addition, the CFR module boosts the accuracy of localization without excessive computation cost. Experimental results on the validation set of KITTI dataset demonstrate that, in the challenging noisy cases, i. e. , adding additional random noisy points around each object, the presented approach goes far beyond state-of-theart approaches. Furthermore, for the 3D object detection task of the KITTI benchmark, our approach ranks the first place on Pedestrian class, by using the point clouds as the only input. The running speed is around 29 frames per second.

PDF Details

TCS Journal 2020 Journal Article

The k-power domination problem in weighted trees

  • Changjie Cheng
  • Changhong Lu
  • Yu Zhou

The power domination problem of the graph comes from how to choose the node location of the least phase measurement units in the electric power system. In the actual electric power system, because of the difference in the cost of phase measurement units at different nodes, it is more practical to study the power domination problem with the weighted graph. In this paper, we present a dynamic programming style linear-time algorithm for k-power domination problem in weighted trees.

Details DOI

AIIM Journal 2020 Journal Article

Upper-limb functional assessment after stroke using mirror contraction: A pilot study

  • Yu Zhou
  • Jia Zeng
  • Hongze Jiang
  • Yang Li
  • Jie Jia
  • Honghai Liu

The clinical assessment after stroke depends on the rating scale, usually lack of quantitative feedback such as biomedical signal captured from stroke patients. This study attempts to develop a unified assessment framework for persons after stroke via surface electromyography (sEMG) bias from bilateral limbs, based on four types of selected movements, namely forward lift arm, lateral lift arm, forearm internal/external rotation, forearm pronation/supination. Eleven healthy subjects and six stroke patients are recruited to participate in the experiment to perform the bilateral-mirrored paradigm with six channels of sEMG signals recorded from each of their arms. The linear discriminant analysis (LDA), random forest algorithm (RF) and support vector machine (SVM) are adopted, trained and used for stroke patients qualitative recognition. The bilateral bias diagnosis algorithm (BBDA) is developed to evaluate the stroke severity quantitatively based on the similarity index (SI) of the sEMG. The results reveal that: (1) the sEMG feature bias of bilateral arms for stroke patients is different from that of healthy people; (2) the RF and SVM demonstrate a better performance with an average recognition accuracy of 0. 92 ± 0. 12 and 0. 93 ± 0. 12 than LDA (0. 84 ± 0. 20) in distinguishing stroke patients from healthy subjects; (3) there is a strong positive correlation between SI and the Fugl-Meyer score (r = 0. 93). These research findings indicate that the dominant qualitative assessment after stroke could be complementary by its counterpart quantitative solutions, and stroke rehabilitation could be automated with less involvement of professional therapists.

Details DOI

AAAI Conference 2020 Conference Paper

Video Cloze Procedure for Self-Supervised Spatio-Temporal Learning

  • Dezhao Luo
  • Chang Liu
  • Yu Zhou
  • Dongbao Yang
  • Can Ma
  • Qixiang Ye
  • Weiping Wang

We propose a novel self-supervised method, referred to as Video Cloze Procedure (VCP), to learn rich spatial-temporal representations. VCP first generates “blanks” by withholding video clips and then creates “options” by applying spatiotemporal operations on the withheld clips. Finally, it fills the blanks with “options” and learns representations by predicting the categories of operations applied on the clips. VCP can act as either a proxy task or a target task in self-supervised learning. As a proxy task, it converts rich self-supervised representations into video clip operations (options), which enhances the flexibility and reduces the complexity of representation learning. As a target task, it can assess learned representation models in a uniform and interpretable manner. With VCP, we train spatial-temporal representation models (3D-CNNs) and apply such models on action recognition and video retrieval tasks. Experiments on commonly used benchmarks show that the trained models outperform the state-ofthe-art self-supervised models with significant margins.

PDF Details

JBHI Journal 2019 Journal Article

Towards Zero Re-Training for Long-Term Hand Gesture Recognition via Ultrasound Sensing

  • Xingchen Yang
  • Dalin Zhou
  • Yu Zhou
  • Youjia Huang
  • Honghai Liu

While myoelectric pattern recognition is a prevailing way for gesture recognition, the inherent nonstationarity of electromyography signals hinders its long-term application. This study aims to prove a hypothesis that morphological information of muscle contraction detected by ultrasound image is potentially suitable for long-term use. A set of ultrasound-based algorithms are proposed to realize robust hand gesture recognition over multiple days, with user training only at the first day. A markerless calibration algorithm is first presented to position the ultrasound probe during donning and doffing; an algorithm combining speeded-up robust features and bag-of-features model being immune to ultrasound probe shift and rotation is then introduced; a self-enhancing classification method is next adopted to update classification model automatically by incorporating useful knowledge from testing data; finally the performance of long-term hand gesture recognition with zero re-training is validated by a six-day experiment of six healthy subjects, whose outcomes strongly support the hypothesis with about 94% of gesture recognition accuracy for each testing day. This study confirms the feasibility of adoption of ultrasound sensing for long-term musculature related applications.

Details DOI

ICRA Conference 2018 Conference Paper

Historical Data is Useful for Navigation Planning: Data Driven Route Generation for Autonomous Ship

  • Wei-Chian Tan
  • Ching-Yen Weng
  • Yu Zhou
  • Kie Hian Chua
  • I-Ming Chen 0001

This work presents a method for automated generation of navigation plan for autonomous or robotic surface vessel. Historical Automatic Identification System (AIS) data is of significant value to this problem. The method joins AIS locations of a same vessel at different time and locations in a region into a route. Next, it automatically computes navigation plans using nearest neighbour based path retrieval relying on two representations, Ship Feature and Navigation Feature. Before starting service, existing AIS records in the form of ship properties and corresponding route are preprocessed and stored in the form of Ship and Navigation Feature. During online retrieval, given input constraints in vector form, nearest neighbour of this query vector in the same space is found and corresponding path of the neighbour is returned as recommended path. Analysis was done in four and two dimensional spaces for Ship and Navigation Feature respectively. Application of the method is demonstrated in two regions of Australian, covering Bass Strait and Great Australian Bight.

Details

IS Journal 2016 Journal Article

Monocular Depth-Ordering Reasoning with Occlusion Edge Detection and Couple Layers Inference

  • Anlong Ming
  • Tianfu Wu
  • Jianxiang Ma
  • Fang Sun
  • Yu Zhou

A depth-ordering reasoning approach first provides novel occlusion edge detection, generating precise same-layer relationship judgment and producing reliable region proposals for the depth-ordering inference. Specifically, a novel sparsity-induced regression model learns a discriminative feature subspace. In addition, kernel ridge regression assigns the occlusion label for each edge. The kernel trick guarantees linearly separable edges in a rich, high-dimensional feature space. Secondly, a couple layers inference approach infers the final depth order. In the semilocal layer, a novel triple descriptor judges the foreground relationship. In the global layer, the inference is executed by finding a valid path on a directed graph model. The proposed approach is validated on the Cornell depth-order and NYU 2 datasets.

Details DOI

IJCAI Conference 2015 Conference Paper

A New Input Method for Human Translators: Integrating Machine Translation Effectively and Imperceptibly

  • Guoping Huang
  • Jiajun Zhang
  • Yu Zhou
  • Chengqing Zong

Computer-aided translation (CAT) system is the most popular tool which helps human translators perform language translation efficiently. To further improve the efficiency, there is an increasing interest in applying the machine translation (MT) technology to upgrade CAT. Post-editing is a standard approach: human translators generate the translation by correcting MT outputs. In this paper, we propose a novel approach deeply integrating MT into CAT systems: a well-designed input method which makes full use of the knowledge adopted by MT systems, such as translation rules, decoding hypotheses and n-best translation lists. Our proposed approach allows human translators to focus on choosing better translation results with less time rather than just complete translation themselves. The extensive experiments demonstrate that our method saves more than 14% time and over 33% keystrokes, and it improves the translation quality as well by more than 3 absolute BLEU scores compared with the strong baseline, i. e. , post-editing using Google Pinyin.

PDF Details

NeurIPS Conference 2012 Conference Paper

Fusion with Diffusion for Robust Visual Tracking

  • Yu Zhou
  • Xiang Bai
  • Wenyu Liu
  • Longin Latecki

A weighted graph is used as an underlying structure of many algorithms like semi-supervised learning and spectral clustering. The edge weights are usually deter-mined by a single similarity measure, but it often hard if not impossible to capture all relevant aspects of similarity when using a single similarity measure. In par-ticular, in the case of visual object matching it is beneficial to integrate different similarity measures that focus on different visual representations. In this paper, a novel approach to integrate multiple similarity measures is pro-posed. First pairs of similarity measures are combined with a diffusion process on their tensor product graph (TPG). Hence the diffused similarity of each pair of ob-jects becomes a function of joint diffusion of the two original similarities, which in turn depends on the neighborhood structure of the TPG. We call this process Fusion with Diffusion (FD). However, a higher order graph like the TPG usually means significant increase in time complexity. This is not the case in the proposed approach. A key feature of our approach is that the time complexity of the dif-fusion on the TPG is the same as the diffusion process on each of the original graphs, Moreover, it is not necessary to explicitly construct the TPG in our frame-work. Finally all diffused pairs of similarity measures are combined as a weighted sum. We demonstrate the advantages of the proposed approach on the task of visual tracking, where different aspects of the appearance similarity between the target object in frame t and target object candidates in frame t+1 are integrated. The obtained method is tested on several challenge video sequences and the experimental results show that it outperforms state-of-the-art tracking methods.

PDF Details

ICRA Conference 2007 Conference Paper

Laser-activated RFID-based Indoor Localization System for Mobile Robots

  • Yu Zhou
  • Wenfei Liu
  • Peisen Huang

Localization is a fundamental problem in autonomous mobile robot navigation. This paper introduces a new artificial landmark-based localization system for mobile robots navigating in indoor environments. Laser-activated RFID tag is designed and used as the artificial landmark in the proposed localization system. The robot localization is realized via the combination of the stereo vision and laser-activated RFID based on the principle of triangulation. The localization system functions like an indoor GPS. Preliminary research shows that the proposed system is promising to provide a robust and accurate indoor localization method for mobile robots.

Details

IJCAI Conference 2005 Conference Paper

A Machine Learning Approach to Identification and Resolution of One-Anaphora

  • Hwee Tou Ng
  • Yu Zhou
  • Robert Dale
  • Mary

We present a machine learning approach to identifying and resolving one-anaphora. In this approach, the system first learns to distinguish different uses of instances of the word one; in the second stage, the antecedents of those instances of one that are classified as anaphoric are then determined. We evaluated our approach on written texts drawn from the informative domains of the British National Corpus (BNC), and achieved encouraging results. To our knowledge, this is the first learningbased system for the identification and resolution of one-anaphora.

PDF

NeurIPS Conference 2003 Conference Paper

Impact of an Energy Normalization Transform on the Performance of the LF-ASD Brain Computer Interface

  • Yu Zhou
  • Steven Mason
  • Gary Birch

This paper presents an energy normalization transform as a method to reduce system errors in the LF-ASD brain-computer interface. The energy normalization transform has two major benefits to the system performance. First, it can increase class separation between the active and idle EEG data. Second, it can desensitize the system to the signal amplitude variability. For four subjects in the study, the benefits resulted in the performance improvement of the LF-ASD in the range from 7. 7% to 18. 9%, while for the fifth subject, who had the highest non-normalized accuracy of 90. 5%, the performance did not change notably with normalization.

PDF Details

ICRA Conference 2000 Conference Paper

The Effect of Material Properties and Gripping Force on Micrograsping

  • Yu Zhou
  • Bradley J. Nelson

This paper presents our work in developing a force controlled microgripper and micrograsping strategies using optical beam deflection techniques. The optical beam deflection sensor is based on modified atomic force microscopy techniques and is able to resolve forces below a nano-Newton. A variety of gripper fingers made from materials with different conductivity and surface roughness is analyzed theoretically and experimentally using the force sensor. These results provide insight into the mechanics of micromanipulation, and the results are used to develop micrograsping strategies. A design of a microfabricated force controlled microgripper is presented along with initial experimental results in applying various gripping forces to microparts. The results demonstrate the important role gripping force plays in the grasping and releasing of microparts.

Details

ICRA Conference 1998 Conference Paper

Fusing Force and Vision Feedback for Micromanipulation

  • Yu Zhou
  • Bradley J. Nelson
  • Barmeshwar Vikramaditya

We present experimental results that investigate the integration of two disparate sensing modalities, force and vision, for sensor-based microassembly. By integrating these sensing modes, we are able to provide feedback in a task-oriented frame of reference over a broad range of motion with an extremely high precision. An optical microscope is used to provide visual feedback down to micron resolutions. We have developed an optical beam deflection sensor to provide nanonewton level force feedback or nanometric level position feedback. The value of integrating these two disparate sensing modalities is demonstrated during controlled micropart impact experiments. Using force feedback alone to control micropart contact transitions, impact forces of over 140 nN were generated before the desired contact force of 2 nN was achieved. When visual servoing is integrated with the force control framework, micropart impact forces of only 9 nN and final contact forces of 2 nN were easily achieved.

Details
v2026.09.13