Author name cluster

Bo Du

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.

74 papers

2 author rows

EAAI Journal 2026 Journal Article

Exploring magnetic actuation automation: Learning from noisy demonstrations via adaptive sampling policy

Xutian Deng
Jianhui Zhao
Zhiyong Yuan
Bo Du
Miao Li
Zhijian Yang

Magnetic actuation enables contactless manipulation of miniaturized objects and holds great potential for vascular interventional surgery. Achieving effective operation in such complex environments requires robust and adaptable control algorithms. Data-driven approaches offer a promising avenue, but collecting sufficient high-quality demonstrations is costly and challenging. As a result, available training data are often sparse and noisy, limiting the robustness, generalization, and direct applicability of learning-based methods in real-world vascular procedures. To overcome these challenges, we propose a learning-based framework that leverages noisy demonstrations to automate dual-arm magnetic actuation in simulated vascular environments. The framework integrates multimodal inputs, including visual, pose, and force information, and employs an adaptive sampling policy to identify the most informative demonstrations. This design enables end-to-end control of robotic joints while reducing data requirements and enhancing robustness. Extensive experiments show that our method outperforms mainstream learning algorithms in offline benchmarks and achieves zero-shot deployment on a magnetic actuation prototype, successfully performing in-vitro aortic vascular intervention procedures using only 20 noisy demonstration trajectories. Quantitatively, compared with the uniform sampling baseline, our method reduces task completion time by 6. 5 s and increases valid magnetic actuation steps by 5. 44% in real-world execution, highlighting its superior robustness, sample efficiency, and clinical applicability in vascular interventional tasks.

Details DOI

EAAI Journal 2026 Journal Article

Learning electromagnetic diffusion policies from mixed demonstrations in magnetic-assisted surgical contexts

Xutian Deng
Hang Qi
Jianhui Zhao
Zhiyong Yuan
Bo Du
Miao Li
Tingbao Zhang
Zhijian Yang

Electromagnetic actuation is increasingly important in medical robotics for contactless manipulation and the miniaturization of surgical tools, yet achieving autonomous operation with sufficient feasibility and dexterity remains challenging. In this work, we propose a mixed demonstration framework combining virtual and physical data to enrich training diversity, improve generalization, and bridge the simulation-to-reality gap in electromagnetic actuation. We further introduce a conditional generative policy based on diffusion models for electromagnetic field planning, which bypasses traditional multi-step pipelines by directly generating time-horizon control signals. Our approach supports real-time deployment with diverse interfaces (e. g. , human, audio, and large language models) and significantly outperforms regression- and estimation-based baselines across three surgical instruments and working contexts. These results demonstrate robust and dexterous autonomy, highlighted by strong transferability and scalability across real-world scenarios.

Details DOI

JBHI Journal 2026 Journal Article

MsGA: Gestational Age Estimation with Multi-plane Unified Measurements Driven by Anatomic Segmentation

Mingjun Huang
Junbo Zhang
Wei Hu
Chao Sun
Xiantao Cai
Bo Du

An accurate estimation of gestational age is critical for prenatal care and clinical decision-making. Existing ultrasound-based gestational age estimation methods are limited by the insufficient information representation capacity of conventional medical segmentation models, noise interference in ultrasound images, and inter-observer variability in traditional geometry-based measurement methods. To address these challenges, we propose the MsGA model to estimate gestational age with multi-plane unified measurements driven by anatomic segmentation. In the anatomic segmentation stage, a lightweight and high-performance LGF-UNet module is proposed, which utilizes the Deep Patch Embedding module to expand the receptive field, the Local-Global Fusion Transformer block to enhance local-global feature fusion, and the Focusing Attention Bottleneck module to suppress ultrasound noise via an adaptive threshold. In the measurement stage, a Point Regression module is introduced to refine biometric landmark localization. Furthermore, we create a fully annotated ultrasound plane dataset for the estimation of gestational age across various gestational stages. Extensive experiments on the dataset have demonstrated the effectiveness of the whole model and each module. Our MsGA model is superior to existing models with fewer parameters and achieves state-of-the-art performance on the Gestational Age Estimation task.

Details DOI

AAAI Conference 2026 Conference Paper

Rethinking Visual Token Reduction in LVLMs Under Cross-Modal Misalignment

Rui Xu
Yunke Wang
Yong Luo
Bo Du

Large Vision-Language Models (LVLMs) encode visual inputs as dense sequences of patch-level tokens to capture fine-grained semantics. These visual tokens often outnumber their textual counterparts by a large margin, leading to substantial computational overhead and limiting the scalability of LVLMs in practice. Previous efforts have explored visual token reduction either prior to or within the large language models (LLMs). However, most in-LLM reduction approaches rely on text-conditioned interactions, implicitly assuming that textual tokens can reliably capture the importance of visual tokens. In this work, we revisit this assumption and reveal causal, semantic, and spatial forms of cross-modal misalignment. These misalignments undermine the effectiveness of text-guided visual token reduction. To address this, we introduce VisionDrop, a training-free, visual-only pruning framework that selects informative visual tokens based on intra-modal (visual-to-visual) attention, without relying on textual signals. To further suppress redundancy throughout the model hierarchy, we treat the visual encoder and the LLM as a unified system and design a progressive pruning pipeline. Our method performs dominant token selection and lightweight contextual merging at multiple stages, enabling fine-grained visual information to be retained even under aggressive token budgets. Extensive experiments across diverse benchmarks show that VisionDrop achieves consistent improvements over existing approaches, despite requiring no additional training or complex modifications. Notably, when integrated with LLaVA-NeXT-7B, VisionDrop achieves a 2.7x reduction in inference latency and 6x in FLOPs, while retaining 95.71% of the original performance.

PDF Details DOI

NeurIPS Conference 2025 Conference Paper

AiDE-Q: Synthetic Labeled Datasets Can Enhance Learning Models for Quantum Property Estimation

Xinbiao Wang
Yuxuan Du
Zihan Lou
Yang Qian
Kaining Zhang
Yong Luo
Bo Du
Dacheng Tao

Quantum many-body problems are central to various scientific disciplines, yet their ground-state properties are intrinsically challenging to estimate. Recent advances in deep learning (DL) offer potential solutions in this field, complementing prior purely classical and quantum approaches. However, existing DL-based models typically assume access to a large-scale and noiseless labeled dataset collected by infinite sampling. This idealization raises fundamental concerns about their practical utility, especially given the limited availability of quantum hardware in the near term. To unleash the power of these DL-based models, we propose AiDE-Q (\underline{a}utomat\underline{i}c \underline{d}ata \underline{e}ngine for \underline{q}uantum property estimation), an effective framework that addresses this challenge by iteratively generating high-quality synthetic labeled datasets. Specifically, AiDE-Q utilizes a confidence-check method to assess the quality of synthetic labels and continuously improves the employed DL models with the identified high-quality synthetic dataset. To verify the effectiveness of AiDE-Q, we conduct extensive numerical simulations on a diverse set of quantum many-body and molecular systems, with up to 50 qubits. The results show that AiDE-Q enhances prediction performance for various reference learning models, with improvements of up to $14. 2\\%$. Moreover, we exhibit that a basic supervised learning model integrated with AiDE-Q outperforms advanced reference models, highlighting the importance of a synthetic dataset. Our work paves the way for more efficient and practical applications of DL for quantum property estimation.