arXiv subjects

Yuqing Feng

Publications and source records attributed to Yuqing Feng.

3 recordsLinked to original sources

arXiv2026

SurgNarrator: A Generative Retrieval Framework for Surgical Video Understanding

Surgical procedures unfold as structured and recurring clinical events, whose real-time understanding via intraoperative surgical videos is critical for intraoperative decision-making and support. However, existing video understanding methods force a trade-off: autoregressive video-language models support comprehensive reasoning but are not practical for time-sensitive clinical applications, whereas contrastive models offer low latency but struggle with complex scene understanding. Recently, generative retrieval has been explored for general-domain video understanding, but transferring it to surgery is not trivial because near-identical visual appearances may indicate semantically distinct events, and the terminology involved is highly surgery-specific. To this end, we propose SurgNarrator, a new generative retrieval framework tailored for surgical video understanding. We construct a well-curated surgery-centric vocabulary from surgical captions to define a clinically meaningful retrieval space. We then adapt the pre-trained Qwen3-VL-Embedding-8B to learn discriminative clinical representations with a temporally-aware contrastive objective. During inference, a hierarchical, procedure-aware retrieval strategy narrows the search space to the relevant procedure type, delivering fast and effective responses. Our method is comprehensively evaluated on twelve benchmarks in a zero-shot setting and achieves consistent performance gains over state-of-the-art baselines, while reducing output-stage latency by more than two orders of magnitude compared with the generative baseline.

cs.CV↗

arXiv2025

Impacts of flow velocity and microbubbles on water flushing in a horizontal pipeline

Water flushing to remove particle sediment is essential for safe and continuous transport of many industrial slurries through pipelines. An efficient flushing strategy may reduce water consumption and the cost associated with water usage, and help water conservation for sustainability. In this study, a computational fluid dynamics (CFD) model coupled with the kinetic theory of granular flow for the flushing process is presented. The CFD models were validated against field data collected from a coal slurry pipeline of 128 $km$ in length, 0.575~$m$ in diameter, achieving an average error of less than 15\% for outlet solid concentration over time. A parametric study evaluated the effects of water velocity (1.88-5.88~$m/s$), bubble size (50~$\mu m$, 150~$\mu m$, and 1000~$\mu m$) and bubble volume fraction (0.05-0.2) on flushing performance including pipeline cleanness, cleanness efficiency, and water consumption. The obtained outcomes indicate that higher water velocity is preferred and an increase in water velocity from $1.88~m/s$ to $5.88~m/s$ reduces the water consumption by $28\%$. Large bubbles may hinder the flushing process and increase the water consumption by $23\%$. Remarkably, small bubbles facilitates the flushing process and lead to $35\%$ reduction in water consumption. These effects are attributed to the turbulent characteristics in the pipelines in presence of microbubbles.

physics.flu-dyn↗

arXiv2025

City-Level Foreign Direct Investment Prediction with Tabular Learning on Judicial Data

To advance the United Nations Sustainable Development Goal on promoting sustained, inclusive, and sustainable economic growth, foreign direct investment (FDI) plays a crucial role in catalyzing economic expansion and fostering innovation. Precise city-level FDI prediction is quite important for local government and is commonly studied based on economic data (e.g., GDP). However, such economic data could be prone to manipulation, making predictions less reliable. To address this issue, we try to leverage large-scale judicial data which reflects judicial performance influencing local investment security and returns, for city-level FDI prediction. Based on this, we first build an index system for the evaluation of judicial performance over twelve million publicly available adjudication documents according to which a tabular dataset is reformulated. We then propose a new Tabular Learning method on Judicial Data (TLJD) for city-level FDI prediction. TLJD integrates row data and column data in our built tabular dataset for judicial performance indicator encoding, and utilizes a mixture of experts model to adjust the weights of different indicators considering regional variations. To validate the effectiveness of TLJD, we design cross-city and cross-time tasks for city-level FDI predictions. Extensive experiments on both tasks demonstrate the superiority of TLJD (reach to at least 0.92 R2) over the other ten state-of-the-art baselines in different evaluation metrics.

cs.AI↗