arXiv subjects

Fengnian Zhao

Publications and source records attributed to Fengnian Zhao.

3 recordsLinked to original sources

arXiv2026

FlowForge: A Staged Local Rollout Engine for Flow-Field Prediction

Deep learning surrogates for CFD flow-field prediction often rely on large, complex models, which can be slow and fragile when data are noisy or incomplete. We introduce FlowForge, a staged local rollout engine that predicts future flow fields by compiling a locality-preserving update schedule and executing it with a shared lightweight local predictor. Rather than producing the next frame in a single global pass, FlowForge rewrites spatial sites stage by stage so that each update conditions only on bounded local context exposed by earlier stages. This compile-execute design aligns inference with short-range physical dependence, keeps latency predictable, and limits error amplification from global mixing. Across PDEBench, CFDBench, and BubbleML, FlowForge matches or improves upon strong baselines in pointwise accuracy, delivers consistently better robustness to noise and missing observations, and maintains stable multi-step rollout behavior while reducing per-step latency.

cs.LG↗

arXiv2026

CXR-LT 2026 Challenge: Multi-Center Long-Tailed and Zero Shot Chest X-ray Classification

Chest X-ray (CXR) interpretation is hindered by the long-tailed distribution of pathologies and the open-world nature of clinical environments. Existing benchmarks often rely on closed-set classes from a single institution, failing to capture the prevalence of rare diseases or the appearance of novel findings. To address this, we present the CXR-LT challenge. The first event, CXR-LT 2023, established a large-scale benchmark for long-tailed multi-label CXR classification and identified key challenges in rare disease recognition. CXR-LT 2024 further expanded the label space and introduced a zero-shot task to study generalization to unseen findings. Building on the success of CXR-LT 2023 and 2024, this third iteration of the benchmark introduces a multi-center dataset comprising over 145,000 images from PadChest and NIH Chest X-ray datasets. Additionally, all development and test sets in CXR-LT 2026 are annotated by radiologists, providing a more reliable and clinically grounded evaluation than report-derived labels. The challenge defines two core tasks this year: (1) Robust Multi-Label Classification on 30 known classes and (2) Open-World Generalization to 6 unseen (out-of-distribution) rare disease classes. This paper summarizes the overview of the CXR-LT 2026 challenge. We describe the data collection and annotation procedures, analyze solution strategies adopted by participating teams, and evaluate head-versus-tail performance, calibration, and cross-center generalization gaps. Our results show that vision-language foundation models improve both in-distribution and zero-shot performance, but detecting rare findings under multi-center shift remains challenging. Our study provides a foundation for developing and evaluating AI systems in realistic long-tailed and open-world clinical conditions.

cs.CV↗

arXiv2021

Soft-Median Choice: An Automatic Feature Smoothing Method for Sound Event Detection

In Sound Event Detection (SED) systems, the lengths of median filters for post-processing have never been optimized during training due to several problems. No gradient is received by the lengths so they cannot be learned during back-propagation. The median-filtering inserted in the models also causes block in gradient flowing and the smoothing process misleads the model by ignoring errors. To resolve these problems, we provide different channels of features smoothed to different extents along with the original feature, so the model can optimize the weights while cognizing all the errors. We then use a linear layer to integrate the results and produce a linear combination. We further design the soft-median function to dredge the gradient flow. The proposed framework is called Soft-Median Choice (SMC). Experiments show that the SMC block not only automatically smooths the features based on the training set, but also forces the model to extract common features shared by all the frames of a sound event. The performance of the proposed method outperforms the baseline by over 10% of Event-Based F1 Score (EBFS) in both the validation and the evaluation set, and also slightly outperforms the single model of the state-of-the-art SED system.

eess.AS↗