Searcharxiv⌕ Search

arXiv subjects

Zheyun Qin

Publications and source records attributed to Zheyun Qin.

3 recordsLinked to original sources

LogicEye: Agent Driven Multimodal Dataset for Interpretable Ophthalmic Diagnosis

Clinical adoption of ophthalmic Artificial Intelligence remains limited by low interpretability and a lack of finely annotated multimodal data. Most existing datasets provide only coarse disease labels without the granular pathological findings required for differential diagnosis. We present LogicEye$^{Agent}$, a dual-pool inference and verification agent that emulates clinical reasoning by jointly verifying positive inclusion signs and negative exclusion criteria. Using LogicEye$^{Agent}$, we curate LogicEye$^{Data}$, a large-scale multimodal dataset of 804,461 fundus and OCT images paired with structured pathological reports spanning 38 diseases and 29 subtypes, with quality assured through joint verification by multimodal large language models and clinical experts. Building on this dataset, we further develop LogicEye$^{Bench}$, which incorporates a category-constrained loss for hierarchical disease recognition and report generation. Grounding coarse labels into verified visual findings effectively curtails diagnostic hallucinations and establishes a solid foundation for interpretable vision-language modeling in ophthalmology. The code is available at:https://github.com/Eden35/LogicEye-Agent-Driven-Multimodal-Dataset-for-Interpretable-Ophthalmic-Diagnosis.

cs.CV↗

MCoT-MVS: Multi-level Vision Selection by Multi-modal Chain-of-Thought Reasoning for Composed Image Retrieval

Composed Image Retrieval (CIR) aims to retrieve target images based on a reference image and modified texts. However, existing methods often struggle to extract the correct semantic cues from the reference image that best reflect the user's intent under textual modification prompts, resulting in interference from irrelevant visual noise. In this paper, we propose a novel Multi-level Vision Selection by Multi-modal Chain-of-Thought Reasoning (MCoT-MVS) for CIR, integrating attention-aware multi-level vision features guided by reasoning cues from a multi-modal large language model (MLLM). Specifically, we leverage an MLLM to perform chain-of-thought reasoning on the multimodal composed input, generating the retained, removed, and target-inferred texts. These textual cues subsequently guide two reference visual attention selection modules to selectively extract discriminative patch-level and instance-level semantics from the reference image. Finally, to effectively fuse these multi-granular visual cues with the modified text and the imagined target description, we design a weighted hierarchical combination module to align the composed query with target images in a unified embedding space. Extensive experiments on two CIR benchmarks, namely CIRR and FashionIQ, demonstrate that our approach consistently outperforms existing methods and achieves new state-of-the-art performance. Code and trained models are publicly released.

cs.CV↗

A Universal Model Combining Differential Equations and Neural Networks for Ball Trajectory Prediction

This paper presents a data driven universal ball trajectory prediction method integrated with physics equations. Existing methods are designed for specific ball types and struggle to generalize. This challenge arises from three key factors. First, learning-based models require large datasets but suffer from accuracy drops in unseen scenarios. Second, physics-based models rely on complex formulas and detailed inputs, yet accurately obtaining ball states, such as spin, is often impractical. Third, integrating physical principles with neural networks to achieve high accuracy, fast inference, and strong generalization remains difficult. To address these issues, we propose an innovative approach that incorporates physics-based equations and neural networks. We first derive three generalized physical formulas. Then, using a neural network and observed trajectory points, we infer certain parameters while fitting the remaining ones. These formulas enable precise trajectory prediction with minimal training data: only a few dozen samples. Extensive experiments demonstrate our method superiority in generalization, real-time performance, and accuracy.

cs.LG↗