arXiv subjects

Puzhen Wu

Publications and source records attributed to Puzhen Wu.

13 recordsLinked to original sources

arXiv2026

Agents as Knowledge Integrator and Utilizer in Multimodal Recommendation

Online platforms increasingly rely on multimodal recommender systems to rank products, media, and other Web content. Existing methods usually inject visual and textual features into item representations or build homogeneous graphs from modality-level similarity, but the resulting signals can remain misaligned with the recommendation objective. We study this semantic gap from a knowledge-integration perspective: multimodal content should be interpreted together with user behavior before it is used to construct recommendation graphs or adjust rankings. We propose AgentMMRec, an agent-based multimodal recommendation framework with two coordinated roles. The Integrator Agent infers behavior- and multimodal-aware user preferences and item properties from training interactions and item content, then stores them in a reusable knowledge memory. The Utilizer Agent consumes this memory to refine modality-specific item-item graphs, construct behavior-aware homogeneous graphs, and rerank candidate lists under a frozen evaluation-time memory. This design differs from direct LLM feature augmentation and pure LLM reranking because the generated knowledge is first converted into graph structure and model representations before recommendation. Experiments on three Amazon multimodal recommendation datasets show that AgentMMRec consistently improves Recall and NDCG over recent multimodal baselines, remains effective under sparsity and item cold-start settings, and can transfer its constructed knowledge to existing backbones.

cs.IR↗

arXiv2026

DocOCR-Eval: A Correction-Based Framework for OCR Tool Selection Without Ground Truth

Document parsing is a foundational step for document understanding tasks such as visual question answering and key information extraction, as it transforms unstructured scanned images into structured representations by extracting textual, visual, and layout information. While numerous Optical Character Recognition (OCR) engines and multimodal large language models (MLLMs) have been developed for this purpose, selecting an appropriate document parsing solution for a given document collection remains challenging, particularly in label-scarce settings. In this work, we conduct a systematic evaluation of text recognition performance across a diverse set of OCR engines and state-of-the-art MLLMs on multiple scanned document benchmarks spanning different domains and languages. Motivated by the limited contextual reasoning capabilities of many OCR engines and the high cost of manual annotations, we propose DocOCR-Eval, an annotation-free evaluation framework for automatic OCR assessment and selection. DocOCR-Eval employs a three-staged correction and ranking strategy to approximate annotation-based tool ordering without ground-truth labels. We show that aggregating across multiple MLLMs progressively improves alignment with annotation-based rankings. Extensive experiments further demonstrate that reliable OCR tool selection can be achieved in realistic, label-limited settings, providing practical guidance for deploying document parsing systems across diverse real-world document collections.

cs.LG↗

arXiv2026

CLIP-Guided Unsupervised Semantic-Aware Exposure Correction

Improper exposure often leads to severe loss of details, color distortion, and reduced contrast. Exposure correction still faces two critical challenges: (1) the ignorance of object-wise regional semantic information causes the color shift artifacts; (2) real-world exposure images generally have no ground-truth labels, and its labeling entails massive manual editing. To tackle the challenges, we propose a new unsupervised semantic-aware exposure correction network. It contains an adaptive semantic-aware fusion module, which effectively fuses the semantic information extracted from a pre-trained Fast Segment Anything Model into a shared image feature space. Then the fused features are used by our multi-scale residual spatial mamba group to restore the details and adjust the exposure. To avoid manual editing, we propose a pseudo-ground truth generator guided by CLIP, which is fine-tuned to automatically identify exposure situations and instruct the tailored corrections. Also, we leverage the rich priors from the FastSAM and CLIP to develop a semantic-prompt consistency loss to enforce semantic consistency and image-prompt alignment for unsupervised training. Comprehensive experimental results illustrate the effectiveness of our method in correcting real-world exposure images and outperforms state-of-the-art unsupervised methods both numerically and visually.

cs.CV↗

arXiv2026

Docs2Synth: A Synthetic Data Trained Retriever Framework for Scanned Visually Rich Documents Understanding

Document understanding (VRDU) in regulated domains is particularly challenging, since scanned documents often contain sensitive, evolving, and domain specific knowledge. This leads to two major challenges: the lack of manual annotations for model adaptation and the difficulty for pretrained models to stay up-to-date with domain-specific facts. While Multimodal Large Language Models (MLLMs) show strong zero-shot abilities, they still suffer from hallucination and limited domain grounding. In contrast, discriminative Vision-Language Pre-trained Models (VLPMs) provide reliable grounding but require costly annotations to cover new domains. We introduce Docs2Synth, a synthetic-supervision framework that enables retrieval-guided inference for private and low-resource domains. Docs2Synth automatically processes raw document collections, generates and verifies diverse QA pairs via an agent-based system, and trains a lightweight visual retriever to extract domain-relevant evidence. During inference, the retriever collaborates with an MLLM through an iterative retrieval--generation loop, reducing hallucination and improving response consistency. We further deliver Docs2Synth as an easy-to-use Python package, enabling plug-and-play deployment across diverse real-world scenarios. Experiments on multiple VRDU benchmarks show that Docs2Synth substantially enhances grounding and domain generalization without requiring human annotations.

cs.AI↗

arXiv2025

A Disease-Aware Dual-Stage Framework for Chest X-ray Report Generation

Radiology report generation from chest X-rays is an important task in artificial intelligence with the potential to greatly reduce radiologists' workload and shorten patient wait times. Despite recent advances, existing approaches often lack sufficient disease-awareness in visual representations and adequate vision-language alignment to meet the specialized requirements of medical image analysis. As a result, these models usually overlook critical pathological features on chest X-rays and struggle to generate clinically accurate reports. To address these limitations, we propose a novel dual-stage disease-aware framework for chest X-ray report generation. In Stage~1, our model learns Disease-Aware Semantic Tokens (DASTs) corresponding to specific pathology categories through cross-attention mechanisms and multi-label classification, while simultaneously aligning vision and language representations via contrastive learning. In Stage~2, we introduce a Disease-Visual Attention Fusion (DVAF) module to integrate disease-aware representations with visual features, along with a Dual-Modal Similarity Retrieval (DMSR) mechanism that combines visual and disease-specific similarities to retrieve relevant exemplars, providing contextual guidance during report generation. Extensive experiments on benchmark datasets (i.e., CheXpert Plus, IU X-ray, and MIMIC-CXR) demonstrate that our disease-aware framework achieves state-of-the-art performance in chest X-ray report generation, with significant improvements in clinical accuracy and linguistic quality.

cs.CV↗

arXiv2025

Graph-Reward-SQL: Execution-Free Reinforcement Learning for Text-to-SQL via Graph Matching and Stepwise Reward

Reinforcement learning (RL) has been widely adopted to enhance the performance of large language models (LLMs) on Text-to-SQL tasks. However, existing methods often rely on execution-based or LLM-based Bradley-Terry reward models. The former suffers from high execution latency caused by repeated database calls, whereas the latter imposes substantial GPU memory overhead, both of which significantly hinder the efficiency and scalability of RL pipelines. To this end, we propose a novel reward model framework for RL-based Text-to-SQL named Graph-Reward-SQL, which employs the GMNScore outcome reward model. We leverage SQL graph representations to provide accurate reward signals while significantly reducing time cost and GPU memory usage. Building on this foundation, we further introduce StepRTM, a stepwise reward model that provides intermediate supervision over Common Table Expression (CTE) subqueries. This encourages both functional correctness and readability of SQL. Extensive comparative and ablation experiments on standard benchmarks, including Spider and BIRD, demonstrate that our method consistently outperforms existing reward models.

cs.LG↗

arXiv2025

AMD-Mamba: A Phenotype-Aware Multi-Modal Framework for Robust AMD Prognosis

Age-related macular degeneration (AMD) is a leading cause of irreversible vision loss, making effective prognosis crucial for timely intervention. In this work, we propose AMD-Mamba, a novel multi-modal framework for AMD prognosis, and further develop a new AMD biomarker. This framework integrates color fundus images with genetic variants and socio-demographic variables. At its core, AMD-Mamba introduces an innovative metric learning strategy that leverages AMD severity scale score as prior knowledge. This strategy allows the model to learn richer feature representations by aligning learned features with clinical phenotypes, thereby improving the capability of conventional prognosis methods in capturing disease progression patterns. In addition, unlike existing models that use traditional CNN backbones and focus primarily on local information, such as the presence of drusen, AMD-Mamba applies Vision Mamba and simultaneously fuses local and long-range global information, such as vascular changes. Furthermore, we enhance prediction performance through multi-scale fusion, combining image information with clinical variables at different resolutions. We evaluate AMD-Mamba on the AREDS dataset, which includes 45,818 color fundus photographs, 52 genetic variants, and 3 socio-demographic variables from 2,741 subjects. Our experimental results demonstrate that our proposed biomarker is one of the most significant biomarkers for the progression of AMD. Notably, combining this biomarker with other existing variables yields promising improvements in detecting high-risk AMD patients at early stages. These findings highlight the potential of our multi-modal framework to facilitate more precise and proactive management of AMD.

eess.IV↗

arXiv2025

Multi-Cali Anything: Dense Feature Multi-Frame Structure-from-Motion for Large-Scale Camera Array Calibration

Calibrating large-scale camera arrays, such as those in dome-based setups, is time-intensive and typically requires dedicated captures of known patterns. While extrinsics in such arrays are fixed due to the physical setup, intrinsics often vary across sessions due to factors like lens adjustments or temperature changes. In this paper, we propose a dense-feature-driven multi-frame calibration method that refines intrinsics directly from scene data, eliminating the necessity for additional calibration captures. Our approach enhances traditional Structure-from-Motion (SfM) pipelines by introducing an extrinsics regularization term to progressively align estimated extrinsics with ground-truth values, a dense feature reprojection term to reduce keypoint errors by minimizing reprojection loss in the feature space, and an intrinsics variance term for joint optimization across multiple frames. Experiments on the Multiface dataset show that our method achieves nearly the same precision as dedicated calibration processes, and significantly enhances intrinsics and 3D reconstruction accuracy. Fully compatible with existing SfM pipelines, our method provides an efficient and practical plug-and-play solution for large-scale camera setups. Our code is publicly available at: https://github.com/YJJfish/Multi-Cali-Anything

cs.CV↗

arXiv2025

MDVT: Enhancing Multimodal Recommendation with Model-Agnostic Multimodal-Driven Virtual Triplets

The data sparsity problem significantly hinders the performance of recommender systems, as traditional models rely on limited historical interactions to learn user preferences and item properties. While incorporating multimodal information can explicitly represent these preferences and properties, existing works often use it only as side information, failing to fully leverage its potential. In this paper, we propose MDVT, a model-agnostic approach that constructs multimodal-driven virtual triplets to provide valuable supervision signals, effectively mitigating the data sparsity problem in multimodal recommendation systems. To ensure high-quality virtual triplets, we introduce three tailored warm-up threshold strategies: static, dynamic, and hybrid. The static warm-up threshold strategy exhaustively searches for the optimal number of warm-up epochs but is time-consuming and computationally intensive. The dynamic warm-up threshold strategy adjusts the warm-up period based on loss trends, improving efficiency but potentially missing optimal performance. The hybrid strategy combines both, using the dynamic strategy to find the approximate optimal number of warm-up epochs and then refining it with the static strategy in a narrow hyper-parameter space. Once the warm-up threshold is satisfied, the virtual triplets are used for joint model optimization by our enhanced pair-wise loss function without causing significant gradient skew. Extensive experiments on multiple real-world datasets demonstrate that integrating MDVT into advanced multimodal recommendation models effectively alleviates the data sparsity problem and improves recommendation performance, particularly in sparse data scenarios.

cs.IR↗

arXiv2025

EffiLearner: Enhancing Efficiency of Generated Code via Self-Optimization

Large language models (LLMs) have shown remarkable progress in code generation, but their generated code often suffers from inefficiency, resulting in longer execution times and higher memory consumption. To address this issue, we propose \textbf{EffiLearner}, a self-optimization framework that utilizes execution overhead profiles to improve the efficiency of LLM-generated code. EffiLearner first generates code using an LLM, then executes it locally to capture execution time and memory usage profiles. These profiles are fed back to the LLM, which then revises the code to reduce overhead. To evaluate the effectiveness of EffiLearner, we conduct extensive experiments on the EffiBench, HumanEval, and MBPP with 16 open-source and 6 closed-source models. Our evaluation results demonstrate that through iterative self-optimization, EffiLearner significantly enhances the efficiency of LLM-generated code. For example, the execution time (ET) of StarCoder2-15B for the EffiBench decreases from 0.93 (s) to 0.12 (s) which reduces 87.1% the execution time requirement compared with the initial code. The total memory usage (TMU) of StarCoder2-15B also decreases from 22.02 (Mb*s) to 2.03 (Mb*s), which decreases 90.8% of total memory consumption during the execution process. The source code of EffiLearner was released in https://github.com/huangd1999/EffiLearner

cs.SE↗

arXiv2025

Segregation and Context Aggregation Network for Real-time Cloud Segmentation

Cloud segmentation from intensity images is a pivotal task in atmospheric science and computer vision, aiding weather forecasting and climate analysis. Ground-based sky/cloud segmentation extracts clouds from images for further feature analysis. Existing methods struggle to balance segmentation accuracy and computational efficiency, limiting real-world deployment on edge devices, so we introduce SCANet, a novel lightweight cloud segmentation model featuring Segregation and Context Aggregation Module (SCAM), which refines rough segmentation maps into weighted sky and cloud features processed separately. SCANet achieves state-of-the-art performance while drastically reducing computational complexity. SCANet-large (4.29M) achieves comparable accuracy to state-of-the-art methods with 70.9% fewer parameters. Meanwhile, SCANet-lite (90K) delivers 1390 fps in FP16, surpassing real-time standards. Additionally, we propose an efficient pre-training strategy that enhances performance even without ImageNet pre-training.

cs.CV↗

arXiv2025

DDUNet: Dual Dynamic U-Net for Highly-Efficient Cloud Segmentation

Cloud segmentation amounts to separating cloud pixels from non-cloud pixels in an image. Current deep learning methods for cloud segmentation suffer from three issues. (a) Constrain on their receptive field due to the fixed size of the convolution kernel. (b) Lack of robustness towards different scenarios. (c) Requirement of a large number of parameters and limitations for real-time implementation. To address these issues, we propose a Dual Dynamic U-Net (DDUNet) for supervised cloud segmentation. The DDUNet adheres to a U-Net architecture and integrates two crucial modules: the dynamic multi-scale convolution (DMSC), improving merging features under different reception fields, and the dynamic weights and bias generator (DWBG) in classification layers to enhance generalization ability. More importantly, owing to the use of depth-wise convolution, the DDUNet is a lightweight network that can achieve 95.3% accuracy on the SWINySEG dataset with only 0.33M parameters, and achieve superior performance over three different configurations of the SWINySEg dataset in both accuracy and efficiency.

cs.CV↗

arXiv2025

CP2M: Clustered-Patch-Mixed Mosaic Augmentation for Aerial Image Segmentation

Remote sensing image segmentation is pivotal for earth observation, underpinning applications such as environmental monitoring and urban planning. Due to the limited annotation data available in remote sensing images, numerous studies have focused on data augmentation as a means to alleviate overfitting in deep learning networks. However, some existing data augmentation strategies rely on simple transformations that may not sufficiently enhance data diversity or model generalization capabilities. This paper proposes a novel augmentation strategy, Clustered-Patch-Mixed Mosaic (CP2M), designed to address these limitations. CP2M integrates a Mosaic augmentation phase with a clustered patch mix phase. The former stage constructs a new sample from four random samples, while the latter phase uses the connected component labeling algorithm to ensure the augmented data maintains spatial coherence and avoids introducing irrelevant semantics when pasting random patches. Our experiments on the ISPRS Potsdam dataset demonstrate that CP2M substantially mitigates overfitting, setting new benchmarks for segmentation accuracy and model robustness in remote sensing tasks.

cs.CV↗