arXiv subjects

Zehui Wang

Publications and source records attributed to Zehui Wang.

11 recordsLinked to original sources

arXiv2026

POI Recommendation with LLM-Augmented Multi-Graph Learning and Contrastive Alignment

Point-of-interest (POI) recommendation models based on graph neural networks achieve strong performance by propagating collaborative signals over user-item interactions, yet they struggle with the cold-start problem, where items with few or no interactions are not represented. In this paper, we propose LLM-augmented Multi-Graph Contrastive Learning (LLM-MGCL), a multi-graph neural network that uses semantic and spatial information about items to extend the LightGCN backbone with two auxiliary item-item graphs: a semantic graph constructed from sentence embeddings of LLM-generated photo summaries and keywords, and a geographic graph derived from Haversine distances between business locations. Item embeddings are propagated over all three graphs in parallel, fused additively, and aligned across views through a bidirectional InfoNCE contrastive objective that connects behavioral, semantic, and spatial representations of the same items. Experiments on the Yelp Multimodal Recommendation Dataset show that LLM-MGCL outperforms classical collaborative filtering, matrix factorization, and interaction-only graph neural network baselines. It improves Recall@20 by 52.0% and NDCG@20 by 64.8% over LightGCN while performing on par with the strongest contrastive baseline, Self-supervised Graph Learning (SGL), which is also affected by the cold-start problem. An ablation study reveals that the cross-view contrastive alignment (CA) is the primary driver of these gains, with the best performance achieved when all three graphs are combined. Our results suggest that externally grounded, LLM-derived item knowledge can effectively compensate for missing collaborative signal and mitigate the item cold-start problem in POI recommendation.

cs.IR↗

arXiv2026

GLM-5V-Turbo: Toward a Native Foundation Model for Multimodal Agents

We present GLM-5V-Turbo, a step toward native foundation models for multimodal agents. As foundation models are increasingly deployed in real environments, agentic capability depends not only on language reasoning, but also on the ability to perceive, interpret, and act over heterogeneous contexts such as images, videos, webpages, documents, GUIs. GLM-5V-Turbo is built around this objective: multimodal perception is integrated as a core component of reasoning, planning, tool use, and execution, rather than as an auxiliary interface to a language model. This report summarizes the main improvements behind GLM-5V-Turbo across model design, multimodal training, reinforcement learning, toolchain expansion, and integration with agent frameworks. These developments lead to strong performance in multimodal coding, visual tool use, and framework-based agentic tasks, while preserving competitive text-only coding capability. More importantly, our development process offers practical insights for building multimodal agents, highlighting the central role of multimodal perception, hierarchical optimization, and reliable end-to-end verification.

cs.CV↗

arXiv2026

APEX: Learning Adaptive Priorities for Multi-Objective Alignment in Vision-Language Generation

Multi-objective alignment for text-to-image generation is commonly implemented via static linear scalarization, but fixed weights often fail under heterogeneous rewards, leading to optimization imbalance where models overfit high-variance, high-responsiveness objectives (e.g., OCR) while under-optimizing perceptual goals. We identify two mechanistic causes: variance hijacking, where reward dispersion induces implicit reweighting that dominates the normalized training signal, and gradient conflicts, where competing objectives produce opposing update directions and trigger seesaw-like oscillations. We propose APEX (Adaptive Priority-based Efficient X-objective Alignment), which stabilizes heterogeneous rewards with Dual-Stage Adaptive Normalization and dynamically schedules objectives via P^3 Adaptive Priorities that combine learning potential, conflict penalty, and progress need. On Stable Diffusion 3.5, APEX achieves improved Pareto trade-offs across four heterogeneous objectives, with balanced gains of +1.31 PickScore, +0.35 DeQA, and +0.53 Aesthetics while maintaining competitive OCR accuracy, mitigating the instability of multi-objective alignment.

cs.CV↗

arXiv2025

MM-R5: MultiModal Reasoning-Enhanced ReRanker via Reinforcement Learning for Document Retrieval

Multimodal document retrieval systems enable information access across text, images, and layouts, benefiting various domains like document-based question answering, report analysis, and interactive content summarization. Rerankers improve retrieval precision by reordering retrieved candidates. However, current multimodal reranking methods remain underexplored, with significant room for improvement in both training strategies and overall effectiveness. Moreover, the lack of explicit reasoning makes it difficult to analyze and optimize these methods further. In this paper, We propose MM-R5, a MultiModal Reasoning-Enhanced ReRanker via Reinforcement Learning for Document Retrieval, aiming to provide a more effective and reliable solution for multimodal reranking tasks. MM-R5 is trained in two stages: supervised fine-tuning (SFT) and reinforcement learning (RL). In the SFT stage, we focus on improving instruction-following and guiding the model to generate complete and high-quality reasoning chains. To support this, we introduce a novel data construction strategy that produces rich, high-quality reasoning data. In the RL stage, we design a task-specific reward framework, including a reranking reward tailored for multimodal candidates and a composite template-based reward to further refine reasoning quality. We conduct extensive experiments on MMDocIR, a challenging public benchmark spanning multiple domains. MM-R5 achieves state-of-the-art performance on most metrics and delivers comparable results to much larger models on the remaining ones. Moreover, compared to the best retrieval-only method, MM-R5 improves recall@1 by over 4%. These results validate the effectiveness of our reasoning-enhanced training pipeline. Our code is available at https://github.com/i2vec/MM-R5 .

cs.AI↗

arXiv2025

A Multi-Granularity Retrieval Framework for Visually-Rich Documents

Retrieval-augmented generation (RAG) systems have predominantly focused on text-based retrieval, limiting their effectiveness in handling visually-rich documents that encompass text, images, tables, and charts. To bridge this gap, we propose a unified multi-granularity multimodal retrieval framework tailored for two benchmark tasks: MMDocIR and M2KR. Our approach integrates hierarchical encoding strategies, modality-aware retrieval mechanisms, and vision-language model (VLM)-based candidate filtering to effectively capture and utilize the complex interdependencies between textual and visual modalities. By leveraging off-the-shelf vision-language models and implementing a training-free hybrid retrieval strategy, our framework demonstrates robust performance without the need for task-specific fine-tuning. Experimental evaluations reveal that incorporating layout-aware search and VLM-based candidate verification significantly enhances retrieval accuracy, achieving a top performance score of 65.56. This work underscores the potential of scalable and reproducible solutions in advancing multimodal document retrieval systems.

cs.IR↗

arXiv2023

A Survey on Point-of-Interest Recommendations Leveraging Heterogeneous Data

Tourism is an important application domain for recommender systems. In this domain, recommender systems are for example tasked with providing personalized recommendations for transportation, accommodation, points-of-interest (POIs), etc. Among these tasks, in particular the problem of recommending POIs that are of likely interest to individual tourists has gained growing attention in recent years. Providing POI recommendations to tourists can however be especially challenging due to the variability of the user's context. With the rapid development of the Web and today's multitude of online services, vast amounts of data from various sources have become available, and these heterogeneous data represent a huge potential to better address the challenges of POI recommendation problems. In this work, we provide a survey of published research on the problem of POI recommendation between 2021 and 2023. The literature was surveyed to identify the information types, techniques and evaluation methods employed. Based on the analysis, it was observed that the current research tends to focus on a relatively narrow range of information types and there is a significant potential in improving POI recommendation by leveraging heterogeneous data. As the first information-centric survey on POI recommendation research, this study serves as a reference for researchers aiming to develop increasingly accurate, personalized and context-aware POI recommender systems.

cs.IR↗

arXiv2023

Practical Policy Optimization with Personalized Experimentation

Many organizations measure treatment effects via an experimentation platform to evaluate the casual effect of product variations prior to full-scale deployment. However, standard experimentation platforms do not perform optimally for end user populations that exhibit heterogeneous treatment effects (HTEs). Here we present a personalized experimentation framework, Personalized Experiments (PEX), which optimizes treatment group assignment at the user level via HTE modeling and sequential decision policy optimization to optimize multiple short-term and long-term outcomes simultaneously. We describe an end-to-end workflow that has proven to be successful in practice and can be readily implemented using open-source software.

cs.LG↗

arXiv2022

Looper: An end-to-end ML platform for product decisions

Modern software systems and products increasingly rely on machine learning models to make data-driven decisions based on interactions with users, infrastructure and other systems. For broader adoption, this practice must (i) accommodate product engineers without ML backgrounds, (ii) support finegrain product-metric evaluation and (iii) optimize for product goals. To address shortcomings of prior platforms, we introduce general principles for and the architecture of an ML platform, Looper, with simple APIs for decision-making and feedback collection. Looper covers the end-to-end ML lifecycle from collecting training data and model training to deployment and inference, and extends support to personalization, causal evaluation with heterogenous treatment effects, and Bayesian tuning for product goals. During the 2021 production deployment Looper simultaneously hosted 440-1,000 ML models that made 4-6 million real-time decisions per second. We sum up experiences of platform adopters and describe their learning curve.

cs.LG↗

arXiv2022

On the initiation of fiber fuse damage in high-power ytterbium-doped fiber lasers

Fiber fuse effect can occur spontaneously and propagate along optical fibers to cause wide-spread damage; it threatens all applications involving optical fibers. This paper presents two results. First, it establishes that the initiation of fiber fuse (IFF) in silica fibers is caused by defect-induced absorption. Critical temperatures and critical optical powers for IFF are simulated for the first time using a 3D solid-state heat transfer model with heat source generated by defect-induced absorption. In this method, formation energies of the defects can be uniquely determined, which offers critical information on the chemical reasons for fiber fuse. Second, this paper offers a method to evaluate operating temperatures of fiber lasers. General analytical solutions of the operating temperatures along gain fibers are deduced. Results of 976-nm laser-diode-(LD)-pumped and 1018-nm tandem-pumped ytterbium-doped fiber (YDF) amplifiers using 10/130-μm YDFs are calculated. Potential limits caused by fiber fuse are discussed.

physics.optics↗

arXiv2021

Personalization for Web-based Services using Offline Reinforcement Learning

Large-scale Web-based services present opportunities for improving UI policies based on observed user interactions. We address challenges of learning such policies through model-free offline Reinforcement Learning (RL) with off-policy training. Deployed in a production system for user authentication in a major social network, it significantly improves long-term objectives. We articulate practical challenges, compare several ML techniques, provide insights on training and evaluation of RL models, and discuss generalizations.

cs.LG↗

arXiv2020

Predictive Precompute with Recurrent Neural Networks

In both mobile and web applications, speeding up user interface response times can often lead to significant improvements in user engagement. A common technique to improve responsiveness is to precompute data ahead of time for specific activities. However, simply precomputing data for all user and activity combinations is prohibitive at scale due to both network constraints and server-side computational costs. It is therefore important to accurately predict per-user application usage in order to minimize wasted precomputation ("predictive precompute"). In this paper, we describe the novel application of recurrent neural networks (RNNs) for predictive precompute. We compare their performance with traditional machine learning models, and share findings from their large-scale production use at Facebook. We demonstrate that RNN models improve prediction accuracy, eliminate most feature engineering steps, and reduce the computational cost of serving predictions by an order of magnitude.

cs.LG↗