Author name cluster

Xinyuan Lu

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

3 papers

1 author row

EAAI Journal 2025 Journal Article

A sample average approximation-based approach for the last mile delivery and pickup problem with load-dependent travel time under uncertainties

Hongyuan Luo
Deyun Wang
Yanhui Li
Xinyuan Lu
Mingyun Gao
Hao Chen

This paper studies an electric cargo bicycle last mile Delivery and Pickup Problem with Load-Dependent Travel Time under Uncertainties (DPLTTU), in which the travel speed depends on the gradient of the road and the load of the electric cargo bicycle. Moreover, the uncertainties in actual transportation and service processes are considered in the DPLTTU. We formulate the studied DPLTTU as a Stochastic Programming with Recourse (SPR) model. To solve this SPR model, the Sample Average Approximation (SAA)-based algorithms are proposed, where a Simulated Annealing (SA) algorithm and an Adaptive Large Neighborhood Search (ALNS) algorithm are proposed to solve the integer programming model converted by SPR model. An effective parallel computing strategy is applied into speeding up the solving processed of SAA-based algorithms. In order to validate the efficiency and effectiveness of the developed SPR model and SAA-based algorithms, a large number of numerical experiments are conducted. The experimental results highlight the performance of the developed model and algorithms, and demonstrate that the proposed model and algorithms can generate a more risk-resistant solution when uncertain elements are taken into account, which will help logistics companies make suitable decisions when addressing the issue of last mile delivery and pickup.

Details DOI

NeurIPS Conference 2024 Conference Paper

MMLONGBENCH-DOC: Benchmarking Long-context Document Understanding with Visualizations

Yubo Ma
Yuhang Zang
Liangyu Chen
Meiqi Chen
Yizhu Jiao
Xinze Li
Xinyuan Lu
Ziyu Liu

Understanding documents with rich layouts and multi-modal components is a long-standing and practical task. Recent Large Vision-Language Models (LVLMs) have made remarkable strides in various tasks, particularly in single-page document understanding (DU). However, their abilities on long-context DU remain an open problem. This work presents MMLONGBENCH-DOC, a long-context, multi- modal benchmark comprising 1, 082 expert-annotated questions. Distinct from previous datasets, it is constructed upon 135 lengthy PDF-formatted documents with an average of 47. 5 pages and 21, 214 textual tokens. Towards comprehensive evaluation, answers to these questions rely on pieces of evidence from (1) different sources (text, image, chart, table, and layout structure) and (2) various locations (i. e. , page number). Moreover, 33. 7\% of the questions are cross-page questions requiring evidence across multiple pages. 20. 6\% of the questions are designed to be unanswerable for detecting potential hallucinations. Experiments on 14 LVLMs demonstrate that long-context DU greatly challenges current models. Notably, the best-performing model, GPT-4o, achieves an F1 score of only 44. 9\%, while the second-best, GPT-4V, scores 30. 5\%. Furthermore, 12 LVLMs (all except GPT-4o and GPT-4V) even present worse performance than their LLM counterparts which are fed with lossy-parsed OCR documents. These results validate the necessity of future research toward more capable long-context LVLMs.

PDF Details DOI

IJCAI Conference 2022 Conference Paper

Deep Video Harmonization With Color Mapping Consistency

Xinyuan Lu
Shengyuan Huang
Li Niu
Wenyan Cong
Liqing Zhang

Video harmonization aims to adjust the foreground of a composite video to make it compatible with the background. So far, video harmonization has only received limited attention and there is no public dataset for video harmonization. In this work, we construct a new video harmonization dataset HYouTube by adjusting the foreground of real videos to create synthetic composite videos. Moreover, we consider the temporal consistency in video harmonization task. Unlike previous works which establish the spatial correspondence, we design a novel framework based on the assumption of color mapping consistency, which leverages the color mapping of neighboring frames to refine the current frame. Extensive experiments on our HYouTube dataset prove the effectiveness of our proposed framework. Our dataset and code are available at https: //github. com/bcmi/Video-Harmonization-Dataset-HYouTube.

PDF Details DOI