Eric Xing — Multimodal Retrieval & Vision-Language Models

Eric Xing

I'm a 4th year PhD student in the Multimodal Vision Research Lab at WashU advised by Nathan Jacobs. I've generally been working on multimodal learning and search. I did my BS in Computer Science with a minor in Math at Western Kentucky University.

Email / CV / LinkedIn / Scholar / GitHub

Research

Lately, my research has spanned vision-language models and multimodal learning. I am currently interested in efficient search paradigms, multimodal reranking, and hierarchical representation learning.

Publications

	Tessellating The Earth Dan Cher, Hamza Iqbal, Eric Xing, Brian Wei, Nathan Jacobs European Conference on Computer Vision (ECCV'26) arXiv / code / website / bibtex A geolocation encoder that learns representations by tessellating the Earth's sphere using Voronoi partitioning.
	TerraDiT-Ω: Unified Spatial Control for Satellite Image Synthesis with Any Geospatial Primitive Brian Wei, Srikumar Sastry, Dan Cher, Eric Xing, Nathan Jacobs European Conference on Computer Vision (ECCV'26) arXiv / code / website / bibtex A diffusion transformer for satellite imagery that is promptable with any geospatial primitive (polygons, polylines, bounding boxes, points) grounded in text.
	QuARI: Query Adaptive Retrieval Improvement Eric Xing, Abby Stylianou, Robert Pless, Nathan Jacobs Advances in Neural Information Processing Systems (NeurIPS'25) arXiv / code / website / bibtex Hypernetwork-based framework for dynamic database feature adaptation, for retrieval and large-scale reranking.
	Global and Local Entailment Learning for Natural World Imagery Srikumar Sastry, Aayush Dhakal, Eric Xing, Subash Khanal, Nathan Jacobs 2025 IEEE/CVF International Conference on Computer Vision (ICCV'25) arXiv / website / bibtex We introduce a framework that explicitly enforces the modeling of transitivity-enforced entailment for the partial order of concepts within vision-language models.
	Towards Open-World Generation of Stereo Images and Unsupervised Matching Feng Qiao, Zhexiao Xiong, Eric Xing, Nathan Jacobs 2025 IEEE/CVF International Conference on Computer Vision (ICCV'25) arXiv / code / website / bibtex GenStereo uses a diffusion-based approach with disparity-aware conditioning for strong performance in stereo image generation and unsupervised stereo matching.
	ConText-CIR: Learning from Concepts in Text for Composed Image Retrieval Eric Xing, Pranavi Kolouju, Robert Pless, Abby Stylianou, Nathan Jacobs 2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR'25) arXiv / code / website / bibtex State-of-the-art composed image retrieval by bootstrapping learning from concept representations.
	RANGE: Retrieval Augmented Neural Fields for Multi-Resolution Geo-Embeddings Aayush Dhakal, Srikumar Sastry, Subash Khanal, Adeel Ahmad, Eric Xing, Nathan Jacobs 2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR'25) arXiv / code / bibtex We develop a retrieval-augmented strategy for geospatial representation learning, significantly improving performance on downstream tasks.
	good4cir: Generating Detailed Synthetic Captions for Composed Image Retrieval Pranavi Kolouju, Eric Xing, Robert Pless, Nathan Jacobs, Abby Stylianou 2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW'25) arXiv / code / bibtex Dataset synthesis pipeline for training of powerful composed image retrieval models.
	PSM: Learning Probabilistic Embeddings for Multi-scale Zero-Shot Soundscape Mapping Subash Khanal, Eric Xing, Srikumar Sastry, Aayush Dhakal, Zhexiao Xiong, Adeel Ahmad, Nathan Jacobs MM '24: Proceedings of the 32nd ACM International Conference on Multimedia (ACM MM'24) paper / arXiv / code / bibtex We develop a probabilistic, multi-scale, and metadata-aware embedding space that connects audio, text, and overhead imagery.
	ALISON: Fast and Effective Stylometric Authorship Obfuscation Eric Xing, Saranya Venkatraman, Thai Le, Dongwon Lee AAAI Conference on Artificial Intelligence (AAAI'24) paper / arXiv / code / bibtex An authorship obfuscation method that demonstrates a ~10x speedup over previous methods while outperforming them in terms of attack success, semantic preservation, and fluency.
	Neural Network Decision-Making Criteria Consistency Analysis via Inputs Sensitivity Eric Xing, Liangliang Liu, Xin Xing, Yunni Qu, Nathan Jacobs, Gongbo Liang 2022 26th International Conference on Pattern Recognition (ICPR'22) paper / bibtex Quantification and analysis of neural network decision-making criteria inconsistency and three algorithms to mitigate this inconsistency with minimal performance sacrifice.
	Beware the Black-Box of Medical Image Generation: An Uncertainty Analysis by the Learned Feature Space Yunni Qu, David Yan, Eric Xing, Fengbo Zheng, Jie Zhang, Liangliang Liu, Gongbo Liang 2022 44th Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC'22) paper / bibtex Quantitative and clustering-based analyses of learned feature spaces of U-Net architectures for medical image generation.
	A Toolkit for Assessments in Introductory Programming Courses Eric Xing, Guangming Xing Proceedings of the 54th ACM Technical Symposium on Computer Science Education V. 2 (SIGCSE'23) paper / bibtex A versatile online exam toolkit with plagiarism and cheating detection as part of the vLab learning management system.
	Motorcycle Safety Investigation in Kentucky Using Machine and Deep Learning Techniques Eric Xing, Kirolos Haleem International Conference on Transportation and Development 2022 (ICTD'22) paper / bibtex Interpretability over a new state-of-the-art pipeline for motorcycle crash severity prediction for the analysis of factors contributing to severe motorcycle crashes.

Thank you to Jon Barron for the site template.

bibtex copied