Is this you? As a journalist, you can create a free Muck Rack account to customize your profile, list your contact preferences, and upload a portfolio of your best work.
Claim your profile
Articles by Xiaobo Xia
Computer Science > Computer Vision and Pattern Recognition
Multimodal classification processes and relates information from multiple modalities to achieve more accurate predictions. However, existing methods typically rely on high-quality ground-truth labels, which are difficult to obtain in real-world scenarios. While sample-selection methods for learning with noisy labels aim to identify correctly labeled examples from noisy data, traditional methods primarily focus on unimodal settings and fail to exploit multimodal information fully.
Computer Science > Machine Learning
Benign overfitting is extensively studied when learning from a single high-dimensional input, but its behavior under heterogeneous input fusion remains largely unexplored. We study this question for minimum-norm linear interpolation under a heterogeneous Gaussian design, comparing two statistically dependent input blocks with their fusion while holding the underlying population task fixed.
Computer Science > Computation and Language
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
FreeAct: Freeing Activations for LLM Quantization
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Erythropoiesis–inosine metabolic axis failure underlying retinal neurodegeneration in glaucoma: novel diagnoses and therapies - Experimental & Molecular Medicine
Abstract Glaucoma, long considered an ocular-limited, age-dependent and hypoxia-driven neurodegeneration, is here reframed as a systemic erythroid–inosine axis failure that originates in the bone marrow yet culminates in retinal ganglion cell (RGC) death.
SPD-Faith Bench: Diagnosing and Improving Faithfulness in Chain-of-Thought for Multimodal Large Language Models
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
[2601.14888] What Makes Low-Bit Quantization-Aware Training Work for Reasoning LLMs? A Systematic Study
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
[2511.12034] Calibrated Multimodal Representation Learning with Missing Modalities
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
[2507.17343] Principled Multimodal Representation Learning
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
L-MTP: Leap Multi-Token Prediction Beyond Adjacent Context for Large Language Models
Work on one of the world's most important websites and make an impact on open science.
[2503.14963] Continual Multimodal Contrastive Learning
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Overexpression of plant chitin receptors in wheat confers broad‐spectrum resistance to fungal diseases
CONFLICT OF INTEREST The authors declare no conflict of interest. Supporting Information Filename Description tpj17035-sup-0001-FiguresS1-S9.pdfPDF document, 10.5 MB Figure S1. Phylogenetic analysis of CERK1 protein in plants. Figure S2. Amino acid alignment of AtCERK1 and its homologs. Figure S3. Molecular docking of CERK1 homologs with chitin. Figure S4. Molecular docking of CERK1 and CEBiP of grass homologs with chitin. Figure S5.
[2403.14774] Few-Shot Adversarial Prompt Learning on Vision-Language Models
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
IDEAL: Influence-Driven Selective Annotations Empower In-Context Learners in Large Language Models
[Submitted on 16 Oct 2023 ( v1 ), last revised 20 Jan 2024 (this version, v2)] Title:IDEAL: Influence-Driven Selective Annotations Empower In-Context Learners in Large Language Models Download a PDF of the paper titled IDEAL: Influence-Driven Selective Annotations Empower In-Context Learners in Large Language Models, by Shaokun Zhang and 6 other authors Download PDF HTML (experimental) Abstract:In-context learning is a promising paradigm that utilizes in-context examples as prompts for the...
Investigating the Resistance Mechanism of Wheat Varieties to Fusarium Head Blight Using Comparative Metabolomics
This is an early access version, the complete PDF, HTML, and XML versions will be available soon.
Phenotypic screen identifies calcineurin-sparing FK506 analogs as BMP potentiators for treatment of acute kidney injury
Highlights • Phenotypic screen identifies FK506 as BMP potentiator sequestering FKBP12 from BMPRI • Calcineurin inhibition is not required for BMP potentiation • Calcineurin-sparing FK506 analogs identified via late-stage functionalization • OxtFK preserves renal functions in vitro and in vivo with high safety margins Summary Acute kidney injury (AKI) is a life-threatening disease with no known curative or preventive therapies.
Cross-talk between MYOC p. Y437H mutation and TGF-β2 in the pathology of glaucoma
Int J Med Sci 2020; 17(8):1062-1070. doi:10.7150/ijms.43614 Research Paper Yang Yang1,2, AL Sabri Waled Abdulghani Abdulatef1, LuSi Zhang3, Haibo Jiang1, Zhou Zeng1, Haibo Li1,4 , Xiaobo Xia1,4 1. Eye center of Xiangya Hospital, Central South University, Changsha 410008, Hunan Province, China 2. Department of Ophthalmology, the First People's Hospital of Yueyang, Yueyang, 414000, Hunan Province, China 3. The School of Life Sciences, Central South University, Changsha 410078, Hunan Province, China 4.
Angle stability and outflow in dual blade ab interno trabeculectomy with active versus passive chamber management
To compare intraoperative angle stability and postoperative outflow of two ab interno trabeculectomy devices that excise the trabecular meshwork with or without active aspiration and irrigation. We hypothesized that anterior segment optical coherence tomography (AS-OCT) allows for a quantitative comparison of intraoperative angle stability in a microincisional glaucoma surgery (MIGS) pig eye training model.
Show More
loading
Actions
Is this you?
As a journalist, you can create a free Muck Rack account to customize your profile, list your contact preferences, and upload a portfolio of your best work.Get in touch with Xiaobo
Contact Xiaobo, search articles and posts on X, monitor coverage, and track replies from one place.
Learn more about Muck Rack