您正在关注
关注用户错误。
该用户不允许他人关注他们。
您已经关注了这个用户。
您的会员计划仅允许0个关注。升级在此。
成功取消关注
取消关注用户错误
您已成功推荐
推荐用户错误。
出问题了，请刷新页面并重新尝试。
联系Chang L.询问您的工作详情
登录并通过聊天讨论所有细节。
文件夹
文件夹
Publication of Academic Papers in AI Conferences (AAAI 2024)
Research Experience of Survey Paper in DL Research
Research Paper of A Complete Image Inpainting Project
Research Paper of A Complete Diffusion Model Project
GitHub Repositories of Up-to-Date Research Paper Lists
GitHub Repositories of Up-to-Date Research Paper Lists
GitHub Repositories of Up-to-Date Research Paper Lists
Extensive Working Experience on OCR-Related Projects
Extensive Working Experience on OCR-Related Projects
Extensive Working Experience on OCR-Related Projects
Publication of Academic Papers in AI Conferences (AAAI 2024)
Research Experience of Survey Paper in DL Research
Research Paper of A Complete Image Inpainting Project
Research Paper of A Complete Diffusion Model Project
GitHub Repositories of Up-to-Date Research Paper Lists
GitHub Repositories of Up-to-Date Research Paper Lists
GitHub Repositories of Up-to-Date Research Paper Lists
Extensive Working Experience on OCR-Related Projects
Extensive Working Experience on OCR-Related Projects
Extensive Working Experience on OCR-Related Projects
评论
变更已保存
这里没有评论！
经验
DL reseracher
9月 2021 - 现在
I am a DL researcher (ranked 57th worldwide). I have skillful experience in deep learning-based research, including computer vision, generative models, etc. I am familiar with model architecture such as Transformer, GANs, diffusion models, VAEs, which are highly related to my published papers. I can provide services including academic paper proofreading (LaTeX), Deep Learning/Computer Vision/Generative Models Project tutoring, academic presentation slide production.
教育
PhD (expected finished year 2027)
(2 年)
出版物
[Under Review] A Systematic Review of Deep Learning-based Research on Radiology Report Generation
arXiv
This paper surveys the radiology report generation task with respect to the method categories, datasets, evaluation metrics, result analyses, discussion of limitations, and futuristic development direction. The original paper is available at: https://arxiv.org/abs/2311.14199.
[Under Review] Late-Constraint Diffusion Guidance for Controllable Image Synthesis
arXiv
The core purpose of this study is to utilize a lightweight condition adapter to change the sampling process of diffusion models with external conditions, similar to ControlNet. Experiments and analyses illustrate that the proposed method demonstrate superior efficiency and proficiency when handling conditional image generation. The original paper is available at: https://arxiv.org/abs/2305.11520.
[Under Review] Towards Interactive Image Inpainting via Robust Sketch Refinement
arXiv
The core purpose of this study is to utilize user-drawn sketches to guide the image inpainting process. Since user sketches can be inaccurate, we propose an automatic sketch simulation strategy and a sketch refinement network to calibrate sketches, ensuring the inpainting process to be guided more precisely. The original paper is available at: https://arxiv.org/abs/2306.00407.
[AAAI 2024] Bootstrapping Large Language Models for Radiology Report Generation
The MIT Press
The core purpose of this study is to propose a specific paradigm for large language models (LLM) fine-tuning, so that we can bootstrap LLMs from the general domain to the medical domain. Once transferred to the medical domain, the astonishing generation ability of LLM enables itself to produce elaborated radiology reports for medical usage. The original paper is available at: https://ojs.aaai.org/index.php/AAAI/article/view/29826.
联系Chang L.询问您的工作详情
登录并通过聊天讨论所有细节。
认证
资质
浏览相似作品
邀请已发送成功！
谢谢！我们已通过电子邮件向您发送了索取免费积分的链接。
发送电子邮件时出现问题。请再试一次。
加载预览
授予地理位置权限。
您的登录会话已过期而且您已经登出，请再次登录。