Viewing profile — jimdavid
jimdavid
HN member- Joined
- Thu, Oct 23, 2025, 2:46 AM UTC
- HN karma
- 1
- Public activity
- 6 items
- HN profile
- View on Hacker News ↗
About jimdavid
This is Dr. David (Dawei Leng) majored in VLM, multimodal understanding and multimodal generation, the following is my linkedin profile:
https://www.linkedin.com/in/daweileng/
Connect with me if you've same research interests.
Recent public activity
-
comment
Comment #46821307
We’re sharing an open-source project on makeup transfer built on a Diffusion Transformer (DiT) backbone. The goal is to transfer makeup from a reference face to a source face while…
- story
-
comment
Comment #45692828
[dead]
-
comment
Comment #45689872
[dead]
-
comment
Comment #45678020
Hey, pretty nice work! Are you using any CLIP-like model for image retrieval? If so, would you try FG-CLIP 2 ( https://360cvgroup.github.io/FG-CLIP ) and see how it'll improve the …
-
comment
Comment #45677838
Did anyone check the token feature dimension? If we're talking about compression, "token length" is just one of the dimensions.