SFT Accept/Reject paper-review models: {3B,7B} x {text,vision} x {arxiv,ICLR}. Cards have prompt format + per-model metrics.
AI & ML interests
None defined yet.
Recent Activity
View all activity
Papers
SWEeper-Bench: Can Agents Discover Bugs in Interactive Software?
VisionFoundry: Teaching VLMs Visual Perception with Synthetic Images
i1: A Simple and Fully Open Recipe for Strong Text-to-Image Models
-
Vero: An Open RL Recipe for General Visual Reasoning
Paper • 2604.04917 • Published • 31 -
zlab-princeton/Vero-Qwen3T-8B
Image-Text-to-Text • 9B • Updated • 20 • 3 -
zlab-princeton/Vero-Qwen3I-8B
Image-Text-to-Text • 9B • Updated • 37 • 2 -
zlab-princeton/Vero-Qwen25-7B
Image-Text-to-Text • 8B • Updated • 24 • 2
[ICLR 2026] SoFlow: Solution Flow Models for One-Step Generative Modeling.
SFT Accept/Reject paper-review models: {3B,7B} x {text,vision} x {arxiv,ICLR}. Cards have prompt format + per-model metrics.
i1: A Simple and Fully Open Recipe for Strong Text-to-Image Models
-
Vero: An Open RL Recipe for General Visual Reasoning
Paper • 2604.04917 • Published • 31 -
zlab-princeton/Vero-Qwen3T-8B
Image-Text-to-Text • 9B • Updated • 20 • 3 -
zlab-princeton/Vero-Qwen3I-8B
Image-Text-to-Text • 9B • Updated • 37 • 2 -
zlab-princeton/Vero-Qwen25-7B
Image-Text-to-Text • 8B • Updated • 24 • 2
[ICLR 2026] SoFlow: Solution Flow Models for One-Step Generative Modeling.
Model Checkpoints for [CVPR 2026] Stronger Normalization-Free Transformers