Xin Wen

xwen99

·

https://wen-xin.info

AI & ML interests

self-supervised learning, object discovery

Recent Activity

upvoted a paper 4 days ago

Towards Physics of Multimodal Pretraining: Knowledge Flow, Modality Synergy, Early Unification, and Recipes

upvoted a paper 3 months ago

SpecBench: Measuring Reward Hacking in Long-Horizon Coding Agents

upvoted a paper 3 months ago

LongLive-2.0: An NVFP4 Parallel Infrastructure for Long Video Generation

View all activity

Organizations

liked a dataset 8 months ago

Ritvik19/Sudoku-Dataset

Viewer • Updated Mar 22, 2024 • 17M • 2.28k • 13

liked a Space 11 months ago

FineVision: Open Data is All You Need

A new open-source dataset for training VLMs

liked a Space over 3 years ago

BLIP2

image captioning, VQA

liked 3 models over 3 years ago

Salesforce/blip2-flan-t5-xxl

Image-Text-to-Text • 12B • Updated Feb 3, 2025 • 704 • 94

google/flan-t5-xl

3B • Updated Nov 28, 2023 • 152k • 535

facebook/opt-iml-1.3b

Text Generation • Updated Jan 26, 2023 • 1.74k • 28

liked a Space over 3 years ago

BLIP2

Generate captions and answer questions from an image