Cong Wei

@CongWei1230

Interning @nvidia | CS PhD @UWaterloo | MS & B.CS @UofT | Prev Intern @AIatMeta @VectorInst

congwei1230.github.io

Joined July 2023

Replies Reposts Media

Cong Wei
@CongWei1230
Aug 28
Thanks @_akhaliq for sharing our work! How intelligent is your video generation model? Video generators are emerging as backbones for World Action Models (WAMs). VGI-Bench probes the reasoning and action-relevant priors they encode across 27 tasks and 810 instances.
00:00
AK
@_akhaliq
Aug 28
VGI-Bench Probing Visual Intelligence in Video Generation Models paper: huggingface.co/papers/2608.19…
1
Cong Wei
@CongWei1230
Aug 27
How intelligent is your video generation model? Video generators are emerging as backbones for World Action Models (WAMs). VGI-Bench probes the reasoning and action-relevant priors they encode across 27 tasks and 810 instances. Thanks for sharing our work!
00:00
DailyPapers
@HuggingPapers
Aug 27
Microsoft Research released VGI-Bench on Hugging Face A new benchmark probing visual intelligence in video generation models, with 27 tasks and 810 instances across 4 domains - even the strongest model only reaches 51.0%.
Cong Wei
@CongWei1230
Aug 27
As image generation becomes more agentic, benchmarks must evolve with it. AgentGen-Bench tests models on real-world tasks. @SpaceXAI’s Grok Imagine 2.0 (Low) ranks #2. Great work—we look forward to evaluating the latest model 🚀 huggingface.co/spaces/CongWei… #AgentGenBench
Hexiang (Frank) Hu
@hexiang
Aug 20
Imagine 2 is designed for real-world usefulness. This is it working in the hands of art professionals — assets rebuilt from screenshots, already usable. Thanks Dogan for sharing this feedback over♥️
Cong Wei
@CongWei1230
Jul 16
Thanks for sharing our work! Try our Live Demo: tinyurl.com/searchgendemo Read our project blog: haozheh3.github.io/SearchGen/
DailyPapers
@HuggingPapers
Jul 15
SearchGen Image generators fabricate what they don't know. Alibaba's Qwen team and collaborators propose SearchGen, which searches for and organizes multimodal context to ground generations, and knows when not to search.
Cong Wei
@CongWei1230
Jul 15
💡LLM agents should do more than rewrite prompts—they should actively search for and organize useful multimodal context to support the generator. Introducing SearchGen, which co-trains an LLM agent and a generator to expand the knowledge boundary of image generation. Try our
GIF
1