oliveryanzuolu/crafter-cache-81f-256px
Viewer • Updated • 99.8k
Yanzuo Lu
oliveryanzuolu
AI & ML interests
None yet
Recent Activity
updated a dataset about 2 hours ago
oliveryanzuolu/crafter-cache-81f-256px published a dataset about 2 hours ago
oliveryanzuolu/crafter-cache-81f-256px updated a bucket 4 days ago
mvp-lab/crafter-trajectories-bucketOrganizations
updated a
dataset about 2 hours ago
published a
dataset about 2 hours ago
updated a
bucket 4 days ago
published a
bucket 4 days ago
updated a
model 14 days ago
published a
model 14 days ago
updated a
dataset about 2 months ago
published a
dataset about 2 months ago
updated a
bucket about 2 months ago
published a
bucket about 2 months ago
authored a
paper 3 months ago
posted an update 3 months ago
Post
125
Excited to share RAVEN, my first PhD project. Paper, code, and models are all released.
RAVEN is for real-time autoregressive video generation. Instead of simply appending future chunks, we train the model to better remember and use its own generated history, leading to more realistic and natural long-horizon videos.
Technically, RAVEN repacks self-rollouts into interleaved clean historical endpoints and noisy denoising states, aligning training-time attention with inference-time extrapolation.
We also introduce CM-GRPO: by reformulating consistency-model sampling as a conditional Gaussian transition kernel, online RL can directly optimize the sampler transition used at inference.
Project Page: https://yanzuo.lu/raven
Paper: https://arxiv.org/abs/2605.15190
Code: https://github.com/mvp-ai-lab/RAVEN
Model: mvp-lab/RAVEN
RAVEN is for real-time autoregressive video generation. Instead of simply appending future chunks, we train the model to better remember and use its own generated history, leading to more realistic and natural long-horizon videos.
Technically, RAVEN repacks self-rollouts into interleaved clean historical endpoints and noisy denoising states, aligning training-time attention with inference-time extrapolation.
We also introduce CM-GRPO: by reformulating consistency-model sampling as a conditional Gaussian transition kernel, online RL can directly optimize the sampler transition used at inference.
Project Page: https://yanzuo.lu/raven
Paper: https://arxiv.org/abs/2605.15190
Code: https://github.com/mvp-ai-lab/RAVEN
Model: mvp-lab/RAVEN
commented a
paper 3 months ago
updated a
collection 3 months ago
upvoted a paper 3 months ago
submitted a
paper to Daily Papers 3 months ago
updated a
model 3 months ago
updated a
collection 3 months ago
published a
model 3 months ago
upvoted a paper 4 months ago