arxiv:2607.07534
Qingyan PRO
QingyanBai
AI & ML interests
Generative Models, UMMs, and Agents.
Recent Activity
upvoted a paper 3 days ago
SMELT: Scaling Laws for Compute-Matched MoE Looped Transformers upvoted a paper 3 days ago
Dream-RSI: Recursive Self-Improvement through Evolving Worlds upvoted a paper 3 days ago
DeepSeek-V4.1-Flash: Pushing the Limits of KV Cache CompressionOrganizations
