wenhua cheng
wenhuach
AI & ML interests
Model Compression, CV
Recent Activity
new activity
4 days ago
Intel/Qwen3-235B-A22B-Instruct-2507-gguf-q2ks-mixed-AutoRound:Thinking version has been deleted?
reacted
to
their
post
with 🔥
11 days ago
🚀 SignRoundV2 for LLM quantization: PTQ-level cost, QAT-level accuracy — yes, even at 2 bits.
https://huggingface.co/papers/2512.04746
Organizations
