SignRoundV2: Closing the Performance Gap in Extremely Low-Bit Post-Training Quantization for LLMs
Paper • 2512.04746 • Published • 14
wenhua cheng
wenhuach
AI & ML interests
Model Compression, CV
Recent Activity
new activity 5 days ago
Intel/GLM-5-int4-mixed-AutoRound:This model always predicts some few nonsense sequences liked a model 10 days ago
Intel/Step-3.5-Flash-int4-mixed-AutoRound new activity 11 days ago
Intel/Qwen3.5-122B-A10B-int4-AutoRound:Does the A100 work?Organizations
authored a
paper 4 months ago