QuantWM: Temporally Consistent 2-Bit KV Cache Quantization for Video World Models
QuantWM is presented, a training-free 2-bit KV cache quantization framework for video world models that significantly improves visual quality and temporal consistency, while outperforming existing methods across benchmarks with up to 6.20 KV cache memory compression and limited additional overhead.
Jia-Qi Zhao, Xiao-Bin Hu, Bo Yin et al.
· 0 citations