Pith. sign in

Post train- ing 4-bit quantization of convolutional networks for rapid- deployment

1 Pith paper cite this work. Polarity classification is still indexing.

1 Pith paper citing it

citation-role summary

background 1

citation-polarity summary

fields

cs.LG 1

years

2025 1

verdicts

CONDITIONAL 1

roles

background 1

polarities

unclear 1

representative citing papers

MSQ: Memory-Efficient Bit Sparsification Quantization

cs.LG · 2025-07-30 · conditional · novelty 6.0

MSQ computes and prunes least significant bits of weights directly from the full-precision parameters, cutting training memory and time for mixed-precision quantization compared with bit-splitting methods.

citing papers explorer

Showing 1 of 1 citing paper.

  • MSQ: Memory-Efficient Bit Sparsification Quantization cs.LG · 2025-07-30 · conditional · none · ref 1

    MSQ computes and prunes least significant bits of weights directly from the full-precision parameters, cutting training memory and time for mixed-precision quantization compared with bit-splitting methods.