NeUQI improves low-bit uniform quantization of LLMs by relaxing the integer zero-point constraint and efficiently searching a near-optimal scale, beating existing PTQ baselines at 2-4 bits.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
NeUQI: Near-Optimal Uniform Quantization Parameter Initialization for Low-Bit LLMs
NeUQI improves low-bit uniform quantization of LLMs by relaxing the integer zero-point constraint and efficiently searching a near-optimal scale, beating existing PTQ baselines at 2-4 bits.