LSQ
LSQ+: Improving low-bit quantization through learnable offsets and better initializationLLM quantization · first seen Apr 20, 2020
superseded — cited as a baseline and beaten by newer methods
0 papers critique it · 2 beat it on benchmarks
Beaten on benchmarks
Head-to-head results where a newer method reports beating LSQ. Values are copied from the source paper's tables — verify against the cited paper.
- GABFusion: Rethinking Feature Fusion for Low-Bit Quantization of Multi-Task Networks
+ Ours beats LSQ · mAP [YOLOv5s W4A4]
78.9 vs 76.9
- GABFusion: Rethinking Feature Fusion for Low-Bit Quantization of Multi-Task Networks
+ Ours beats LSQ · mAP [YOLOv5s W3A3]
66.8 vs 59.9
- GABFusion: Rethinking Feature Fusion for Low-Bit Quantization of Multi-Task Networks
+ Ours beats LSQ · mAP [YOLOv11s W4A4]
84.2 vs 82.9
- GABFusion: Rethinking Feature Fusion for Low-Bit Quantization of Multi-Task Networks
+ Ours beats LSQ · mAP [YOLOv11s W3A3]
78.9 vs 75.5
- MEC-Quant: Maximum Entropy Coding for Extremely Low Bit Quantization-Aware Training
MEC-Qaunt (Setting (A)) beats LSQ · Res18 [W/A 4/4]
90.22 vs 88.69
- MEC-Quant: Maximum Entropy Coding for Extremely Low Bit Quantization-Aware Training
MEC-Qaunt (Setting (A)) beats LSQ · Res50 [W/A 4/4]
91.25 vs 90.01
- MEC-Quant: Maximum Entropy Coding for Extremely Low Bit Quantization-Aware Training
MEC-Qaunt (Setting (A)) beats LSQ · MBV1 [W/A 4/4]
85.44 vs 82.39
- MEC-Quant: Maximum Entropy Coding for Extremely Low Bit Quantization-Aware Training
MEC-Qaunt (Setting (A)) beats LSQ · MBV2 [W/A 4/4]
85.52 vs 84.45
- MEC-Quant: Maximum Entropy Coding for Extremely Low Bit Quantization-Aware Training
MEC-Qaunt (Setting (B)) beats LSQ · Res18 [W/A 2/4]
89.98 vs 88.36
- MEC-Quant: Maximum Entropy Coding for Extremely Low Bit Quantization-Aware Training
MEC-Qaunt (Setting (B)) beats LSQ · Res50 [W/A 2/4]
91.23 vs 90.01
- MEC-Quant: Maximum Entropy Coding for Extremely Low Bit Quantization-Aware Training
MEC-Qaunt (Setting (A)) beats LSQ · MBV1 [W/A 2/4]
80.79 vs 78.15
- MEC-Quant: Maximum Entropy Coding for Extremely Low Bit Quantization-Aware Training
MEC-Qaunt (Setting (A)) beats LSQ · MBV2 [W/A 2/4]
81.20 vs 78.15
Newer alternatives
Recent methods in the same sub-problem, not yet superseded in the knowledge base.
- Nov 8, 2025
- Sep 19, 2025