Superseded baseline#42 of 80 most-superseded
N2UQ
LLM quantization
superseded — cited as a baseline and beaten by newer methods
1 papers critique it · 1 beat it on benchmarks
What papers say
Verbatim critique sentences, each from a paper that cites N2UQ as a baseline.
“even with the most powerful non-uniform quantization method N2UQ, the quantized model still suffers a 3.8% accuracy loss”
— GABFusion: Rethinking Feature Fusion for Low-Bit Quantization of Multi-Task Networks
Beaten on benchmarks
Head-to-head results where a newer method reports beating N2UQ. 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 N2UQ · mAP [YOLOv5s W4A4]
84.2 vs 82.1
- GABFusion: Rethinking Feature Fusion for Low-Bit Quantization of Multi-Task Networks
+ Ours beats N2UQ · mAP [YOLOv5s W3A3]
78.0 vs 75.5
- GABFusion: Rethinking Feature Fusion for Low-Bit Quantization of Multi-Task Networks
+ Ours beats N2UQ · mAP [YOLOv11s W4A4]
87.6 vs 86.2
- GABFusion: Rethinking Feature Fusion for Low-Bit Quantization of Multi-Task Networks
+ Ours beats N2UQ · mAP [YOLOv11s W3A3]
84.3 vs 83.0
Newer alternatives
Recent methods in the same sub-problem, not yet superseded in the knowledge base.
- Nov 8, 2025
- Sep 19, 2025