AI Paper Atlas

Research map / Architectures and efficiency / Quantization

NIPQ: Noise Proxy-Based Integrated Pseudo-Quantization

Juncheol Shin, Junhyuk So, Sein Park, Seungyeop Kang, Sungjoo Yoo, Eunhyeok Park

CVPR 2023 · Architectures and efficiency › Quantization · approach: Ultra-low-bit LLM quantization · 20 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Use NIPQ mixed precision instead of LSQMobileNet-v2 · ImageNetTop-1 accuracy (%): 70.46 -> 71.52Improved
Use NIPQ BOP-constrained mixed precision instead of LSQMobileNet-v2 · ImageNetTop-1 accuracy (%): 70.46 -> 72.26Improved
Add knowledge distillationNIPQ mixed-precision quantization · ImageNet, MobileNet-v2Top-1 accuracy (%): 71.52 -> 72.16Improved
Add truncation rather than using min-max quantizationPseudo-quantization weight training · GLUE, BERT-base, weight quantizationscore: At 3-bit, MNLI: 31.82 vs 81.88; CoLA: 0.02 vs 53.03Improved
Add stochastic rounding of bit-width during training (ablation)NIPQ · CIFAR-100, MobileNet-v2accuracyImproved
Add post-training batch-normalization statistics update (ablation)NIPQ · CIFAR-100, MobileNet-v2, 1.0 GBlopsTop-1 accuracy: 70.45 -> 72.99Improved
Add late-stage QAT finetuning (ablation)NIPQ · CIFAR-100, MobileNet-v2, 1.0 GBlopsTop-1 accuracy: 70.45 -> 73.29Improved
Use late-stage QAT finetuning rather than BN update alone (ablation)NIPQ · CIFAR-100, MobileNet-v2, 1.0 GBlopsTop-1 accuracy: 72.99 -> 73.29Improved
Use a uniform PQN distribution instead of sampling the quantization-error distribution (ablation)NIPQ · Quantization experimentsfinal qualityNo clear effect

Later papers that compared with this method

2 later accepted papers report NIPQ as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy