Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning
IntraQ: Learning Synthetic Images With Intra-Class Heterogeneity for Zero-Shot Network Quantization
Yunshan Zhong, Mingbao Lin, Gongrui Nan, Jianzhuang Liu, Baochang Zhang, Yonghong Tian, Rongrong Ji
Check 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 changed | Where | Measured | Effect |
|---|---|---|---|
| Use IntraQ rather than DSG+IL | MobileNetV1 · ImageNet W4A4 | top-1 accuracy (%): 51.36% vs. 42.19%; 9.17% accuracy improvements | Improved |
| Use IntraQ rather than GZNQ | ResNet-18 · ImageNet W4A4 | top-1 accuracy (%): 66.47% vs. 64.50%; 1.97% accuracy increases | Improved |
| Add local object reinforcement to ZeroQ+IL (ablation) | 4-bit ResNet-18 · ImageNet | top-1 accuracy (%): 63.38% to 66.14% | Improved |
| Add marginal distance constraint to ZeroQ+IL (ablation) | 4-bit ResNet-18 · ImageNet | top-1 accuracy (%): 63.38% to 63.77% | Improved |
| Add soft inception loss to ZeroQ+IL (ablation) | 4-bit ResNet-18 · ImageNet | top-1 accuracy (%): 63.38% to 63.60% | Improved |
| Add marginal distance constraint and soft inception loss to ZeroQ+IL (ablation) | 4-bit ResNet-18 · ImageNet | top-1 accuracy (%): 63.38% to 64.05% | Improved |
| Add local object reinforcement and marginal distance constraint to ZeroQ+IL (ablation) | 4-bit ResNet-18 · ImageNet | top-1 accuracy (%): 63.38% to 66.32% | Improved |
| Add local object reinforcement and soft inception loss to ZeroQ+IL (ablation) | 4-bit ResNet-18 · ImageNet | top-1 accuracy (%): 63.38% to 66.30% | Improved |
| Add local object reinforcement, marginal distance constraint, and soft inception loss to ZeroQ+IL (ablation) | 4-bit ResNet-18 · ImageNet | top-1 accuracy (%): 63.38% to 66.47% | Improved |
| Add flip preprocessing to local object reinforcement (ablation) | IntraQ on 4-bit ResNet-18 · ImageNet | top-1 accuracy (%): 66.47 -> 66.34 | Hurt |
| Add rotation preprocessing to local object reinforcement (ablation) | IntraQ on 4-bit ResNet-18 · ImageNet | top-1 accuracy (%): 66.47 -> 66.31 | Hurt |
| Add rotation and flip preprocessing to local object reinforcement (ablation) | IntraQ on 4-bit ResNet-18 · ImageNet | top-1 accuracy (%): 66.47 -> 66.42 | Hurt |
Later papers that compared with this method
1 later accepted paper reports IntraQ as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- Relation-Guided Adversarial Learning for Data-Free Knowledge Transfer (IJCV 2024 · better on 1 of 1 shared result)
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
- ZeroQ: A Novel Zero Shot Quantization Framework (CVPR 2020 · similarity 0.89)
- SynQ: Accurate Zero-shot Quantization by Synthesis-aware Fine-tuning (ICLR 2025 · similarity 0.89)
- Sharpness-Aware Data Generation for Zero-shot Quantization (ICML 2024 · similarity 0.89)
- Hard Sample Matters a Lot in Zero-Shot Quantization (CVPR 2023 · similarity 0.88)
- It's All in the Teacher: Zero-Shot Quantization Brought Closer to the Teacher (CVPR 2022 · similarity 0.88)
- Zero-Shot Quantization via Weight-Space Arithmetic (NeurIPS 2026 · similarity 0.86)
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.
