The DOI matches its Crossref record.
PackQViT: Faster Sub-8-bit Vision Transformers via Full and Packed Quantization on the Mobile
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
PackQViT introduces an outlier-aware training approach, differing from the method in [43] which calculates the power-of-two factor during calibration.
“PackQViT [69] introduces an outlier-aware training approach, differing from the method in [43] which calculates the power-of-two factor during calibration.”
From Model Quantization and Hardware Acceleration for Vision Transformers: A Comprehensive Survey (2024) · cites this paper as [69]
PackQViT substitutes the natural constant e with 2 in the softmax equation, resulting in no observable loss in accuracy after retraining.
“PackQViT [69] eschews the second-order polynomial approximation in favor of a more straightforward approach that substitutes the natural constant e with 2 in the softmax equation. This simplification results in no observable loss in accuracy after retraining.”
From Model Quantization and Hardware Acceleration for Vision Transformers: A Comprehensive Survey (2024) · cites this paper as [69]
Citations
Cites
Still being checked.
Cited by
Sorted by citations
Deployment-Aware Compression of Large Language Models for Edge Intelligence: A System-Level Survey
Dritsas, E., & Trigka, M. (2026) · IEEE Transactions on Artificial Intelligence · doi:10.1109/tai.2026.3656155 · Cited by 10
Model Quantization and Hardware Acceleration for Vision Transformers: A Comprehensive Survey
Du, D., Gong, G., & Chu, X. (2024) · arXiv · doi:10.48550/arxiv.2405.00314 · Cited by 4
“4) Outlier-Aware Training : PackQViT [69] introduces an outlier-aware training approach, differing from the method in [43] which calculates the power-of-two factor during calibration.”
MethodRefining Datapath for Microscaling ViTs
Xiao, C., Cheng, J., & Zhao, Y. (2025) · International Conference on Field Programmable Logic and Applications (FPL) · doi:10.1109/fpl68686.2025.00044 · Cited by 2
DIMQ: Distribution-Induced Multi-Peak Quantization for Deep Neural Networks
Feng, C., Zhou, G., Lin, H. H., Lin, Y., Wu, X., Zhao, Q., & Xie, S. (2026) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2026.3707561
Single domain generalized polyp detection in colonoscopy scene utilizing vision foundation models
Gan, T., Cai, Y., & Ye, X. (2026) · Medical Image Analysis · doi:10.1016/j.media.2026.104268
Wan, R., Pu, H., Liang, Y., Xu, J., & Wang, D. (2026) · The Visual Computer · doi:10.1007/s00371-026-04733-2
Xia, R., Liu, G., Li, Q., Li, G., Zhao, Z., Peng, X., Hong, C., Zhang, H., Guo, X., & Liu, J. (2026) · ACM Transactions on Architecture and Code Optimization · doi:10.1145/3841638
Selective Ternary and Multi-Precision Post-Training Quantization for Vision Transformers
George, S. B., Kumar, S. N., & Krishnan, M. S. (2026) · International Conference on Smart Communication and Sustainable Technologies (ICSCST) · doi:10.1109/icscst70031.2026.11709677
Metadata
- Authors
- Dong, P., Lu, L., Wu, C., Lyu, C., Yuan, G., Tang, H., & Wang, Y.
- Venue
- Advances in Neural Information Processing Systems 36
- Published
- 2023
- Pages
- 9015-9028
- ISSN
- 1049-5258 (ISSN-L)
- OpenAlex
- W7133228638
- Cited by
- 7
- References
- 0
- Citations by year
- 2026: 6 · 2025: 1
- FWCI
- 0.75
- Type
- conference paper
- Language
- en
- Open access
- closed
- Locations
- neural information processing systems (publishedVersion)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- Physical Sciences › Computer Science › Artificial Intelligence › Domain Adaptation and Few-Shot Learning
- Physical Sciences › Engineering › Electrical and Electronic Engineering › Low-power high-performance VLSI design
- Keywords
- vision transformer, mobile deployment, low-bit precision, model compression, neural network quantization
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
