PrismQuant: Optimal Null-Space Rotations for Grouped Quantizers
Smaller activation outliers do not necessarily imply better low-bit quantization: their alignment with the quantizer matters. We introduce PrismQuant, a quantizer-aware rotation framework that aligns the leading activation eigenspace with the constant group subspace of asymmetric grouped INT4. The affine offsets represent the energy in this subspace without widening the range within the group. We formulate rotation design as a Ky Fan trace maximization and derive a closed-form solution that is provably optimal for this alignment objective. Compact Householder transformations and their compact-WY representation enable gradient-free construction and efficient application at both foldable an...
| Authors | Yanlong Chen, Yining Chen, Song Zhang et al. |
|---|---|
| Primary category | cs.AI |
| Published | 26 September 2026 |
| Last updated at source | 14 October 2026 |
Connected in the knowledge graph
- ForeverBlue/PrismQuant-Qwen3-30B-A3B-BaseModel · cited by
- ForeverBlue/PrismQuant-Llama-3.1-70BModel · cited by
- ForeverBlue/PrismQuant-Llama-3.1-8BModel · cited by
- ForeverBlue/PrismQuant-Llama-3.2-3BModel · cited by
- ForeverBlue/PrismQuant-Qwen3-BaseModel · cited by
Models and datasets referencing this paper
Silytics shows public metadata from arXiv and links to the original. Information may be incomplete or out of date. Last refreshed 17 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.