Guardado en:
| Autores principales: | , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2026
|
| Materias: | |
| Acceso en línea: | https://arxiv.org/abs/2602.15563 |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866912909785825280 |
|---|---|
| author | Maskey, Sohir Eichenberg, Constantin Messner, Johannes Orr, Douglas |
| author_facet | Maskey, Sohir Eichenberg, Constantin Messner, Johannes Orr, Douglas |
| contents | Quantization-aware training (QAT) is an effective method to drastically reduce the memory footprint of LLMs while keeping performance degradation at an acceptable level. However, the optimal choice of quantization format and bit-width presents a challenge in practice. The full design space of quantization is not fully explored in the context of QAT, and the precise trade-off between quantization and downstream performance is poorly understood, as comparisons often rely solely on perplexity-based evaluations. In this work, we address these shortcomings with an empirical study of QAT in the low-bit regime. We show that k-means based weight quantization outperforms integer formats and can be implemented efficiently on standard hardware. Furthermore, we find that, under a fixed inference memory budget, the best performance on generative downstream tasks is achieved with $1$-bit quantized weights. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_15563 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | 1-Bit Wonder: Improving QAT Performance in the Low-Bit Regime through K-Means Quantization Maskey, Sohir Eichenberg, Constantin Messner, Johannes Orr, Douglas Machine Learning Quantization-aware training (QAT) is an effective method to drastically reduce the memory footprint of LLMs while keeping performance degradation at an acceptable level. However, the optimal choice of quantization format and bit-width presents a challenge in practice. The full design space of quantization is not fully explored in the context of QAT, and the precise trade-off between quantization and downstream performance is poorly understood, as comparisons often rely solely on perplexity-based evaluations. In this work, we address these shortcomings with an empirical study of QAT in the low-bit regime. We show that k-means based weight quantization outperforms integer formats and can be implemented efficiently on standard hardware. Furthermore, we find that, under a fixed inference memory budget, the best performance on generative downstream tasks is achieved with $1$-bit quantized weights. |
| title | 1-Bit Wonder: Improving QAT Performance in the Low-Bit Regime through K-Means Quantization |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2602.15563 |