Megadose AI progress, ranked and analyzed.

HyQuant: Hybrid-Precision Quantization for LLM Attention

· HF Daily Papers ·
HyQuant keeps most attention data low-bit, but protects a small set of high-impact tokens and local states in full precision.

The paper targets the error spike that low-bit quantization can cause inside LLM attention. HyQuant selects vertical-line tokens and local-window states using lightweight attention-pattern signals, then leaves those regions high precision while quantizing the rest. It applies the design in prefill and decode, including KV-cache compression with fused dequantization and attention computation. The authors report nearly lossless accuracy across tasks, models, and datasets. HF Daily Papers' note

score 4

Categories: Research