AI RESEARCH
Initialisation Determines the Basin: Efficient Codebook Optimisation for Extreme LLM Quantization
arXiv CS.CL
•
ArXi:2604.08118v1 Announce Type: new Additive quantization enables extreme LLM compression with O(1) lookup-table dequantization, making it attractive for edge deployment. Yet at 2-bit precision, it often fails catastrophically, even with extensive search and finetuning. We show that the dominant bottleneck is codebook initialisation. Greedy sequential initialisation frequently places the model in poor optimisation regions that subsequent beam search and PV-tuning struggle to overcome.