Author

H. Ogilvy

h-index 9 View on Semantic Scholar

We have 1 of 23 papers

We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.

Not the right person? Other researchers publish under this name.

Preprint Jul 2026

Variable Bit-width Quantization: Learning Per-Group Precision for"Bigger-but-Smaller"Language Models

Variable Bit-width Quantization (VBQ), a training-time method in which each contiguous group of 64 weights learns its own resolution from {1,2,4,8} bits via a Gumbel-Softmax relaxation, is introduced, trained jointly by an alternating optimization that gives the precision logits a clean, task-aligned signal.

H. Ogilvy · 0 citations