Do Language Models Need a Trainable Input Embedding Table? Fixed Minimal Token Codes at 1.7B-Class Scale
arXiv:2610.04002v1 Announce Type: new Abstract: A trainable input embedding table assigns each vocabulary item an independently adjustable vector. We investigate whether this token-specific parameterization is required for substantial language-modeling capability, or whether a shared Transformer ca…
Read the full story at arXiv cs.CL ↗
Timeline · 1 report
- 2026-10-06 04:00 · arXiv cs.CL
Do Language Models Need a Trainable Input Embedding Table? Fixed Minimal Token Codes at 1.7B-Class Scale