MetaTT factorizes all transformer adapter weights into a single shared tensor-train, achieving LoRA-competitive accuracy with up to 30-40x fewer trainable parameters and a DMRG-inspired rank-adaptive optimizer.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
MetaTT: A Global Tensor-Train Adapter for Parameter-Efficient Fine-Tuning
MetaTT factorizes all transformer adapter weights into a single shared tensor-train, achieving LoRA-competitive accuracy with up to 30-40x fewer trainable parameters and a DMRG-inspired rank-adaptive optimizer.