TensorSLM applies per-vector tensor-train SVD to compress SLM token embeddings training-free, showing competitive task performance at roughly 2x embedding compression on Raspberry Pi with an estimated, pre-decoder energy saving.
Online embedding compression for text classification using low rank matrix factorization
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
TensorSLM: Energy-efficient Embedding Compression of Sub-billion Parameter Language Models on Low-end Devices
TensorSLM applies per-vector tensor-train SVD to compress SLM token embeddings training-free, showing competitive task performance at roughly 2x embedding compression on Raspberry Pi with an estimated, pre-decoder energy saving.