Attention Editing converts pre-trained LLMs to new attention architectures through layer-wise teacher-forced optimization and model-level distillation, preserving performance with efficiency gains.
Towards economical inference: Enabling deepseek’s multi-head latent attention in any transformer-based llms.arXiv preprint arXiv:2502.14837
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CL 2verdicts
CONDITIONAL 2roles
background 1polarities
background 1representative citing papers
A3 splits Transformer layers into QK, OV, and MLP components and derives analytical low-rank approximations that reduce hidden dimensions while minimizing each component's functional loss, yielding better perplexity than prior low-rank methods on LLaMA models.
citing papers explorer
-
Attention Editing: A Versatile Framework for Cross-Architecture Attention Conversion
Attention Editing converts pre-trained LLMs to new attention architectures through layer-wise teacher-forced optimization and model-level distillation, preserving performance with efficiency gains.
-
A3 : an Analytical Low-Rank Approximation Framework for Attention
A3 splits Transformer layers into QK, OV, and MLP components and derives analytical low-rank approximations that reduce hidden dimensions while minimizing each component's functional loss, yielding better perplexity than prior low-rank methods on LLaMA models.