OrthoRank selects tokens for computation by their orthogonality to the sink token in normalized hidden states, yielding better perplexity and accuracy than layer pruning at matched sparsity without additional training.
Fast and robust early-exiting framework for autoregressive language models with synchronized parallel decoding
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
-
OrthoRank: Token Selection via Sink Token Orthogonality for Efficient LLM inference
OrthoRank selects tokens for computation by their orthogonality to the sink token in normalized hidden states, yielding better perplexity and accuracy than layer pruning at matched sparsity without additional training.