CD-LM interleaves token-level generation with retrieval of multi-token chunks from a trie datastore, yielding exact sequence probabilities via a dynamic program and improving perplexity and inference speed without training.
The parametrization ϕ is none
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Chunk-Distilled Language Modeling
CD-LM interleaves token-level generation with retrieval of multi-token chunks from a trie datastore, yielding exact sequence probabilities via a dynamic program and improving perplexity and inference speed without training.