A hybrid decoding scheme with a fast TDT draft decoder and selective transformer patches matches baseline word error rate while cutting decoder latency roughly threefold.
Fast inference from transformers via speculative decoding
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
eess.AS 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
support 1representative citing papers
citing papers explorer
-
Hybrid Decoding: Rapid Pass and Selective Detailed Correction for Sequence Models
A hybrid decoding scheme with a fast TDT draft decoder and selective transformer patches matches baseline word error rate while cutting decoder latency roughly threefold.