3 papers
cs.SD2026
Whisper-Flash: Acoustically Conditioned Parallel Drafting for Faster Whisper Decoding
Huapeng Zhou, Huayu Wang, Junkai Wu +2
Whisper is a widely used encoder-decoder model for speech recognition. Its encoder reads an utterance in one parallel pass, but its decoder writes the transcript one token at a tim…
cs.SD2026
Alignment Drift in Single-Model Speculative Decoding for ASR: Mechanism, Correction, and Cost
Xinyu Wang, Huapeng Zhou, Ziyu Zhao +5
Speculative decoding speeds up generation by letting a cheap draft propose several tokens that a target model checks in one pass. In the single-model form, the draft is a lightweig…
cs.CL2026
TreeSpark: Calibrated, Load-Adaptive Draft Trees for Semi-Autoregressive Speculative Decoding
Huapeng Zhou, Huayu Wang, Xinyu Wang
Speculative decoding accelerates language-model inference by letting a cheap drafter propose tokens that the target model verifies in parallel. Recent block drafters make drafting…