OmniDraft: A Cross-vocabulary, Online Adaptive Drafter for On-device Speculative Decoding
Paper
•
2507.02659
•
Published
•
15
Ramchalam K R
ramchalamkr
AI & ML interests
Small language models, model compression, LLM, TimyML, RL
Recent Activity
upvoted
a
paper
10 days ago
Edge-ASR: Towards Low-Bit Quantization of Automatic Speech Recognition
Models
authored
a paper
25 days ago
OmniDraft: A Cross-vocabulary, Online Adaptive Drafter for On-device
Speculative Decoding
Organizations
None yet
authored
2
papers
25 days ago