Search papers, labs, and topics across Lattice.
1
0
3
Don't fully retrain your draft model after fine-tuning your LLM: EDA restores speculative decoding performance with significantly less compute by adapting only a small, private component and regenerating training data.
