Intra-Prompt Parallel Decoding for Common-Context Question Answering
Intra-Prompt Parallel Decoding (IPPD), a novel inference method that answers multiple common-context questions in parallel within a single prompt without requiring fine-tuning or any changes to the LLM architecture, is presented.
Theodore Glavas, Nikhita Vedula, Dushyanta Dhyani et al.
· 1 citation