An Analysis of Decoding Methods for LLM-based Agents for Faithful Multi-Hop Question Answering

Open in new window