Ad-hoc retrieval is an important problem with many practical applications. It forms the basis of web search, question-answering, and a new generation of virtual assistants being developed by several of the largest software companies in the world. In this report, we continue our exploration of the importance of multiple expressions of information needs. Our thesis is that over-reliance on a single query can lead to suboptimal performance, and that by creating multiple query representations for an information need and combining the relevance signals through fusion and relevance modeling, highly effective systems can be produced. This approach may form the basis for more complex multi-stage retrieval systems in a variety of applications.