Search for: All records

We all have moments when we are struck by a “gut feeling” or a “sixth sense” about something. It could pertain to a relationship or task at work. That sense can be broadly termed intuition. Intuitive decisionmaking is an essential characteristic of individuals who have attained a certain level of expertise. The development of expertise and intuition are heavily influenced by experience. Engineering intuition is defined as an experience-informed skill subconsciously leveraged in problem solving by engineering practitioners when under pressure from constraints such as lack of time. Practicing engineers use and develop intuition regularly on-the-job, but the use of intuition is often discouraged in undergraduate education. The disconnect between intuition’s use in engineering practice and in education, coupled with our limited knowledge of the relationship between intuition, expertise, and experience, presents an important gap in our existing understanding of engineering problem solving and future workforce preparation. Through this Research in the Formation of Engineers (RFE) grant, we seek to address this gap by examining the application of intuition by engineering practitioners to generate knowledge that promotes professional formation and development of a stronger engineering workforce. 

Retrieval-augmented generation (RAG) systems can effectively address user queries by leveraging indexed document corpora to retrieve the relevant contexts. Ranking techniques have been adopted in RAG systems to sort the retrieved contexts by their relevance to the query so that users can select the most useful contexts for their downstream tasks. While many existing ranking methods rely on the similarity between the embedding vectors of the context and query to measure relevance, it is important to note that similarity does not equate to relevance in all scenarios. Some ranking methods use large language models (LLMs) to rank the contexts by putting the query and the candidate contexts in the prompt and asking LLM about their relevance. The scalability of those methods is contingent on the number of candidate contexts and the context window of those LLMs. Also, those methods require fine-tuning the LLMs, which can be computationally expensive and require domain-related data. In this work, we propose a scalable ranking framework that does not involve LLM training. Our framework uses an off-the-shelf LLM to hypothesize the user's query based on the retrieved contexts and ranks the contexts based on the similarity between the hypothesized queries and the user query. Our framework is efficient at inference time and is compatible with many other context retrieval and ranking techniques. Experimental results show that our method improves the ranking performance of retrieval systems in multiple benchmarks.