The CBR-LLM synergies workshop explored opportunities for combining Case-Based Reasoning (CBR) and Large Language Models (LLMs) to foster a deeper understanding of their synergistic potential. The workshop presented the first opportunity for researchers and practitioners to discuss and share their insights on this new, exciting, and rapidly evolving field of AI. 2023 was the year LLMs broke into public attention following the release of ChatGPT.
The potential of LLMs to act as a conversational interface to CBR systems for query formulation and solution explanation seems obvious, but has yet to be explored. In addition, LLMs can be used to limit the knowledge acquisition and modelling eforts for initially creating and representing knowledge in the CBR knowledge containers. Finally, technologies associated with LLMs, such as vector databases and Approximate Nearest Neighbor retrieval, are of great interest to CBR. Conversely, CBR could benefit the function of LLMs by improving prompt engineering, reducing hallucinations, and providing a persistent memory of previous conversations.