RETRO-LI: Small-Scale Retrieval Augmented Generation Supporting Noisy Similarity Searches and Domain Shift Generalization

The retrieval augmented generation (RAG) system such as Retro has been shown to improve language modeling capabilities and reduce toxicity and hallucinations by retrieving from a database of non-parametric memory containing trillions of entries. We introduce RETRO-LI that shows retrieval can also help at a small scale database but it demands more accurate and better neighbors when searching in such a smaller hence sparser non-parametric memory. This can be met by using a proper semantic similarity search. We further propose adding a regularization to the non-parametric memory for the first time: it significantly reduces perplexity when the neighbor search operations are noisy during inference, and it improves generalization when a domain shift occurs. We also show that the RETRO-LI's non-parametric memory can be potentially implemented on analog in-memory computing hardware, exhibiting $O(1)$ search time while causing noise in retrieving neighbors, with minimal (<1%) performance loss.