Exploring the Practicality of Generative Retrieval on Dynamic Corpora
This addresses the challenge of building practical information retrieval systems for dynamic environments, offering incremental improvements over existing methods.
The paper tackled the problem of applying generative retrieval to dynamic corpora, showing that it is more adaptable (4-11% improvement), robust, and efficient in inference FLOPs (x2), indexing time (x6), and storage (x4) compared to dual encoders.
Benchmarking the performance of information retrieval (IR) is mostly conducted with a fixed set of documents (static corpora). However, in realistic scenarios, this is rarely the case and the documents to be retrieved are constantly updated and added. In this paper, we focus on Generative Retrievals (GR), which apply autoregressive language models to IR problems, and explore their adaptability and robustness in dynamic scenarios. We also conduct an extensive evaluation of computational and memory efficiency, crucial factors for real-world deployment of IR systems handling vast and ever-changing document collections. Our results on the StreamingQA benchmark demonstrate that GR is more adaptable to evolving knowledge (4-11%), robust in learning knowledge with temporal information, and efficient in terms of inference FLOPs (x2), indexing time (x6), and storage footprint (x4) compared to Dual Encoders (DE), which are commonly used in retrieval systems. Our paper highlights the potential of GR for future use in practical IR systems within dynamic environments.