IntSR: An Integrated Generative Framework for Search and Recommendation
This work addresses the integration of search and recommendation for applications like Amap, offering incremental improvements in performance metrics.
The paper tackles the problem of integrating search and recommendation tasks, which have distinct query modalities, by proposing IntSR, a generative framework that addresses computational complexity and erroneous pattern learning, resulting in substantial improvements such as a 9.34% increase in GMV, 2.76% increase in CTR, and 7.04% increase in ACC in deployed scenarios.
Generative recommendation has emerged as a promising paradigm, demonstrating remarkable results in both academic benchmarks and industrial applications. However, existing systems predominantly focus on unifying retrieval and ranking while neglecting the integration of search and recommendation (S&R) tasks. What makes search and recommendation different is how queries are formed: search uses explicit user requests, while recommendation relies on implicit user interests. As for retrieval versus ranking, the distinction comes down to whether the queries are the target items themselves. Recognizing the query as central element, we propose IntSR, an integrated generative framework for S&R. IntSR integrates these disparate tasks using distinct query modalities. It also addresses the increased computational complexity associated with integrated S&R behaviors and the erroneous pattern learning introduced by a dynamically changing corpus. IntSR has been successfully deployed across various scenarios in Amap, leading to substantial improvements in digital asset's GMV(+9.34%), POI recommendation's CTR(+2.76%), and travel mode suggestion's ACC(+7.04%).