Splitting mixed search-recommend behavior into multiple interests lifts both tasks

Multi Interests for Joint Search-Recommendation Modeling

Xiangchen Pan, Wei Wei, Huakang Niu, Zhicong Cheng

cs.IR

2026-08-11

MIJSR mines structural and semantic multi-interests from mixed search-recommend sequences, lifting search HR@5 by about 4.5% on KuaiSAR.

What problem this solves

Search and recommendation are two routes into user preference: search is the user actively stating a need, recommendation is passive consumption. A growing line of work models them jointly, feeding a single mixed behavior sequence to serve both tasks.

The trouble is that existing methods encode search and recommendation behaviors into one unified sequence and lose sight of the fact that a user's interests differ by scenario. Someone who browses gadgets and designer toys long-term may suddenly need to book a flight; in a unified model that short-term travel intent gets drowned out by long-standing historical interests.

Method

MIJSR (Multi-Interest Joint Search-Recommendation) pulls the interests apart inside the mixed sequence. Three modules:

Results

On two public datasets, KuaiSAR and Amazon Kindle, against UniSAR and other baselines:

TaskDatasetMetricMIJSR vs baseline
SearchKuaiSARHR@50.7811 vs 0.7412 (+4.5%)
SearchKuaiSARNDCG@50.6635 vs 0.6397 (+3.7%)
RecKuaiSARHR@50.5115 vs 0.4887
SearchKindleHR@50.8265 vs 0.8149

Search gains are clearer; recommendation also moves up slightly.

Why it matters

For teams building unified search and recommendation, splitting interests by scenario is a sensible, effective direction, especially for preserving short-term intent inside a mixed sequence. This is an incremental improvement, and every module reuses off-the-shelf components (contrastive learning, PLE, clustering), so the bar to deploy is low.

Limitations

The authors acknowledge a multi-task seesaw: search and recommendation can trade off, one rising as the other falls, and they early-stop on average NDCG@5 to balance it. The cluster count k for semantic clustering is sensitive: too few is coarse, too many is sparse and noisy. The reproduced UniSAR baseline does not match the original paper, which the authors attribute to the seesaw and early-stopping differences, so the absolute gain may not be fully comparable.

Terms

Source

What people are saying

Related papers

All paper explainers