A5060290590- Recommender Systems and Techniques
- Topic Modeling
- Advanced Graph Neural Networks
- Expert finding and Q&A systems
- Advanced Bandit Algorithms Research
- Data Stream Mining Techniques
- Image and Video Quality Assessment
Google (United States)
2023-2024
The reasoning and generalization capabilities of LLMs can help us better understand user preferences item characteristics, offering exciting prospects to enhance recommendation systems. Though effective while user-item interactions are abundant, conventional systems struggle recommend cold-start items without historical interactions. To address this, we propose utilizing as data augmenters bridge the knowledge gap on during training. We employ infer for based textual description behaviors...
Effective exploration is believed to positively influence the long-term user experience on recommendation platforms. Determining its exact benefits, however, has been challenging. Regular A/B tests often measure neutral or even negative engagement metrics while failing capture benefits. We here introduce new experiment designs formally quantify value of by examining effects content corpus, and connecting corpus growth from real-world experiments. Once established values exploration, we...
Recommendation system serves as a conduit connecting users to an incredibly large, diverse and ever growing collection of contents. In practice, missing information on fresh contents needs be filled in order for them exposed discovered by their audience. this context, we are delighted share our success stories building dedicated content recommendation stack large commercial platform also shed light the utilization Large Language Models (LLMs) recommendations within industrial framework. To...
Exposure bias and its induced feedback loop effect are well-known problems in recommender systems. Exploration is believed to be the key break such loops. While classical contextual bandit algorithms as Upper-Confidence-Bound Thompson Sampling have been successful addressing exploration-exploitation trade-off single-task settings with one clear reward signal, modern systems often leverage multiple rich sources of clicks, likes, dislikes, shares, satisfaction survey responses, employ...
Traditional recommendation systems are subject to a strong feedback loop by learning from and reinforcing past user-item interactions, which in turn limits the discovery of novel user interests. To address this, we introduce hybrid hierarchical framework combining Large Language Models (LLMs) classic models for interest exploration. The controls interfacing between LLMs through "interest clusters", granularity can be explicitly determined algorithm designers. It recommends next interests...
Recommendation system serves as a conduit connecting users to an incredibly large, diverse and ever growing collection of contents. In practice, missing information on fresh (and tail) contents needs be filled in order for them exposed discovered by their audience. We here share our success stories building dedicated content recommendation stack large commercial platform. To nominate contents, we built multi-funnel nomination that combines (i) two-tower model with strong generalization power...
Recommendation system serves as a conduit connecting users to an incredibly large, diverse and ever growing collection of contents. In practice, missing information on fresh (and tail) contents needs be filled in order for them exposed discovered by their audience. We here share our success stories building dedicated content recommendation stack large commercial platform. To nominate contents, we built multi-funnel nomination that combines (i) two-tower model with strong generalization power...