Search papers, labs, and topics across Lattice.
This paper details the migration of a production conversational recommendation system from a gradient-boosted multiclass model to a pairwise-binary deep recommender to better handle the complexities of dynamically bundled offerings in customer support. The authors address the limitations of traditional tree-based classifiers in incorporating multimodal signals and maintaining recommendation quality during live interactions. Their approach, which includes attention pooling and various deep learning architectures, demonstrates improved performance over baseline methods, particularly in later stages of conversation.
Migrating from gradient-boosted trees to deep learning not only preserves but enhances recommendation quality in dynamic customer support environments.
Product catalogs in fast-moving service businesses are shifting from static, independently priced SKUs toward dynamically bundled, discount-coupled offerings--a shift that strains the tree-based classifiers traditionally preferred for sparse and highly imbalanced data. These classifiers assume a fixed, slowly changing label space and struggle to incorporate multimodal signals such as tabular data and transcripts. We present the migration of a live, production conversational recommendation system from a gradient-boosted multiclass model to a pairwise-binary deep recommender. Because this system is critical to ecosystem growth initiatives and downstream features like dynamic pitching--surfacing the most relevant pitch text to a support agent in real time during a live customer conversation--maintaining live recommendation quality was a non-negotiable constraint. We detail the techniques that made this migration successful--reformulating recommendation as pairwise binary prediction to learn jointly from user and item features, and enhancing learned representations via negative sampling and noise injection. To efficiently incorporate long, live conversation context, we apply attention pooling over transcript chunks and benchmark it against TF-IDF and sentence-embedding baselines. Finally, we explore multiple architectures (including two-tower models, DeepFM, and their variants) and loss functions such as contrastive loss. Evaluating against a CatBoost baseline across all conversational stages, we demonstrate that our approach achieves parity at conversation beginning and outperforms at later conversational stages.
