Adapting Generative Recommenders for Multi-Turn Interaction

2026-10-06

AI Summary

This study addresses the absence of user intent correction mechanisms in existing generative recommender systems, where directly incorporating dialogue often disrupts history-item mappings and induces catastrophic forgetting. To this end, we propose INTEGER, a novel multi-turn interaction framework that uniquely integrates dynamic routing with anti-forgetting mechanisms. Specifically, INTEGER employs routing tokens to triage interactions, leverages a history re-anchoring strategy to stabilize representation mappings, and incorporates behavior replay techniques to mitigate forgetting. Evaluated on Amazon datasets, the proposed framework achieves a 13.3% improvement in Hit@10 over competitive baselines, effectively reconciling high recommendation accuracy with fluent multi-turn conversational capabilities.

Abstract

Generative recommenders decode items from a user's interaction history, but offer no way for users to correct a recommendation that misses their current intent. Adding conversation is natural since items and words share same output space, yet training the model to converse may overwrite the history-to-item mapping it relies on. We introduce INTEGER (**INTE**ractive **GE**nerative **R**ecommendation), which extends generative recommendation to multi-turn interaction with a learned routing token that lets the model decide when to recommend, history re-anchoring that conditions each item on both past behavior and the dialogue, and behavioral replay with instruction-data rehearsal that prevents forgetting during adaptation. Users can thus give feedback on recommendations within the dialogue, while recommendations stay grounded in behavioral history and accuracy is not traded for fluency. On Amazon Beauty and Toys, INTEGER matches or exceeds the strongest baselines in accuracy with competitive conversation quality, improving Hit@10 by 13.3% on Amazon Beauty, and significantly outperforms the generative recommender it starts from. Our analyses show that INTEGER learns behaviors that naive adaptation fails to acquire, recommending once the user's intent is clear and staying attentive to behavioral history at the moment of recommendation. INTEGER also learns an intent-agnostic replacement over the item space, which suppresses rejected items but points to attribute-aware feedback as the next step.

Problem

• Generative Recommenders
• Multi-Turn Interaction
• Conversational Recommendation
• Catastrophic Forgetting
• User Feedback

Innovation

• Generative Recommender
• Multi-turn Interaction
• Routing Token
• History Re-anchoring
• Behavioral Replay