Shape Your Feed: An LLM-based Agentic System for Conversational Recommendation

Researchers introduced Shape Your Feed (SYF), an LLM-based agentic recommendation framework enabling real-time, multimodal co-curation of content, with a three-tier architecture including Perception Flow, Serving Flow, and Self-Evolution Flow. Offline evaluations show SYF's alignment scoring module achieves 98.85% accuracy, and large-scale online A/B experiments on production traffic demonstrate improved feed relevance and user sentiment.

arXiv:2608.06632v1 Announce Type: new Abstract: Industrial recommendation systems predominantly adopt a passive ranking paradigm that infers user preferences from implicit behavioral signals e.g., clicks, dwell time rather than explicit, natural language inputs. As a result, users experience a persistent discrepancy between their explicit interests and what passive behavioral algorithms deliver, limiting their ability to express nuanced preferences or steer their feed in real time. To address this growing gap between how recommendations are optimized and how users wish to articulate their interests, we present Shape Your Feed SYF , an LLM-based agentic recommendation framework that enables real-time, multimodal co-curation of content. SYF employs a three-tier architecture: i a Perception Flow that captures fine-grained user intent from text prompts, voice commands, and UI interactions; ii a Serving Flow that performs real-time agentic re-ranking and pruning of candidate items, grounded in a persistent Semantic Profile encoding evolving user preferences; and iii a Self-Evolution Flow that aligns system behavior with human judgments via Direct Preference Optimization DPO and an LLM-as-a-Judge ensemble. Offline evaluations show that SYF's alignment scoring module achieves 98.85% accuracy, substantially improving over strong few-shot baselines. Large-scale online A/B experiments on production traffic further demonstrate that SYF improves feed relevance and user sentiment, indicating a practical and scalable path toward interactive, user-steerable recommendation in industrial settings.