STARec: An Efficient Agent Framework for Recommender Systems via Autonomous Deliberate Reasoning
This work addresses the problem of shallow correlation bias and brittleness in sparse-data scenarios for recommender systems, offering a novel agent-based approach that is incremental in enhancing reasoning capabilities.
The authors tackled the limitations of static user modeling and reactive decision-making in recommender systems by introducing STARec, a framework that integrates autonomous deliberative reasoning, resulting in substantial performance gains on benchmarks like MovieLens 1M and Amazon CDs while using only 0.4% of the full training data.
While modern recommender systems are instrumental in navigating information abundance, they remain fundamentally limited by static user modeling and reactive decision-making paradigms. Current large language model (LLM)-based agents inherit these shortcomings through their overreliance on heuristic pattern matching, yielding recommendations prone to shallow correlation bias, limited causal inference, and brittleness in sparse-data scenarios. We introduce STARec, a slow-thinking augmented agent framework that endows recommender systems with autonomous deliberative reasoning capabilities. Each user is modeled as an agent with parallel cognitions: fast response for immediate interactions and slow reasoning that performs chain-of-thought rationales. To cultivate intrinsic slow thinking, we develop anchored reinforcement training - a two-stage paradigm combining structured knowledge distillation from advanced reasoning models with preference-aligned reward shaping. This hybrid approach scaffolds agents in acquiring foundational capabilities (preference summarization, rationale generation) while enabling dynamic policy adaptation through simulated feedback loops. Experiments on MovieLens 1M and Amazon CDs benchmarks demonstrate that STARec achieves substantial performance gains compared with state-of-the-art baselines, despite using only 0.4% of the full training data.