TwiSTAR:Think Fast, Think Slow, Then Act,Generative Recommendation with Adaptive Reasoning

TL;DR

TwiSTAR introduces an adaptive framework for generative recommendation that dynamically balances fast and slow reasoning strategies, improving accuracy and reducing latency.

Contribution

It proposes a system that learns to allocate reasoning effort per user, combining retrieval, ranking, and explicit rationale generation with reinforcement learning.

Findings

Outperforms strong baselines on three datasets.

Achieves higher accuracy with lower inference latency.

Effectively balances fast and slow reasoning strategies.

Abstract

Generative recommendation with Semantic IDs (SIDs) has emerged as a promising paradigm, yet existing methods apply a fixed inference strategy, either fast direct generation or slow chain-of-thought reasoning, uniformly across all user histories. This approach creates a trade-off: fast recommendation model produces suboptimal accuracy on hard samples, while always invoking slow reasoning incurs prohibitive latency and wastes computation on easy cases. To address this, we propose Think Fast, Think Slow, Then Act, a framework that learns to adaptively allocate reasoning effort per user sequence. Our system equips an LLM with three complementary tools: a fast SID-based retriever, a lightweight candidate ranker, and a slow reasoning model that generates explicit rationales before recommending. Crucially, we inject collaborative commonsense into the slow model by transforming item-to-item knowledge into natural language explanations. A planner, trained through supervised warm-up followed by agentic reinforcement learning, dynamically decides which tool to invoke. Experiments on three datasets demonstrate that our method outperforms strong baselines, achieving consistent accuracy gains while reducing inference latency compared to uniform slow reasoning.