Closed-loop LLM simulations find generative recommenders form fewer exposure-level information cocoons than traditional sequential baselines on Amazon data, though tokenization strategy and model scale affect concentration in generated SID space.
Choosing the best of both worlds: Diverse and novel recommendations through multi-objective reinforcement learning,
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2verdicts
UNVERDICTED 2representative citing papers
Introduces semantic Pareto-DQN for multi-objective recommendation that sustains trajectory variance to improve diversity and fairness on MovieLens with limited engagement loss.
citing papers explorer
-
Do Generative Recommenders Deepen the Information Cocoon? A Closed-Loop Simulation with LLM-powered User Simulators
Closed-loop LLM simulations find generative recommenders form fewer exposure-level information cocoons than traditional sequential baselines on Amazon data, though tokenization strategy and model scale affect concentration in generated SID space.
-
Breaking the Filter Bubble: A Semantic Pareto-DQN Framework for Multi-Objective Recommendation
Introduces semantic Pareto-DQN for multi-objective recommendation that sustains trajectory variance to improve diversity and fairness on MovieLens with limited engagement loss.