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.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.IR 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
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.