Search

Few direct matches — filled in with the latest updates.

Tag: #latent-cache1 results

MobCache Scales LLM Mobility Sims

MobCache Scales LLM Mobility Sims

MobCache is a mobility-aware cache framework that enables scalable LLM-based human mobility simulations by reusing reconstructible reasoning caches. It encodes reasoning steps as latent embeddings for recombination and uses a lightweight decoder trained via mobility-constrained distillation. Experiments demonstrate major efficiency gains while matching state-of-the-art performance.