When Classic Cache Policies Fail: Learning-Augmented Replacement for Semantic Retrieval Buffers

When Classic Cache Policies Fail: Learning-Augmented Replacement for Semantic Retrieval Buffers

Research addresses cache management for LLM agents by formalizing semantic cache replacement as an online problem with switching costs, proposing SOLAR—a learning-augmented frame…

Hugging Face · Daily Papers ·Yushi Sun, Bowen Cao · ·▲ 4 upvotes

Este artigo está em destaque na seleção diária de papers do Hugging Face, curada pela comunidade de pesquisa em IA.

Autores: Yushi Sun, Bowen Cao, Wai Lam

  • 4 upvotes da comunidade
  • Temas: semantic cache replacement, online semantic cache replacement, switching costs, embedding similarity, regret accumulation, Bayesian online learning

Resumo

Resumo original (em inglês), extraído do paper:

Research addresses cache management for LLM agents by formalizing semantic cache replacement as an online problem with switching costs, proposing SOLAR—a learning-augmented framework that outperforms traditional heuristics through regret-based modification timing and Bayesian content selection.

Onde ler

compartilhar: