{"ID":2827679,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2512.16843","arxiv_id":"2512.16843","title":"LLMCache: Layer-Wise Caching Strategies for Accelerated Reuse in Transformer Inference","abstract":"Transformer-based language models have achieved remarkable performance across a wide range of tasks, yet their high inference latency poses a significant challenge for real-timeand large-scale deployment. While existing caching mechanisms,such as token-level key-value caches, offer speedups in autore-gressive decoding, they are limited in scope and applicability. In this paper, we present LLMCache, a novel layer-wise caching framework that accelerates transformer inference by reusing intermediate activations based on semantic similarity of input sequences. Unlike prior work, LLMCache is model-agnostic,operates across both encoder and decoder architectures, and supports caching at arbitrary transformer layers. We introduce a lightweight fingerprinting mechanism for matching seman-tically similar inputs and propose adaptive eviction strategies to manage cache staleness. Experiments on BERT and GPT-2 across SQuAD, WikiText-103, and OpenBookQA show up to 3.1 X speedup in inference time with \u003c0.5% accuracy degradation. Our results highlight LLMCache as a practical and general-purpose solution for optimizing transformer inference in real-world applications","short_abstract":"Transformer-based language models have achieved remarkable performance across a wide range of tasks, yet their high inference latency poses a significant challenge for real-timeand large-scale deployment. While existing caching mechanisms,such as token-level key-value caches, offer speedups in autore-gressive decoding,...","url_abs":"https://arxiv.org/abs/2512.16843","url_pdf":"https://arxiv.org/pdf/2512.16843v1","authors":"[\"Harsh Vardhan Bansal\"]","published":"2025-12-18T18:18:57Z","proceeding":"cs.CL","tasks":"[\"cs.CL\",\"cs.AI\"]","methods":"[\"Transformer\",\"Large Language Model\",\"Language Model\"]","has_code":false}