conftrace_

cross-layer attention

2 papers

Explore in graph

Also known as

CLA

Co-occurring keywords · papers mentioning both

multi-query attention (1) autoregressive decoding (1) feature distillation (1) model heterogeneity (1) memory reduction (1) grouped-query attention (1) large language model (1) key-value cache (1) memory optimization (1) memory efficiency (1)

Papers

FedCD: Towards Consolidated Distillation for Heterogeneous Federated Learning AAAI 2026 Reducing Transformer Key-Value Cache Size with Cross-Layer Attention NeurIPS 2024