conftrace_

multi-query attention

3 papers

Explore in graph

Also known as

MQA

Co-occurring keywords · papers mentioning both

grouped-query attention (3) memory efficiency (2) cross-layer attention (1) autoregressive decoding (1) inference speedup (1) memory reduction (1) key-value caching (1) transformer decoding (1) large language model (1) transformer model (1)

Papers

MLKV: Multi-Layer Key-Value Heads for Memory Efficient Transformer Decoding NAACL 2025 Reducing Transformer Key-Value Cache Size with Cross-Layer Attention NeurIPS 2024 GQA: Training Generalized Multi-Query Transformer Models from Multi-Head Checkpoints EMNLP 2023