paper-with-me

Papers

EL-Attention: Memory Efficient Lossless Attention for Generation

2021-05-11 · Yu Yan, Jiusheng Chen, Weizhen Qi, Nikhil Bhendawade, Yeyun Gong, Nan Duan, Ruofei Zhang

Transformer model with multi-head attention requires caching intermediate results for efficient inference in generation tasks. However, cache brings new memory-related costs and prevents leveraging larger batch size for faster speed. We propose memory-efficient lossless attention (called EL-attention) to address this issue. It avoids heavy operations for building multi-head keys and values, cache for them is not needed. EL-attention constructs an ensemble of attention results by expanding query while keeping key and value shared. It produces the same result as multi-head attention with less GPU memory and faster inference speed. We conduct extensive experiments on Transformer, BART, and GPT-2 for summarization and question generation tasks. The results show EL-attention speeds up existing models by 1.6x to 5.3x without accuracy loss.

📄 PDF Abstract BibTeX arXiv:2105.04779

Code (1)

microsoft/fastseq 공식 구현 pytorch

Tasks

GPUQuestion GenerationQuestion-Generation

Methods 이 논문이 사용한 방법론

Attention 설명 없음
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Absolute Position Encodings Absolute Position Encodings are a type of position embeddings for [Transformer-based models] where positional encodings are…
Position-Wise Feed-Forward Layer 설명 없음
Cosine Annealing Cosine Annealing is a type of learning rate schedule that has the effect of starting with a large learning rate that is relatively rapidly decreased to a minimum value before…
Label Smoothing Label Smoothing is a regularization technique that introduces noise for the labels. This accounts for the fact that datasets may have mistakes in them, so maximizing the…
Residual Connection 설명 없음
BPE Byte Pair Encoding, or BPE, is a subword segmentation algorithm that encodes rare and unknown words as sequences of subword units. The intuition is that various word…

Similar Papers 제목 키워드 기반

LoMA: Lossless Compressed Memory Attention

2024-01-16 · Yumeng Wang, Zhenyang Xiao

Large Language Models (LLMs) face limitations due to the high demand on GPU memory and computational resources when handling long contexts. While sparsify the Key-Value (KV) cache of transformer model is a typical strate…

GPU

Slim attention: cut your context memory in half without loss of accuracy -- K-cache is all you need for MHA

2025-03-07 · Nils Graef, Andrew Wasielewski

Slim attention shrinks the context memory size by 2x for transformer models with MHA (multi-head attention), which can speed up inference by up to 2x for large context windows. Slim attention is an exact, mathematically …

AllDecoder

Towards Memory-Efficient Autoregressive Video Generation via Instance-Specific Parametric Absorption

2026-07-01 · Xiaomeng Fu, Jia Li, Yiming Hu, Yong Wang 외 arxiv

Autoregressive (AR) streaming models have emerged as a powerful paradigm for long video generation. However, the linearly growing Key-Value (KV) cache poses a significant bottleneck, leading to memory overload and degrad…

Video Generation

Lossless KV Cache Compression to 2%

2024-10-20 · Zhen Yang, J. N. Han, Kan Wu, Ruobing Xie 외

Large language models have revolutionized data processing in numerous domains, with their ability to handle extended context reasoning receiving notable recognition. To speed up inference, maintaining a key-value (KV) ca…

Dimensionality ReductionQuantization

Artificial Hippocampus Networks for Efficient Long-Context Modeling

2025-10-08 · Yunhao Fang, Weihao Yu, Shu Zhong, Qinghao Ye 외 arxiv

Long-sequence modeling faces a fundamental trade-off between the efficiency of compressive fixed-size memory in RNN-like models and the fidelity of lossless growing memory in attention-based Transformers. Inspired by the…