paper-with-me

홈 › Papers

LoRAP: Transformer Sub-Layers Deserve Differentiated Structured Compression for Large Language Models

2024-04-15 · Guangyan Li, Yongqiang Tang, Wensheng Zhang

Large language models (LLMs) show excellent performance in difficult tasks, but they often require massive memories and computational resources. How to reduce the parameter scale of LLMs has become research hotspots. In this study, we make an important observation that the multi-head self-attention (MHA) sub-layer of Transformer exhibits noticeable low-rank structure, while the feed-forward network (FFN) sub-layer does not. With this regard, we design a mixed compression model, which organically combines Low-Rank matrix approximation And structured Pruning (LoRAP). For the MHA sub-layer, we propose an input activation weighted singular value decomposition method to strengthen the low-rank characteristic. Furthermore, we discover that the weight matrices in MHA sub-layer have different low-rank degrees. Thus, a novel parameter allocation scheme according to the discrepancy of low-rank degrees is devised. For the FFN sub-layer, we propose a gradient-free structured channel pruning method. During the pruning, we get an interesting finding that the least important 1% of parameter actually play a vital role in model performance. Extensive evaluations on zero-shot perplexity and zero-shot task classification indicate that our proposal is superior to previous structured compression rivals under multiple compression ratios.

📄 PDF Abstract BibTeX arXiv:2404.09695

Code (0)

등록된 구현이 없습니다.

Methods 이 논문이 사용한 방법론

Attention 설명 없음
Dropout Dropout is a regularization technique for neural networks that drops a unit (along with connections) at training time with a specified probability $p$ (a common value is…
Adam 설명 없음
Position-Wise Feed-Forward Layer 설명 없음
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Layer Normalization Unlike batch normalization, Layer Normalization directly estimates the normalization statistics from the summed inputs…
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…
Absolute Position Encodings Absolute Position Encodings are a type of position embeddings for [Transformer-based models] where positional encodings are…

Similar Papers 제목 키워드 기반

LoRAPrune: Structured Pruning Meets Low-Rank Parameter-Efficient Fine-Tuning

2023-05-28 · Mingyang Zhang, Hao Chen, Chunhua Shen, Zhen Yang 외

Large Language Models (LLMs), such as LLaMA and T5, have shown exceptional performance across various tasks through fine-tuning. Although low-rank adaption (LoRA) has emerged to cheaply fine-tune these LLMs on downstream…

Model CompressionNetwork Pruningparameter-efficient fine-tuning

LoRAP: Low-Rank Aggregation Prompting for Quantized Graph Neural Networks Training

2026-01-21 · Chenyu Liu, Haige Li, Luca Rossi arxiv

Graph Neural Networks (GNNs) are neural networks that aim to process graph data, capturing the relationships and interactions between nodes using the message-passing mechanism. GNN quantization has emerged as a promising…

Dualformer: Time-Frequency Dual Domain Learning for Long-term Time Series Forecasting

2026-01-22 · Jingjing Bai, Yoshinobu Kawahara arxiv

Transformer-based models, despite their promise for long-term time series forecasting (LTSF), suffer from an inherent low-pass filtering effect that limits their effectiveness. This issue arises due to undifferentiated p…

Time Series Forecasting

DialogueTRM: Exploring the Intra- and Inter-Modal Emotional Behaviors in the Conversation

2020-10-15 · Yuzhao Mao, Qi Sun, Guang Liu, Xiaojie Wang 외

Emotion Recognition in Conversations (ERC) is essential for building empathetic human-machine systems. Existing studies on ERC primarily focus on summarizing the context information in a conversation, however, ignoring t…

Emotion RecognitionEmotion Recognition in Conversation

General sum stochastic games with networked information flows

2022-05-05 · Sarah H. Q. Li, Lillian J. Ratliff, Peeyush Kumar

Inspired by applications such as supply chain management, epidemics, and social networks, we formulate a stochastic game model that addresses three key features common across these domains: 1) network-structured player i…

Decision MakingManagementMulti-agent Reinforcement Learningreinforcement-learning+1