paper-with-me

홈 › Papers

Distil-xLSTM: Learning Attention Mechanisms through Recurrent Structures

2025-03-24 · Abdoul Majid O. Thiombiano, Brahim Hnich, Ali Ben Mrad, Mohamed Wiem Mkaouer

The current era of Natural Language Processing (NLP) is dominated by Transformer models. However, novel architectures relying on recurrent mechanisms, such as xLSTM and Mamba, have been proposed as alternatives to attention-based models. Although computation is done differently than with the attention mechanism mechanism, these recurrent models yield good results and sometimes even outperform state-of-the-art attention-based models. In this work, we propose Distil-xLSTM, an xLSTM-based Small Language Model (SLM) trained by distilling knowledge from a Large Language Model (LLM) that shows promising results while being compute and scale efficient. Our Distil-xLSTM focuses on approximating a transformer-based model attention parametrization using its recurrent sequence mixing components and shows good results with minimal training.

📄 PDF Abstract BibTeX arXiv:2503.18565

Code (0)

등록된 구현이 없습니다.

Tasks

Language ModelingLanguage ModellingLarge Language ModelMambaSmall Language Model

Methods 이 논문이 사용한 방법론

Attention 설명 없음
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Multi-Head Attention 설명 없음
Dense Connections Dense Connections, or Fully Connected Connections, are a type of layer in a deep neural network that use a linear operation where every input is connected to every output…
Absolute Position Encodings Absolute Position Encodings are a type of position embeddings for [Transformer-based models] where positional encodings are…
Residual Connection 설명 없음
Mamba Foundation models, now powering most of the exciting applications in deep learning, are almost universally based on the Transformer architecture and its core attention module.…
Adam 설명 없음

Similar Papers 제목 키워드 기반

On Subquadratic Architectures: From Applications to Principles

2026-06-10 · Anamaria-Roberta Hartl, Levente Zólyomi, David Stap, Pieter-Jan Hoedt 외 arxiv

Transformers dominate modern sequence modeling, but their quadratic attention incurs substantial computational cost. Subquadratic architectures offer a scalable alternative. However, it remains unclear which designs yiel…

Effective Distillation to Hybrid xLSTM Architectures

2026-03-16 · Lukas Hauzenberger, Niklas Schmidinger, Thomas Schmied, Anamaria-Roberta Hartl 외 arxiv

There have been numerous attempts to distill quadratic attention-based large language models (LLMs) into sub-quadratic linearized architectures. However, despite extensive research, such distilled models often fail to ma…

xLSTM-SENet: xLSTM for Single-Channel Speech Enhancement

2025-01-10 · Nikolai Lund Kühne, Jan Østergaard, Jesper Jensen, Zheng-Hua Tan

While attention-based architectures, such as Conformers, excel in speech enhancement, they face challenges such as scalability with respect to input sequence length. In contrast, the recently proposed Extended Long Short…

MambaSpeech Enhancement

xLSTM 7B: A Recurrent LLM for Fast and Efficient Inference

2025-03-17 · Maximilian Beck, Korbinian Pöppel, Phillip Lippe, Richard Kurle 외

Recent breakthroughs in solving reasoning, math and coding problems with Large Language Models (LLMs) have been enabled by investing substantial computation budgets at inference time. Therefore, inference speed is one of…

MambaMath

A Large Recurrent Action Model: xLSTM enables Fast Inference for Robotics Tasks

2024-10-29 · Thomas Schmied, Thomas Adler, Vihang Patil, Maximilian Beck 외

In recent years, there has been a trend in the field of Reinforcement Learning (RL) towards large action models trained offline on large-scale datasets via sequence modeling. Existing models are primarily based on the Tr…

MambaReinforcement Learning (RL)