paper-with-me

홈 › Papers

ARWKV: Pretrain is not what we need, an RNN-Attention-Based Language Model Born from Transformer

2025-01-26 · Lin Yueyu, Li Zhiyuan, Peter Yue, Liu Xiao

As is known, hybrid quadratic and subquadratic attention models in multi-head architectures have surpassed both Transformer and Linear RNN models , with these works primarily focusing on reducing KV complexity and improving efficiency. For further research on expressiveness, we introduce our series of models distilled from Qwen 2.5, based on pure native RWKV-7 attention, which aims to make RNN more expressive and demonstrates state tracking ability beyond transformers. We work with QRWK 32B based on RWKV-6 architecture, another approach that reduces the entire knowledge processing time to just 8 hours using 16 AMD MI300X GPUs while maintaining Qwen 2.5's performance. In fact, the distillation process can utilize any LLM, not just Qwen, and enables knowledge transfer from larger LLMs to smaller ones with more fewer tokens. We will explain the detailed process and share our insights on building more powerful foundation models. Please note that this is an ongoing work that will be updated continuously. The model checkpoints and source code are available at \href{https://github.com/yynil/RWKVInside}{https://github.com/yynil/RWKVInside}, \href{https://huggingface.co/RWKV-Red-Team/ARWKV-7B-Preview-0.1}{https://huggingface.co/RWKV-Red-Team/ARWKV-7B-Preview-0.1}.

📄 PDF Abstract BibTeX arXiv:2501.15570

Code (1)

yynil/rwkvinside 공식 구현 pytorch

Tasks

Language ModelingLanguage ModellingTransfer Learning

Methods 이 논문이 사용한 방법론

Attention 설명 없음
Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…
Residual Connection 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…
Absolute Position Encodings Absolute Position Encodings are a type of position embeddings for [Transformer-based models] where positional encodings are…
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…
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Multi-Head Attention 설명 없음

Similar Papers 제목 키워드 기반

Parametric Knowledge is Not All You Need: Toward Honest Large Language Models via Retrieval of Pretraining Data

2026-01-29 · Christopher Adrian Kusuma, Muhammad Reza Qorib, Hwee Tou Ng arxiv

Large language models (LLMs) are highly capable of answering questions, but they are often unaware of their own knowledge boundary, i.e., knowing what they know and what they don't know. As a result, they can generate fa…

What and How does In-Context Learning Learn? Bayesian Model Averaging, Parameterization, and Generalization

2023-05-30 · Yufeng Zhang, Fengzhuo Zhang, Zhuoran Yang, Zhaoran Wang

In this paper, we conduct a comprehensive study of In-Context Learning (ICL) by addressing several open questions: (a) What type of ICL estimator is learned by large language models? (b) What is a proper performance metr…

In-Context Learning

When Do You Need Billions of Words of Pretraining Data?

2020-11-10 · ACL 2021 5 · Yian Zhang, Alex Warstadt, Haau-Sing Li, Samuel R. Bowman

NLP is currently dominated by general-purpose pretrained language models like RoBERTa, which achieve strong performance on NLU tasks through pretraining on billions of words. But what exact knowledge or skills do Transfo…

What Would Elsa Do? Freezing Layers During Transformer Fine-Tuning

2019-11-08 · Jaejun Lee, Raphael Tang, Jimmy Lin

Pretrained transformer-based language models have achieved state of the art across countless tasks in natural language processing. These models are highly expressive, comprising at least a hundred million parameters and …

Linguistic AcceptabilityNatural Language InferenceSemantic SimilaritySemantic Textual Similarity+1

Difference-Masking: Choosing What to Mask in Continued Pretraining

2023-05-23 · Alex Wilf, Syeda Nahida Akter, Leena Mathur, Paul Pu Liang 외

The self-supervised objective of masking-and-predicting has led to promising performance gains on a variety of downstream tasks. However, while most approaches randomly mask tokens, there is strong intuition that decidin…

Self-Supervised Learning