Papers
arxiv:2506.17671

TPTT: Transforming Pretrained Transformer into Titans

Published on Jun 21
· Submitted by ffurfaro on Jun 24
Authors:

Abstract

TPTT enhances large language models with efficient linearized attention and advanced memory management, improving both efficiency and accuracy for long-context inference.

AI-generated summary

Recent advances in large language models (LLMs) have led to remarkable progress in natural language processing, but their computational and memory demands remain a significant challenge, particularly for long-context inference. We introduce TPTT (Transforming Pretrained Transformer into Titans), a novel framework for enhancing pretrained Transformer models with efficient linearized attention mechanisms and advanced memory management. TPTT employs techniques such as Memory as Gate (MaG) and mixed linearized attention (LiZA). It is fully compatible with the Hugging Face Transformers library, enabling seamless adaptation of any causal LLM through parameter-efficient fine-tuning (LoRA) without full retraining. We show the effectiveness of TPTT on the MMLU benchmark with models of approximately 1 billion parameters, observing substantial improvements in both efficiency and accuracy. For instance, Titans-Llama-3.2-1B achieves a 20% increase in Exact Match (EM) over its baseline. Statistical analyses and comparisons with recent state-of-the-art methods confirm the practical scalability and robustness of TPTT. Code is available at https://github.com/fabienfrfr/tptt . Python package at https://pypi.org/project/tptt/ .

Community

Sign up or log in to comment

Models citing this paper 4

Datasets citing this paper 0

No dataset linking this paper

Cite arxiv.org/abs/2506.17671 in a dataset README.md to link it from this page.

Spaces citing this paper 0

No Space linking this paper

Cite arxiv.org/abs/2506.17671 in a Space README.md to link it from this page.

Collections including this paper 0

No Collection including this paper

Add this paper to a collection to link it from this page.