Cross-Model Memory Transfer via Target-Side Reader Adaptation

Hugging Face Daily Papers Papers

Summary

This paper investigates cross-model memory transfer, demonstrating that target-side reader adaptation is crucial for utilizing frozen memory tables across different large language models. It introduces methods for reusable knowledge artifacts with optional adaptation to improve alignment.

Methods for improving knowledge use in large language models typically fall into two regimes. Non-parametric retrieval offers flexible access to external knowledge, but adds retrieval latency, context overhead, and only shallow integration with the backbone. Parametric adaptation is efficient at inference time, but entangles knowledge with model weights and can be hard to update, audit, or transfer. Engram-style hashed memory occupies a middle regime: it stores learned information in an external, addressable table, yet consumes that table through a small learned reader. This raises a basic question: when such a memory is moved across backbones, what matters more, the frozen memory itself or the target-side reader? We study this question through cross-model frozen-memory extraction, in which a memory trained on a source model is frozen and attached to a different target model, with only a lightweight reader trained. Ablations show that learned memory content and correct addressing both matter, but the transferred table becomes useful only through a reader aligned to the target model. In downstream question answering tasks, a dual-layer, four-branch reader nearly closes the gap between same-model and cross-model reuse, achieving an average score of 38.8 under our controlled evaluation protocol. Moreover, when the provider reader is directly compatible with the target interface, the frozen artifact can provide substantial utility without target-side training, while optional reader adaptation yields further improvement. These results suggest that Engram can serve as a reusable external knowledge artifact, provided that the target has access to a compatible reader interface; target-side adaptation can further improve alignment when direct reader reuse is insufficient.
Original Article
View Cached Full Text

Cached at: 08/19/26, 07:56 AM

Paper page - Cross-Model Memory Transfer via Target-Side Reader Adaptation

Source: https://huggingface.co/papers/2608.17050

Abstract

Cross-model reuse of frozen external memory tables depends primarily on aligning a lightweight target-side reader rather than the memory content alone, enabling reusable knowledge artifacts with optional adaptation.

Methods for improving knowledge use in large language models typically fall into two regimes. Non-parametric retrieval offers flexible access to external knowledge, but adds retrieval latency, context overhead, and only shallow integration with the backbone. Parametric adaptation is efficient at inference time, but entangles knowledge with model weights and can be hard to update, audit, or transfer.Engram-stylehashed memoryoccupies a middle regime: it stores learned information in an external, addressable table, yet consumes that table through a small learned reader. This raises a basic question: when such a memory is moved across backbones, what matters more, the frozen memory itself or the target-side reader? We study this question throughcross-model frozen-memory extraction, in which a memory trained on a source model is frozen and attached to a different target model, with only a lightweight reader trained. Ablations show that learned memory content and correct addressing both matter, but the transferred table becomes useful only through a reader aligned to the target model. In downstream question answering tasks, a dual-layer, four-branch reader nearly closes the gap between same-model and cross-model reuse, achieving an average score of 38.8 under our controlled evaluation protocol. Moreover, when the provider reader is directly compatible with the target interface, the frozen artifact can provide substantial utility without target-side training, while optional reader adaptation yields further improvement. These results suggest thatEngramcan serve as a reusable external knowledge artifact, provided that the target has access to a compatible reader interface; target-side adaptation can further improve alignment when direct reader reuse is insufficient.

View arXiv pageView PDFProject pageGitHubAdd to collection

Get this paper in your agent:

hf papers read 2608\.17050

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2608.17050 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2608.17050 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2608.17050 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.

Similar Articles

@vintcessun: A 'knowledge hard drive' that can be plugged across models — the difficulty lies not in moving the memory table, but in configuring the read head for the target model. https://arxiv.org/abs/2608.17050 The paper first trains Engram hash memory with the source model, then freezes the memory and target backbone, only training the reader to complete addressing...

X AI KOLs Timeline

This paper proposes a method for cross-model memory transfer through target-side reader adaptation, using Engram hash memory and a lightweight reader, achieving 38.8 on QA tasks, and applicable to Agent memory updates and audits. The limitation is that it was only tested up to 9B models, with scaling laws unknown.

Understanding Is Done Early: A Depth Division of Labor in Large Language Models and Its Use for Unbounded-Context Memory

arXiv cs.CL

This paper introduces CoMem, a method that exploits the depth-wise division of labor in LLMs to cache intermediate residual tensors and recompute only upper layers for retrieval, enabling bounded read compute and memory independent of stored-context length. Evaluated on Qwen3-8B, CoMem achieves strong long-context performance with significant memory savings and prefill speedups.