Memory Transfer Learning: How Memories are Transferred Across Domains in Coding Agents

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Kim, Kangsan, Kang, Minki, Kim, Taeil, Yang, Yanlai, Ren, Mengye, Hwang, Sung Ju
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917410779430912
author Kim, Kangsan
Kang, Minki
Kim, Taeil
Yang, Yanlai
Ren, Mengye
Hwang, Sung Ju
author_facet Kim, Kangsan
Kang, Minki
Kim, Taeil
Yang, Yanlai
Ren, Mengye
Hwang, Sung Ju
contents Memory-based self-evolution has emerged as a promising paradigm for coding agents. However, existing approaches typically restrict memory utilization to homogeneous task domains, failing to leverage the shared infrastructural foundations, such as runtime environments and programming languages, that exist across diverse real-world coding problems. To address this limitation, we investigate \textbf{Memory Transfer Learning} (MTL) by harnessing a unified memory pool from heterogeneous domains. We evaluate performance across 6 coding benchmarks using four memory representations, ranging from concrete traces to abstract insights. Our experiments demonstrate that cross-domain memory improves average performance by 3.7\%, primarily by transferring meta-knowledge, such as validation routines, rather than task-specific code. Importantly, we find that abstraction dictates transferability; high-level insights generalize well, whereas low-level traces often induce negative transfer due to excessive specificity. Furthermore, we show that transfer effectiveness scales with the size of the memory pool, and memory can be transferred even between different models. Our work establishes empirical design principles for expanding memory utilization beyond single-domain silos. Project page: https://memorytransfer.github.io/
format Preprint
id arxiv_https___arxiv_org_abs_2604_14004
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Memory Transfer Learning: How Memories are Transferred Across Domains in Coding Agents
Kim, Kangsan
Kang, Minki
Kim, Taeil
Yang, Yanlai
Ren, Mengye
Hwang, Sung Ju
Artificial Intelligence
Computation and Language
Memory-based self-evolution has emerged as a promising paradigm for coding agents. However, existing approaches typically restrict memory utilization to homogeneous task domains, failing to leverage the shared infrastructural foundations, such as runtime environments and programming languages, that exist across diverse real-world coding problems. To address this limitation, we investigate \textbf{Memory Transfer Learning} (MTL) by harnessing a unified memory pool from heterogeneous domains. We evaluate performance across 6 coding benchmarks using four memory representations, ranging from concrete traces to abstract insights. Our experiments demonstrate that cross-domain memory improves average performance by 3.7\%, primarily by transferring meta-knowledge, such as validation routines, rather than task-specific code. Importantly, we find that abstraction dictates transferability; high-level insights generalize well, whereas low-level traces often induce negative transfer due to excessive specificity. Furthermore, we show that transfer effectiveness scales with the size of the memory pool, and memory can be transferred even between different models. Our work establishes empirical design principles for expanding memory utilization beyond single-domain silos. Project page: https://memorytransfer.github.io/
title Memory Transfer Learning: How Memories are Transferred Across Domains in Coding Agents
topic Artificial Intelligence
Computation and Language
url https://arxiv.org/abs/2604.14004