Access Sets Matter: Budgeting Expert Reads for Scalable Weight-Space Model Merging

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Wang, Yuanyi, Gu, Yanggan, Lu, Su, Yang, Yifan, Yan, Zhaoyi, Xie, Congkai, Wu, Jianmin, Yang, Hongxia
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866913169825333248
author Wang, Yuanyi
Gu, Yanggan
Lu, Su
Yang, Yifan
Yan, Zhaoyi
Xie, Congkai
Wu, Jianmin
Yang, Hongxia
author_facet Wang, Yuanyi
Gu, Yanggan
Lu, Su
Yang, Yifan
Yan, Zhaoyi
Xie, Congkai
Wu, Jianmin
Yang, Hongxia
contents Weight-space model merging is usually formulated as an algebraic operation on checkpoints, yet at LLM scale the limiting resource is often the set of expert weights that must be read. We introduce MergePipe, a budget-aware execution layer that casts LLM merging as an \emph{expert access-set} problem: given a merge operator and a checkpoint family in a shared weight coordinate system, choose which expert delta blocks to access under an explicit I/O budget. MergePipe indexes parameter blocks, builds deterministic access plans, and executes the induced budgeted merge with replayable manifests. The plan is budget-sound by construction and recovers the full-read merge at full budget; for fixed-coefficient additive operators, the omitted-update error is bounded by the norm of omitted deltas. Across Qwen and Llama merging workloads, MergePipe reduces expert-read I/O by up to an order of magnitude and achieves up to $11\times$ speedups. Representative budget sweeps show $O(10^{-3})$ parameter deviation from full-read merges and no monotonic degradation on downstream benchmarks.
format Preprint
id arxiv_https___arxiv_org_abs_2605_29489
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Access Sets Matter: Budgeting Expert Reads for Scalable Weight-Space Model Merging
Wang, Yuanyi
Gu, Yanggan
Lu, Su
Yang, Yifan
Yan, Zhaoyi
Xie, Congkai
Wu, Jianmin
Yang, Hongxia
Machine Learning
Systems and Control
Weight-space model merging is usually formulated as an algebraic operation on checkpoints, yet at LLM scale the limiting resource is often the set of expert weights that must be read. We introduce MergePipe, a budget-aware execution layer that casts LLM merging as an \emph{expert access-set} problem: given a merge operator and a checkpoint family in a shared weight coordinate system, choose which expert delta blocks to access under an explicit I/O budget. MergePipe indexes parameter blocks, builds deterministic access plans, and executes the induced budgeted merge with replayable manifests. The plan is budget-sound by construction and recovers the full-read merge at full budget; for fixed-coefficient additive operators, the omitted-update error is bounded by the norm of omitted deltas. Across Qwen and Llama merging workloads, MergePipe reduces expert-read I/O by up to an order of magnitude and achieves up to $11\times$ speedups. Representative budget sweeps show $O(10^{-3})$ parameter deviation from full-read merges and no monotonic degradation on downstream benchmarks.
title Access Sets Matter: Budgeting Expert Reads for Scalable Weight-Space Model Merging
topic Machine Learning
Systems and Control
url https://arxiv.org/abs/2605.29489