SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2408.03092 · 2024

Extend Model Merging from Fine-Tuned to Pre-Trained Large Language Models via Weight Disentanglement

arXiv · PDF · Open in the Atlas

Code that ran

5 of 5 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 1 violated it and 0 have no contract verdict; 4 failures of our instrument; 0 no recorded run); 1 of them was also run on the host, not in the jail, by the peer-agreement lane that set its level, measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2408.03092.json · badge: /api/ran/2408.03092.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
yule-BUAA/MergeLLM canonical (classifier label)
classifier basis: census
5 ran · 0 no recorded run
FunctionStatusWhere it lives
is_number Ran
ran, violated the contract we drafted
yule-BUAA/MergeLLM/utils/evaluate_llms_utils.py
pointer only (licence: NONE) · get_code("0fe9cf6c00ef56ea")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, yule-BUAA/MergeLLM when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

batch_data Ran
ran; our contract draft was wrong, not the code
yule-BUAA/MergeLLM/utils/evaluate_llms_utils.py
pointer only (licence: NONE); the stored copy get_code serves, from Rambo-Yi/TLora, is pointer only (licence: NONE) · get_code("d49412b3a690206e")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, yule-BUAA/MergeLLM when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

get_modules_to_merge Ran
ran; our contract draft was wrong, not the code
yule-BUAA/MergeLLM/utils/utils.py
pointer only (licence: NONE) · get_code("3192d7b16d484b87")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, yule-BUAA/MergeLLM when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

get_param_names_to_merge Ran
ran; our contract draft was wrong, not the code
yule-BUAA/MergeLLM/utils/utils.py
pointer only (licence: NONE); the stored copy get_code serves, from walkerworldpeace/mllmerging, is pointer only (licence: NONE) · get_code("dcafd602728bc51e")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, yule-BUAA/MergeLLM when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

mask_input_with_mask_rate Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
yule-BUAA/MergeLLM/model_merging_methods/mask_weights_utils.py
pointer only (licence: NONE) · get_code("99462e4ad36a59db")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, yule-BUAA/MergeLLM when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

Also present in other papers' code

Of 5 samples lifted from this paper's linked repositories, 5 carry a function identity and 0 do not. Of the 5 distinct functions with an identity, 4 are also present in other papers whose code we hold (4 with at least one paper dated earlier, 1 the same month, 3 later) and 1 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

FunctionAlso present inEarliest-datedOthers named
is_number 18 other papers
13 dated earlier, 1 the same month, 4 later
1807.06756 (2018-07)
SySeVR: A Framework for Using Deep Learning to Detect Software Vulnerabilities
1906.02425 (2019-06)
Uncertainty-guided Continual Learning with Bayesian Neural Networks
2005.00558 (2020-05)
POINTER: Constrained Progressive Text Generation via Insertion-based Generative Pre-training
and 15 more not named here
batch_data 6 other papers
3 dated earlier, 3 later
2311.03099 (2023-11)
Language Models are Super Mario: Absorbing Abilities from Homologous Models as a Free Lunch
2406.09044 (2024-06)
MiLoRA: Harnessing Minor Singular Components for Parameter-Efficient LLM Finetuning
2406.11617 (2024-06)
DELLA-Merging: Reducing Interference in Model Merging through Magnitude-Based Sampling
and 3 more not named here
get_param_names_to_merge 4 other papers
1 dated earlier, 3 later
2212.04089 (2022-12)
Editing Models with Task Arithmetic
2503.22178 (2025-03)
AdaRank: Adaptive Rank Pruning for Enhanced Model Merging
2505.19892 (2025-05)
OptMerge: Unifying Multimodal LLM Capabilities and Modalities via Model Merging
and 1 more not named here
mask_input_with_mask_rate 2 other papers
2 dated earlier
2311.03099 (2023-11)
Language Models are Super Mario: Absorbing Abilities from Homologous Models as a Free Lunch
2405.17461 (2024-05)
EMR-MERGING: Tuning-Free High-Performance Model Merging

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Merging Large Language Models (LLMs) aims to amalgamate multiple homologous LLMs into one with all the capabilities. Ideally, any LLMs sharing the same backbone should be mergeable, irrespective of whether they are Fine-Tuned (FT) with minor parameter changes or Pre-Trained (PT) with substantial parameter shifts. However, existing methods often manually assign the model importance, rendering them feasible only for LLMs with similar parameter alterations, such as multiple FT LLMs. The diverse parameter changed ranges between FT and PT LLMs pose challenges for current solutions in empirically determining the optimal combination. In this paper, we make a pioneering effort to broaden the applicability of merging techniques from FT to PT LLMs. We initially examine the efficacy of current methods in merging FT and PT LLMs, discovering that they struggle to deal with PT LLMs. Subsequently, we introduce an approach based on WeIght DisENtanglement (WIDEN) to effectively extend the merging scope, which first disentangles model weights into magnitude and direction components, and then performs adaptive fusion by considering their respective contributions. In the experiments, we merge Qwen1.5-Chat (an FT LLM with instruction-following skills) with Sailor (a PT LLM with multilingual abilities) across 7B and 14B model scales. Results reveal that: (1) existing solutions usually fail when merging Sailor, either losing both abilities or only retaining instruction-following skills; (2) WIDEN successfully injects the multilingual abilities of Sailor into Qwen1.5-Chat and make it proficient in Southeast Asian languages, achieving enhancements in the fundamental capabilities. In light of previous research, we also merge multiple 13B FT LLMs and observe that WIDEN achieves a balanced amalgamation of instruction following, mathematical reasoning, and code generation skills.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2408.03092")
get_code_for_paper("2408.03092")
have("2408.03092")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2408.03092.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2408.03092.svg)](https://syntology.ai/paper/2408.03092)

Connect an agent — have() is free.