Papers › Retraining-Free Merging of Sparse MoE via Hierarchical Clustering

Retraining-Free Merging of Sparse MoE via Hierarchical Clustering

11 Oct 2024arXiv:2410.08589archive 2025-07-28

I-Chun Chen, Hsu-Shen Liu, Wei-Fang Sun, Chen-Hao Chao, Yen-Chang Hsu, Chun-Yi Lee

Sparse Mixture-of-Experts (SMoE) models represent a significant advancement in large language model (LLM) development through their efficient parameter utilization. These models achieve substantial performance improvements at reduced inference costs. However, the deployment of SMoE models faces constraints from extensive memory requirements of expert components in resource-limited environments. To address these limitations, this paper introduces Hierarchical Clustering for Sparsely activated Mixture of Experts (HC-SMoE), a task-agnostic expert merging framework for parameter reduction without retraining. HC-SMoE introduces a novel hierarchical clustering approach based on expert outputs to ensure merging robustness independent of routing decisions. The proposed output-based clustering method enables effective capture of functional relationships between experts for large-scale architectures. We provide theoretical analysis and comprehensive evaluations across multiple zero-shot language tasks to demonstrate HC-SMoE's effectiveness in state-of-the-art models including Qwen and Mixtral. The experimental results validate HC-SMoE's superior performance and practical applicability for real-world deployments.

PaperPDFCodeCode Syntology ran

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, Syntology's MCP tool lists every function and class Syntology harvested from this paper and whether it ran (how to connect): get_harvested_code_for_paper(arxiv_id="2410.08589")

Code

Syntology Ran 15 of 22 code samples harvested from 1 repository linked to this paper; 7 have no recorded run. Of those that ran: 3 ran · our draft was wrong; 4 ran · fixture could not drive it; 8 ran with no contract checked.

By repository: official repository: 22 samples from 1 repository, 15 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

wazenmai/hc-smoe officialmentioned in papermentioned on GitHubpytorchMIT report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

22 samples harvested; 15 ran; 0 honoured the contract we drafted; 7 have no recorded run. Read from Syntology's graph 2026-09-24; that is when this build read the record, not when the samples ran.

3ran · our draft was wrong
4ran · fixture could not drive it
8ran
7unverified

Licence: 0 of the 22 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from wazenmai/HC-SMoE. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself; “our draft was wrong” and “fixture could not drive it” are failures of Syntology's instrument, not of the code.

Each sample ends with its code_sha256, Syntology's identity for that exact code. An agent fetches the stored sample with Syntology's MCP tool get_code(code_sha256="…") (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples from a repository marked official are listed first. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of four reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; or the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar). Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

apply_mask wazenmai/HC-SMoE/hcsmoe/merging/grouping_mixtral.py official repository ran MIT (permissive) · 2663637656dd4380 · report
bhattacharyya_distance wazenmai/HC-SMoE/hcsmoe/merging/overlap.py official repository ran MIT (permissive) · 8af6d52439df5ac1 · report
compute_distance wazenmai/hc-smoe/hcsmoe/merging/clustering.py official repository ran · fixture could not drive it MIT (permissive) · 89eaaed986467ca4 · report
compute_silhouette_score wazenmai/hc-smoe/hcsmoe/merging/clustering.py official repository ran · our draft was wrong fingerprinted MIT (permissive) · 445a25a3378933c0 · report
compute_silhouette_score wazenmai/hc-smoe/hcsmoe/merging/clustering.py official repository ran · fixture could not drive it fingerprinted MIT (permissive) · 9e3b2b903c89aa39 · report
compute_switch_permutation_by_weight_matching wazenmai/HC-SMoE/hcsmoe/merging/permutation.py official repository ran MIT (permissive) · cd57bc0268739af9 · report
hijack wazenmai/HC-SMoE/hcsmoe/merging/grouping_mixtral.py official repository ran MIT (permissive) · 644aac920c458274 · report
hijack wazenmai/HC-SMoE/hcsmoe/merging/grouping_qwen.py official repository ran MIT (permissive) · c20ae1a60dcd8e85 · report
linkage_step wazenmai/hc-smoe/hcsmoe/merging/clustering.py official repository ran · fixture could not drive it MIT (permissive) · 4047127dc3316fbe · report
overlap_rate wazenmai/HC-SMoE/hcsmoe/merging/overlap.py official repository ran fingerprinted MIT (permissive) · 09fbe56b373956ff · report
overlap_rate_bhattacharyya wazenmai/HC-SMoE/hcsmoe/merging/overlap.py official repository ran fingerprinted MIT (permissive) · ff29e60e8a9b900a · report
pairwise_distances wazenmai/hc-smoe/hcsmoe/merging/clustering.py official repository ran · fixture could not drive it fingerprinted MIT (permissive) · 881d32a340b4337a · report
permute_switch_mlp_dense_expert_ wazenmai/HC-SMoE/hcsmoe/merging/permutation.py official repository ran MIT (permissive) · ed870812c5e5c849 · report
remove_col wazenmai/HC-SMoE/hcsmoe/merging/grouping_mixtral.py official repository ran · our draft was wrong fingerprinted MIT (permissive) · ac28dcec5c8aa992 · report
safe_average wazenmai/hc-smoe/hcsmoe/merging/clustering.py official repository ran · our draft was wrong fingerprinted MIT (permissive) · d7c56984b5555b6e · report
align_switch_permutation_for_all_experts_by_weight_matching wazenmai/HC-SMoE/hcsmoe/merging/permutation.py official repository unverified MIT (permissive) · 5f5c679efbd91d5c · report
evaluate_minipile_perplexity wazenmai/HC-SMoE/hcsmoe/evaluation/minipile.py official repository unverified MIT (permissive) · e4b8bffda1c924f8 · report
generate_random_group_labels wazenmai/HC-SMoE/hcsmoe/merging/utils.py official repository unverified MIT (permissive) · 2a3e7e560f218cc4 · report
get_minipile_dataloder wazenmai/HC-SMoE/hcsmoe/evaluation/minipile.py official repository unverified MIT (permissive) · 2ce9c4afb75f7a42 · report
group_experts_by_clustering wazenmai/hc-smoe/hcsmoe/merging/clustering.py official repository unverified MIT (permissive) · 41388c0799d06a0b · report
hierarchical_clustering_dynamic wazenmai/hc-smoe/hcsmoe/merging/clustering.py official repository unverified MIT (permissive) · 83e8336dcb695d24 · report
safe_average wazenmai/HC-SMoE/hcsmoe/merging/clustering.py official repository unverified MIT (permissive) · 1ea84b1496629450 · report

Tasks

ClusteringLanguage ModelingLanguage ModellingLarge Language ModelMixture-of-Experts

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

SET

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections