Papers › Merging Experts into One: Improving Computational Efficiency of Mixture of Experts

Merging Experts into One: Improving Computational Efficiency of Mixture of Experts

15 Oct 2023arXiv:2310.09832archive 2025-07-28

Shwai He, Run-Ze Fan, Liang Ding, Li Shen, Tianyi Zhou, DaCheng Tao

Scaling the size of language models usually leads to remarkable advancements in NLP tasks. But it often comes with a price of growing computational cost. Although a sparse Mixture of Experts (MoE) can reduce the cost by activating a small subset of parameters (e.g., one expert) for each input, its computation escalates significantly if increasing the number of activated experts, limiting its practical utility. Can we retain the advantages of adding more experts without substantially increasing the computational costs? In this paper, we first demonstrate the superiority of selecting multiple experts and then propose a computation-efficient approach called \textbf{\texttt{Merging Experts into One}} (MEO), which reduces the computation cost to that of a single expert. Extensive experiments show that MEO significantly improves computational efficiency, e.g., FLOPS drops from 72.0G of vanilla MoE to 28.6G (MEO). Moreover, we propose a token-level attention block that further enhances the efficiency and performance of token-level MEO, e.g., 83.3\% (MEO) vs. 82.6\% (vanilla MoE) average score on the GLUE benchmark. Our code will be released upon acceptance. Code will be released at: \url{https://github.com/Shwai-He/MEO}.

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="2310.09832")

Code

Syntology Ran 16 of 21 code samples harvested from 1 repository linked to this paper; 5 have no recorded run. Of those that ran: 1 ran · honoured contract; 3 ran · our draft was wrong; 12 ran with no contract checked.

By repository: official repository: 21 samples from 1 repository, 16 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

shwai-he/meo officialmentioned in paperpytorch 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

21 samples harvested; 16 ran; 1 honoured the contract we drafted; 5 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.

1ran · honoured contract
3ran · our draft was wrong
12ran
5unverified

Licence: 21 of the 21 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 Shwai-He/MEO. “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.

MEO Shwai-He/MEO/transformers/models/layers.py official repository ran · metamorphic tier: deterministic no licence file found · pointer only · 0766417d0a9cc795 · report
TokenAtt Shwai-He/MEO/transformers/models/layers.py official repository ran · metamorphic tier: deterministic no licence file found · pointer only · 721cc7eedfa0ef65 · report
_gates_to_load Shwai-He/MEO/transformers/models/layers.py official repository ran · honoured contract fingerprinted no licence file found · pointer only · bd49c412803e1b6e · report
_prob_in_top_k Shwai-He/MEO/transformers/models/layers.py official repository ran · our draft was wrong no licence file found · pointer only · a089868d27926865 · report
booleans_processing shwai-he/meo/transformers/modeling_tf_utils.py official repository ran no licence file found · pointer only · 743ceb9025a7955d · report
convert_pytorch_state_dict_to_flax shwai-he/meo/transformers/modeling_flax_pytorch_utils.py official repository ran no licence file found · pointer only · 5168a172bbb363a5 · report
cv_squared Shwai-He/MEO/transformers/models/layers.py official repository ran · our draft was wrong fingerprinted no licence file found · pointer only · abbff004073dc773 · report
dummy_loss shwai-he/meo/transformers/modeling_tf_utils.py official repository ran no licence file found · pointer only · 3c1eb64dde6b86af · report
generate_batch_splits shwai-he/meo/tasks/language-modeling/run_bart_dlm.py official repository ran no licence file found · pointer only · 26eb4e2856b3d7a1 · report
get_imports shwai-he/meo/transformers/dynamic_module_utils.py official repository ran no licence file found · pointer only · ccaa0edba373622c · report
infer_metric_tags_from_eval_results shwai-he/meo/transformers/modelcard.py official repository ran no licence file found · pointer only · dd8f7751ef859581 · report
keras_serializable shwai-he/meo/transformers/modeling_tf_utils.py official repository ran no licence file found · pointer only · 6a3959a0ffd36577 · report
noisy_top_k_gating Shwai-He/MEO/transformers/models/layers.py official repository ran · our draft was wrong no licence file found · pointer only · 626cbdc393d06c43 · report
parse_keras_history shwai-he/meo/transformers/modelcard.py official repository ran no licence file found · pointer only · a4491b45d19b7a39 · report
quick_gelu shwai-he/meo/transformers/modeling_flax_utils.py official repository ran fingerprinted no licence file found · pointer only · a66d2b78670c85e1 · report
rename_key_and_reshape_tensor shwai-he/meo/transformers/modeling_flax_pytorch_utils.py official repository ran no licence file found · pointer only · 61779b0406a59cfc · report
dtype_byte_size shwai-he/meo/transformers/modeling_flax_utils.py official repository unverified no licence file found · pointer only · 9d98806acb3144b7 · report
flax_shard_checkpoint shwai-he/meo/transformers/modeling_flax_utils.py official repository unverified no licence file found · pointer only · e3ce70ca9258800e · report
get_relative_import_files shwai-he/meo/transformers/dynamic_module_utils.py official repository unverified no licence file found · pointer only · 00cc853a73f3aeb6 · report
get_relative_imports shwai-he/meo/transformers/dynamic_module_utils.py official repository unverified no licence file found · pointer only · 450aeffe7f80f917 · report
is_hf_dataset shwai-he/meo/transformers/modelcard.py official repository unverified no licence file found · pointer only · ec7b1528a936234d · report

Tasks

Computational EfficiencyMixture-of-Experts

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

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