SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2310.09832 · EMNLP · 2023

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

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

arXiv · PDF · Open in the Atlas

Code that ran

18 of 21 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (2 of them constructed an object rather than computing a result; 2 honoured the contract we drafted, 0 violated it and 2 have no contract verdict; 14 failures of our instrument; 3 no recorded run), 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/2310.09832.json · badge: /api/ran/2310.09832.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
shwai-he/meo canonical (classifier label)
classifier basis: census
12 ran · 3 no recorded run
Shwai-He/MEO community
Papers with Code archive
6 ran · 0 no recorded run
FunctionStatusWhere it lives
MEO Ran
ran on a synthesized input · constructed an object; no result was computed
Shwai-He/MEO/transformers/models/layers.py
pointer only (licence: NONE) · get_code("0766417d0a9cc795")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Shwai-He/MEO 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.

TokenAtt Ran
ran on a synthesized input · constructed an object; no result was computed
Shwai-He/MEO/transformers/models/layers.py
pointer only (licence: NONE) · get_code("721cc7eedfa0ef65")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Shwai-He/MEO 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.

_gates_to_load Ran
ran, honoured the contract we drafted
Shwai-He/MEO/transformers/models/layers.py
pointer only (licence: NONE) · get_code("bd49c412803e1b6e")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Shwai-He/MEO 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.

_prob_in_top_k Ran
ran; our contract draft was wrong, not the code
Shwai-He/MEO/transformers/models/layers.py
pointer only (licence: NONE) · get_code("a089868d27926865")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Shwai-He/MEO 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.

booleans_processing Ran
ran; our contract draft was wrong, not the code
shwai-he/meo/transformers/modeling_tf_utils.py
pointer only (licence: NONE); the stored copy get_code serves, from clin0212/hydralora, is pointer only (licence: NONE) · get_code("743ceb9025a7955d")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, shwai-he/meo 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.

convert_pytorch_state_dict_to_flax Ran
ran; our contract draft was wrong, not the code
shwai-he/meo/transformers/modeling_flax_pytorch_utils.py
pointer only (licence: NONE) · get_code("5168a172bbb363a5")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, shwai-he/meo 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.

cv_squared Ran
ran; our contract draft was wrong, not the code
Shwai-He/MEO/transformers/models/layers.py
pointer only (licence: NONE) · get_code("abbff004073dc773")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Shwai-He/MEO 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.

dummy_loss Ran
ran; our contract draft was wrong, not the code
shwai-he/meo/transformers/modeling_tf_utils.py
pointer only (licence: NONE); the stored copy get_code serves, from clin0212/hydralora, is pointer only (licence: NONE) · get_code("3c1eb64dde6b86af")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, shwai-he/meo 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.

generate_batch_splits Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
shwai-he/meo/tasks/language-modeling/run_bart_dlm.py
pointer only (licence: NONE) · get_code("26eb4e2856b3d7a1")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, shwai-he/meo 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_imports Ran
ran; our contract draft was wrong, not the code
shwai-he/meo/transformers/dynamic_module_utils.py
pointer only (licence: NONE); the stored copy get_code serves, from clin0212/hydralora, is pointer only (licence: NONE) · get_code("ccaa0edba373622c")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, shwai-he/meo 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_relative_import_files Ran
ran; our contract draft was wrong, not the code
shwai-he/meo/transformers/dynamic_module_utils.py
pointer only (licence: NONE) · get_code("00cc853a73f3aeb6")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, shwai-he/meo 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_relative_imports Ran
ran; our contract draft was wrong, not the code
shwai-he/meo/transformers/dynamic_module_utils.py
pointer only (licence: NONE) · get_code("450aeffe7f80f917")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, shwai-he/meo 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.

infer_metric_tags_from_eval_results Ran
ran; our contract draft was wrong, not the code
shwai-he/meo/transformers/modelcard.py
pointer only (licence: NONE); the stored copy get_code serves, from shen8424/CORE, is pointer only (licence: NONE) · get_code("dd8f7751ef859581")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, shwai-he/meo 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.

keras_serializable Ran
ran; our contract draft was wrong, not the code
shwai-he/meo/transformers/modeling_tf_utils.py
pointer only (licence: NONE); the stored copy get_code serves, from clin0212/hydralora, is pointer only (licence: NONE) · get_code("6a3959a0ffd36577")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, shwai-he/meo 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.

noisy_top_k_gating Ran
ran; our contract draft was wrong, not the code
Shwai-He/MEO/transformers/models/layers.py
pointer only (licence: NONE) · get_code("626cbdc393d06c43")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Shwai-He/MEO 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.

parse_keras_history Ran
ran; our contract draft was wrong, not the code
shwai-he/meo/transformers/modelcard.py
pointer only (licence: NONE) · get_code("a4491b45d19b7a39")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, shwai-he/meo 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.

quick_gelu Ran
ran, honoured the contract we drafted
shwai-he/meo/transformers/modeling_flax_utils.py
pointer only (licence: NONE); the stored copy get_code serves, from shen8424/CORE, is pointer only (licence: NONE) · get_code("a66d2b78670c85e1")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, shwai-he/meo 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.

rename_key_and_reshape_tensor Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
shwai-he/meo/transformers/modeling_flax_pytorch_utils.py
pointer only (licence: NONE) · get_code("61779b0406a59cfc")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, shwai-he/meo 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.

dtype_byte_size Not yet run
unverified (harvested, no recorded run)
shwai-he/meo/transformers/modeling_flax_utils.py
pointer only (licence: NONE) · get_code("9d98806acb3144b7")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, shwai-he/meo 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. No invocation recorded.

flax_shard_checkpoint Not yet run
unverified (harvested, no recorded run)
shwai-he/meo/transformers/modeling_flax_utils.py
pointer only (licence: NONE) · get_code("e3ce70ca9258800e")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, shwai-he/meo 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. No invocation recorded.

is_hf_dataset Not yet run
unverified (harvested, no recorded run)
shwai-he/meo/transformers/modelcard.py
pointer only (licence: NONE); the stored copy get_code serves, from shen8424/CORE, is pointer only (licence: NONE) · get_code("ec7b1528a936234d")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, shwai-he/meo 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. No invocation recorded.

Also present in other papers' code

Of 21 samples lifted from this paper's linked repositories, 21 carry a function identity and 0 do not. Of the 21 distinct functions with an identity, 19 are also present in other papers whose code we hold (8 with at least one paper dated earlier, 13 the same month, 18 later) and 2 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
infer_metric_tags_from_eval_results 19 other papers
2 dated earlier, 2 the same month, 15 later
2022.findings-emnlp.427 (2022 (year only))
2305.15756 (2023-05)
UniTRec: A Unified Text-to-Text Transformer and Joint Contrastive Learning Framework for Text-based Recommendation
2310.05107 (2023-10)
OV-PARTS: Towards Open-Vocabulary Part Segmentation
and 16 more not named here
get_relative_import_files 18 other papers
5 dated earlier, 2 the same month, 11 later
2211.11694 (2022-11)
Exploring Discrete Diffusion Models for Image Captioning
2304.05568 (2023-04)
Improving Diffusion Models for Scene Text Editing with Dual Encoders
2305.15756 (2023-05)
UniTRec: A Unified Text-to-Text Transformer and Joint Contrastive Learning Framework for Text-based Recommendation
and 15 more not named here
get_relative_imports 18 other papers
5 dated earlier, 2 the same month, 11 later
2211.11694 (2022-11)
Exploring Discrete Diffusion Models for Image Captioning
2304.05568 (2023-04)
Improving Diffusion Models for Scene Text Editing with Dual Encoders
2305.15756 (2023-05)
UniTRec: A Unified Text-to-Text Transformer and Joint Contrastive Learning Framework for Text-based Recommendation
and 15 more not named here
quick_gelu 17 other papers
3 dated earlier, 2 the same month, 12 later
2022.findings-emnlp.427 (2022 (year only))
2303.14582 (2023-03)
Identification of Negative Transfers in Multitask Learning Using Surrogate Models
2305.15756 (2023-05)
UniTRec: A Unified Text-to-Text Transformer and Joint Contrastive Learning Framework for Text-based Recommendation
and 14 more not named here
is_hf_dataset 16 other papers
1 the same month, 15 later
2310.05107 (2023-10)
OV-PARTS: Towards Open-Vocabulary Part Segmentation
2312.09979 (2023-12)
LoRAMoE: Alleviate World Knowledge Forgetting in Large Language Models via MoE-Style Plugin
2402.05827 (2024-02)
On the Robustness of Editing Large Language Models
and 13 more not named here
parse_keras_history 16 other papers
1 the same month, 15 later
2310.05107 (2023-10)
OV-PARTS: Towards Open-Vocabulary Part Segmentation
2312.09979 (2023-12)
LoRAMoE: Alleviate World Knowledge Forgetting in Large Language Models via MoE-Style Plugin
2402.05827 (2024-02)
On the Robustness of Editing Large Language Models
and 13 more not named here
flax_shard_checkpoint 14 other papers
1 dated earlier, 2 the same month, 11 later
2305.15756 (2023-05)
UniTRec: A Unified Text-to-Text Transformer and Joint Contrastive Learning Framework for Text-based Recommendation
2310.05107 (2023-10)
OV-PARTS: Towards Open-Vocabulary Part Segmentation
2310.18738 (2023-10)
TLM: Token-Level Masking for Transformers
and 11 more not named here
dtype_byte_size 12 other papers
1 dated earlier, 2 the same month, 9 later
2305.15756 (2023-05)
UniTRec: A Unified Text-to-Text Transformer and Joint Contrastive Learning Framework for Text-based Recommendation
2310.05107 (2023-10)
OV-PARTS: Towards Open-Vocabulary Part Segmentation
2310.18738 (2023-10)
TLM: Token-Level Masking for Transformers
and 9 more not named here
keras_serializable 8 other papers
2 dated earlier, 2 the same month, 4 later
2109.03808 (2021-09)
Smelting Gold and Silver for Improved Multilingual AMR-to-Text Generation
2303.14582 (2023-03)
Identification of Negative Transfers in Multitask Learning Using Surrogate Models
2310.05107 (2023-10)
OV-PARTS: Towards Open-Vocabulary Part Segmentation
and 5 more not named here
booleans_processing 6 other papers
2 the same month, 4 later
2310.05107 (2023-10)
OV-PARTS: Towards Open-Vocabulary Part Segmentation
2310.18738 (2023-10)
TLM: Token-Level Masking for Transformers
2402.12656 (2024-02)
HyperMoE: Towards Better Mixture of Experts via Transferring Among Experts
and 3 more not named here
convert_pytorch_state_dict_to_flax 6 other papers
1 the same month, 5 later
2310.05107 (2023-10)
OV-PARTS: Towards Open-Vocabulary Part Segmentation
2312.09979 (2023-12)
LoRAMoE: Alleviate World Knowledge Forgetting in Large Language Models via MoE-Style Plugin
2402.05827 (2024-02)
On the Robustness of Editing Large Language Models
and 3 more not named here
dummy_loss 6 other papers
2 the same month, 4 later
2310.05107 (2023-10)
OV-PARTS: Towards Open-Vocabulary Part Segmentation
2310.18738 (2023-10)
TLM: Token-Level Masking for Transformers
2402.12656 (2024-02)
HyperMoE: Towards Better Mixture of Experts via Transferring Among Experts
and 3 more not named here
rename_key_and_reshape_tensor 6 other papers
1 the same month, 5 later
2310.05107 (2023-10)
OV-PARTS: Towards Open-Vocabulary Part Segmentation
2312.09979 (2023-12)
LoRAMoE: Alleviate World Knowledge Forgetting in Large Language Models via MoE-Style Plugin
2402.05827 (2024-02)
On the Robustness of Editing Large Language Models
and 3 more not named here
get_imports 3 other papers
3 later
2312.09979 (2023-12)
LoRAMoE: Alleviate World Knowledge Forgetting in Large Language Models via MoE-Style Plugin
2402.12656 (2024-02)
HyperMoE: Towards Better Mixture of Experts via Transferring Among Experts
2404.19245 (2024-04)
HydraLoRA: An Asymmetric LoRA Architecture for Efficient Fine-Tuning
_gates_to_load
brick of this paper's assembly: helper
2 other papers
2 later · 2 only as an assembly brick
2402.12656 (2024-02)
HyperMoE: Towards Better Mixture of Experts via Transferring Among Experts
2502.18821 (2025-02)
CAMEx: Curvature-aware Merging of Experts
cv_squared
brick of this paper's assembly: metric
2 other papers
2 later · 2 only as an assembly brick
2402.12656 (2024-02)
HyperMoE: Towards Better Mixture of Experts via Transferring Among Experts
2502.18821 (2025-02)
CAMEx: Curvature-aware Merging of Experts
_prob_in_top_k
brick of this paper's assembly: helper
1 other paper
1 later · 1 only as an assembly brick
2402.12656 (2024-02)
HyperMoE: Towards Better Mixture of Experts via Transferring Among Experts
generate_batch_splits 1 other paper
1 dated earlier
2109.10686 (2021-09)
Scale Efficiently: Insights from Pre-training and Fine-tuning Transformers
noisy_top_k_gating
brick of this paper's assembly: helper
1 other paper
1 later
2502.18821 (2025-02)
CAMEx: Curvature-aware Merging of Experts

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

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 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 28.9G (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: https://github.com/Shwai-He/MEO.

For agents

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

get_harvested_code_for_paper("2310.09832")
get_code_for_paper("2310.09832")
have("2310.09832")

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

curl https://syntology.ai/api/ran/2310.09832.json

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

[![Syntology run record](https://syntology.ai/api/ran/2310.09832.svg)](https://syntology.ai/paper/2310.09832)

Connect an agent — have() is free.