SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2604.04356 · 2026

REAM: Merging Improves Pruning of Experts in LLMs

Boris Knyazev, Ali Parviz, Ali Pasand, Saurav Jha, Maryam Hashemzadeh, Min-Joong Lee

arXiv · PDF · Open in the Atlas

Code that ran

20 of 20 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; 1 honoured the contract we drafted, 2 violated it and 0 have no contract verdict; 17 failures of our instrument; 0 no recorded run), measured 2026-11-22; 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/2604.04356.json · badge: /api/ran/2604.04356.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
zai-org/glm-simple-evals canonical (classifier label)
classifier basis: census
20 ran · 0 no recorded run
FunctionStatusWhere it lives
are_equal_under_sympy Ran
ran, violated the contract we drafted
zai-org/glm-simple-evals/evals/grading/grader.py
pointer only (licence: NONE) · get_code("bc7ab5cea18531ef")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zai-org/glm-simple-evals 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.

count_unknown_letters_in_expr Ran
ran, honoured the contract we drafted
zai-org/glm-simple-evals/evals/grading/grader.py
pointer only (licence: NONE) · get_code("368d3db6347269c3")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zai-org/glm-simple-evals 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.

estimate_pass_at_k Ran
ran; our contract draft was wrong, not the code
zai-org/glm-simple-evals/evals/livecodebench_eval.py
pointer only (licence: NONE) · get_code("9df10eb02582b09f")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zai-org/glm-simple-evals 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.

extract_boxed_answer Ran
ran; our contract draft was wrong, not the code
zai-org/glm-simple-evals/evals/math_eval.py
pointer only (licence: NONE) · get_code("09b470ed75dcda56")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zai-org/glm-simple-evals 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.

extract_function_name Ran
ran; our contract draft was wrong, not the code
zai-org/glm-simple-evals/evals/scicode_eval.py
pointer only (licence: NONE) · get_code("235feea5d71e36b7")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zai-org/glm-simple-evals 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.

extract_python_script Ran
ran; our contract draft was wrong, not the code
zai-org/glm-simple-evals/evals/scicode_eval.py
pointer only (licence: NONE) · get_code("15619befbc78e624")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zai-org/glm-simple-evals 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.

format_example Ran
ran; our contract draft was wrong, not the code
zai-org/glm-simple-evals/evals/mmlu_pro_eval.py
pointer only (licence: NONE) · get_code("24ff49e5a1d92207")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zai-org/glm-simple-evals 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.

format_judge Ran
ran; our contract draft was wrong, not the code
zai-org/glm-simple-evals/evals/hle_eval.py
pointer only (licence: NONE) · get_code("6ec9d0ddf8275c25")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zai-org/glm-simple-evals 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.

format_message Ran
ran; our contract draft was wrong, not the code
zai-org/glm-simple-evals/evals/hle_eval.py
pointer only (licence: NONE) · get_code("ba6ce7a502b69fbe")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zai-org/glm-simple-evals 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.

format_question Ran
ran; our contract draft was wrong, not the code
zai-org/glm-simple-evals/evals/gpqa_eval.py
pointer only (licence: NONE) · get_code("aaef2696f445a0a2")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zai-org/glm-simple-evals 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_random_string Ran
ran; our contract draft was wrong, not the code
zai-org/glm-simple-evals/evaluate.py
pointer only (licence: NONE) · get_code("707dde13249b8653")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zai-org/glm-simple-evals 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_final_answer Ran
ran; our contract draft was wrong, not the code
zai-org/glm-simple-evals/evals/common.py
pointer only (licence: NONE) · get_code("bb0b145ae09c560c")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zai-org/glm-simple-evals 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_function_from_code Ran
ran; our contract draft was wrong, not the code
zai-org/glm-simple-evals/evals/scicode_eval.py
pointer only (licence: NONE) · get_code("75198a88e6bc8bc6")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zai-org/glm-simple-evals 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_generic_question_template_answer Ran
ran; our contract draft was wrong, not the code
zai-org/glm-simple-evals/evals/livecodebench_eval.py
pointer only (licence: NONE) · get_code("a56c4308177361e9")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zai-org/glm-simple-evals 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_question_template_answer Ran
ran; our contract draft was wrong, not the code
zai-org/glm-simple-evals/evals/livecodebench_eval.py
pointer only (licence: NONE) · get_code("3708234c4dc55955")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zai-org/glm-simple-evals 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.

last_boxed_only_string Ran
ran; our contract draft was wrong, not the code
zai-org/glm-simple-evals/evals/math_eval.py
pointer only (licence: NONE) · get_code("0b14c648516c38a7")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zai-org/glm-simple-evals 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.

normalize_answer Ran
ran; our contract draft was wrong, not the code
zai-org/glm-simple-evals/evals/grading/math_normalize.py
pointer only (licence: NONE); the stored copy get_code serves, from PKU-SEC-Lab/SPEX, is pointer only (licence: NONE) · get_code("101b1e54fcdadf90")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zai-org/glm-simple-evals 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.

preprocess Ran
ran; our contract draft was wrong, not the code
zai-org/glm-simple-evals/evals/mmlu_pro_eval.py
pointer only (licence: NONE) · get_code("6096219faae6b01a")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zai-org/glm-simple-evals 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.

remove_boxed Ran
ran; our contract draft was wrong, not the code
zai-org/glm-simple-evals/evals/math_eval.py
pointer only (licence: NONE) · get_code("f3bbe264b05aadd3")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zai-org/glm-simple-evals 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.

should_allow_eval Ran
ran, violated the contract we drafted
zai-org/glm-simple-evals/evals/grading/grader.py
pointer only (licence: NONE) · get_code("95f02454eddf082a")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zai-org/glm-simple-evals 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 20 samples lifted from this paper's linked repositories, 20 carry a function identity and 0 do not. Of the 20 distinct functions with an identity, 7 are also present in other papers whose code we hold (7 with at least one paper dated earlier, 5 the same month, 4 later) and 13 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
remove_boxed 20 other papers
19 dated earlier, 1 the same month · 1 only as an assembly brick
2103.03874 (2021-03)
Measuring Mathematical Problem Solving With the MATH Dataset
2309.06275 (2023-09)
Re-Reading Improves Reasoning in Large Language Models
2309.12284 (2023-09)
MetaMath: Bootstrap Your Own Mathematical Questions for Large Language Models
and 17 more not named here
last_boxed_only_string 17 other papers
16 dated earlier, 1 the same month
2309.06275 (2023-09)
Re-Reading Improves Reasoning in Large Language Models
2310.05914 (2023-10)
NEFTune: Noisy Embeddings Improve Instruction Finetuning
2024.findings-emnlp.407 (2024 (year only))
MultiAgent Collaboration Attack: Investigating Adversarial Attacks in Large Language Model Collaborations via Debate
and 14 more not named here
should_allow_eval 13 other papers
9 dated earlier, 1 the same month, 3 later
2305.20050 (2023-05)
Let's Verify Step by Step
2406.12809 (2024-06)
Can Large Language Models Always Solve Easy Problems if They Can Solve Harder Ones?
2502.06703 (2025-02)
Can 1B LLM Surpass 405B LLM? Rethinking Compute-Optimal Test-Time Scaling
and 10 more not named here
count_unknown_letters_in_expr 12 other papers
8 dated earlier, 1 the same month, 3 later
2305.20050 (2023-05)
Let's Verify Step by Step
2502.06703 (2025-02)
Can 1B LLM Surpass 405B LLM? Rethinking Compute-Optimal Test-Time Scaling
2503.10460 (2025-03)
Light-R1: Curriculum SFT, DPO and RL for Long COT from Scratch and Beyond
and 9 more not named here
normalize_answer 10 other papers
8 dated earlier, 1 the same month, 1 later
2404.02078 (2024-04)
Advancing LLM Reasoning Generalists with Preference Trees
2406.12809 (2024-06)
Can Large Language Models Always Solve Easy Problems if They Can Solve Harder Ones?
2412.01981 (2024-12)
Free Process Rewards without Process Labels
and 7 more not named here
are_equal_under_sympy 4 other papers
2 dated earlier, 2 later
2305.20050 (2023-05)
Let's Verify Step by Step
2602.10048 (2026-02)
Long Chain-of-Thought Compression via Fine-Grained Group Policy Optimization
2605.10195 (2026-05)
Breaking the Reward Barrier: Accelerating Tree-of-Thought Reasoning via Speculative Exploration
and 1 more not named here
estimate_pass_at_k 2 other papers
2 dated earlier
2309.00071 (2023-09)
YaRN: Efficient Context Window Extension of Large Language Models
2602.11639 (2026-02)
PACE: Prefix-Protected and Difficulty-Aware Compression for Efficient Reasoning

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

Mixture-of-Experts (MoE) large language models (LLMs) are among the top-performing architectures. The largest models, often with hundreds of billions of parameters, pose significant memory challenges for deployment. Traditional approaches to reduce memory requirements include weight pruning and quantization. Motivated by the Router-weighted Expert Activation Pruning (REAP) that prunes experts, we propose a novel method, Routerweighted Expert Activation Merging (REAM). Instead of removing experts, REAM groups them and merges their weights, better preserving original performance. We evaluate REAM against REAP and other baselines across multiple MoE LLMs on diverse multiple-choice (MC) question answering and generative (GEN) benchmarks. Our results reveal a trade-off between MC and GEN performance that depends on the mix of calibration data. By controlling the mix of general, math and coding data, we examine the Pareto frontier of this trade-off and show that REAM often outperforms the baselines and in many cases is comparable to the original uncompressed models.

For agents

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

get_harvested_code_for_paper("2604.04356")
get_code_for_paper("2604.04356")
have("2604.04356")

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

curl https://syntology.ai/api/ran/2604.04356.json

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

[![Syntology run record](https://syntology.ai/api/ran/2604.04356.svg)](https://syntology.ai/paper/2604.04356)

Connect an agent — have() is free.